1 RESTORATION OF RED SANDERS (Pterocarpus santalinus L.) FORESTS FOR ECOLOGICAL AND ECONOMIC BENEFITS By SIDHANAND KUKRETY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Sidhanand Kukrety
3 To my parents who se encouragement gives me strength to follow my dreams, to my mother in law who is an inspir ation and to the memor y of my father in law, who would have loved this endeavor
4 ACKNOWLEDGMENTS This dissertation is the labor of love gained following a dream. While working with the Andhra Pradesh Forest Department (APFD) in Chittoor district, India, I wi tnessed the problems associated with Red Sanders forests What follo wed changed course of my job with the department and I moved to Gainesville, Florida to pursue my doctoral research The n I realized that my advisor Dr. Janaki R.R. Alavalapati also had a similar desire. But for Dr. Alavalapati this research would not have taken place at all and to him I owe my deepest gratitude and appreciation. He chaired my committee even when he was busy as the H ead of the D epartment Forest Resources and Environmental C onservation at Virginia Polytechnic Institute and State University. Dr. Shibu Jose, my committee co chair later told me that they (both) had been talking about this subject for the last ten years. I hope their faith in me was justified. I benefitted from Dr. Jose immense experience in designing the field study in India and many other subjects including personal matters. He also spared his valuable time in guiding me even whe n his work as the Director Center for Agroforestry Missouri left him little t ime to spare. My other committee members, Dr. Alan Long, Dr. Alan Hodges and Dr. Carrie Reinhardt Adams were of tremendous help and support during my research. I thank Dr. Salvador Gezan, who helped me in refining the analysis of ecological data. I thank t he School of Forest Resources and Conservation (SFRC) United States Department of Agriculture and United States Department of Energy for f unding this study through Bioenergy Incentives and Forest Sustainability (BIFS) project F unding support provided by The Rufford Small Grants Foundation for field work is gratefully acknowledged I also acknowledge the encouragement and support received from my senior Indian Forest Service colleagues at Hyderabad, India. Field work in India would not
5 have been possible withou t t he help and logistic support from the A PFD I specially thank Dr. P. Subba Raghavaiah, Mr. T. V. Subba Reddy, Mr. C. P. Vinod Kumar Mr. G. Ramalingam for their help in field work and Mr. P. Srinivas for help with the GIS data For the logistic support and for data collection invaluable help provided by Mr. M. Khader Valli and his dedicated team beyond the call of duty is greatly appreciated. I thank all the faculty members Red San ders wood traders, farmers and A PFD personnel for p articipat ing in the survey. I thank Professor G. Sudarshanam for sharing his work and photographs on Red Sanders. The period of stay at Gainesville would not have been smooth without the support of faculty, staff and friends at the School of Forest Resources and Conservation. I specially thank Fredrick Rossi, Andres Susaeta, Ajay Sharma Ilan Kaufer, Jean Gael Emptaz Collomb, Ming Yuan Huang, Nilesh Timilsina Puneet Dwivedi, Pankaj Lal and Tyler Nesbit for their friendship. My parents and my m other in law who are great source of inspiration strongly encouraged me and stood by me to pursue my doctorate Their unflinching support unconditional love and abundant blessings remind me of the sacrifices parents make for their kids. To them I owe my deepest gratitude and love. My wife Swati has been a constant companion, a source of intellectual and emotional support all through the trials and tribulations and n o words of appreciation would suffice for her contribution in my life and during this study Finally, I convey my love and gratitude to my son Suchindram who kept pace with me and with his studies and has embarked upon his own quest for knowledge
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 Tropical Forest Degradation in India ................................ ................................ ....... 15 Ecological Restoration ................................ ................................ ............................ 16 The Red Sanders Forests ................................ ................................ ....................... 17 Importance to Society ................................ ................................ ....................... 18 Unique RS Wood Qualities ................................ ................................ ............... 19 RS W ood Trade ................................ ................................ ................................ 21 Challenges of Restoration ................................ ................................ ....................... 22 Ecological Aspects of RS Forests ................................ ................................ .... 22 Socioeconomic Status of Surrounding Communities: ................................ ....... 24 Adoption of RS by Private Landowners ................................ ............................ 24 Problem Statement ................................ ................................ ................................ 26 Study Objectives ................................ ................................ ................................ ..... 28 Organization of the Dissertation ................................ ................................ .............. 28 2 RESTORATION OF THE ENDANGERED Pterocarpus santalinus L. ENHANCING SURVIVAL AND GROWTH OF ADVANCED REGENERATION USING SILVICULTURAL TOOLS ................................ ................................ ........... 32 Restoration of Endangered Species ................................ ................................ ....... 3 2 Red Sanders The Species ................................ ................................ ............. 33 Ch allenges of R estoration ................................ ................................ ................ 34 Restoration Strategies ................................ ................................ ...................... 35 Material and Methods ................................ ................................ ............................. 36 Study Site ................................ ................................ ................................ ......... 36 Study Design ................................ ................................ ................................ .... 38 Statistical Analyses ................................ ................................ .......................... 40 Results ................................ ................................ ................................ .................... 42 Survival ................................ ................................ ................................ ............. 42 Growth ................................ ................................ ................................ .............. 43 Height growth ................................ ................................ ............................. 43 Root Collar Diameter growth ................................ ................................ ...... 43 Volume growth ................................ ................................ ........................... 44
7 Discussion ................................ ................................ ................................ .............. 45 Effect of Treatmen ts on Survivals ................................ ................................ ..... 45 Effect of Treatments on Seedling Growth ................................ ......................... 47 Summary and Prescriptions ................................ ................................ .................... 50 3 ENERGETIC, ENVIRONMENTAL AND ECONOMIC EVALUATION OF ETHANOL PRODUCTION FROM FOREST GRASS BIOMASS IN INDIA A LIFE CYCLE ASSESSMENT APPROACH ................................ ............................. 59 India Lignocellulosic Ethanol Scenario ................................ ................................ ... 59 Methods ................................ ................................ ................................ .................. 62 Goal and Scope ................................ ................................ ................................ 63 System Boundaries ................................ ................................ .......................... 64 Functional Unit ................................ ................................ ................................ 64 Assumptions ................................ ................................ ................................ ..... 65 Inventory ................................ ................................ ................................ ................. 65 Biomass Characteri stics and Harvest ................................ ............................... 66 Transportation ................................ ................................ ................................ .. 67 Ethanol Production ................................ ................................ ........................... 67 Energy Production ................................ ................................ ............................ 69 Allocation of Energy and Material Use ................................ ............................. 70 Net Energy ................................ ................................ ................................ ....... 71 Emissions ................................ ................................ ................................ ......... 72 Lifecycle Impact Assessments ................................ ................................ ................ 73 Interpretation ................................ ................................ ................................ ........... 74 Economics of E thanol Production ................................ ................................ ........... 74 Cost of Ethanol ................................ ................................ ................................ 74 Socioeconomic Benefits ................................ ................................ ................... 75 Sensitivity, Uncertainty and Risk Analysis ................................ .............................. 75 Results and Discussions ................................ ................................ ......................... 76 Energy Balance ................................ ................................ ................................ 77 Emissions ................................ ................................ ................................ ......... 78 Impacts and Interpretations Ethanol P roduction ................................ ............ 78 LCIA of Gasoline Use & Grass B urning vs Ethanol Production & E 10 Use ..... 78 Net emissions scenario A vs scenario B ................................ ................. 79 Net impacts scenario A vs scenario B ................................ ..................... 79 Economics of Ethanol Production ................................ ................................ ........... 80 Cost of Ethanol ................................ ................................ ................................ 80 Socioeconomic Impact ................................ ................................ ..................... 80 Sensitivity, Uncertainty and Risk Analysis ................................ .............................. 81 Summary, Conclusions and Policy Implications ................................ ...................... 83 4 SANDERS WO OD MARKETING IN ANDHRA PRADESH, INDIA ....................... 101 Red Sanders F orests ................................ ................................ ............................ 101 SWOT AHP Framework ................................ ................................ ...................... 105
8 Method s ................................ ................................ ................................ ................ 107 Results and Discussion ................................ ................................ ......................... 110 Administrators Stakeholder Group ................................ ................................ 110 Landowners Stakeholder Group ................................ ................................ ..... 112 Traders Stakeholder Group ................................ ................................ ............ 114 NGOs & Academia Stakeholder Group ................................ .......................... 115 Overall Perceptions of All Stakeholder Groups ................................ ..................... 117 Summary and conclusions ................................ ................................ .................... 118 5 CONCLUSIONS, POLICY IMPLICATIONS AND FUTURE RESEARCH ............. 129 Conclusions ................................ ................................ ................................ .......... 129 Ecological Benefits ................................ ................................ ............................... 133 Policy Implications ................................ ................................ ................................ 134 Limitations of the Study and Scope of Future Research ................................ ....... 135 APPEND IX A RED SANDERS SELECTED PHOTOGRAPHS ................................ ................ 137 B SUPPLEMENT TO CHAPTER 3 DESCRIPTION OF IMPACT CATEGORIES .. 143 Global Warming Potential ................................ ................................ ..................... 143 Acidification Potential ................................ ................................ ............................ 143 Smog Potential ................................ ................................ ................................ ..... 144 Human Health ................................ ................................ ................................ ....... 144 Water Use ................................ ................................ ................................ ............. 145 LIST OF REFERENCES ................................ ................................ ............................. 146 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 159
9 LIST OF TABLES Table page 2 1 Description of variables ................................ ................................ ...................... 52 2 2 Description of categorical variables for study of treatment effects ...................... 53 2 3 Results for survival and growth study analyses ................................ .................. 53 2 4 Results for different treatments ................................ ................................ ........... 54 2 5 Survival and growth by seedling characteristics and favored treatments ........... 54 3 1 Sources of data used in the study ................................ ................................ ...... 86 3 2 Inputs, costs and data sources ................................ ................................ ........... 87 3 3 Risk and sensitivity analysis values ................................ ................................ .... 88 3 4 Energy balance of ethanol production system ................................ .................... 89 3 5 Net emissions in ethanol production ................................ ................................ ... 90 3 6 Emissions Scenario A vs Scenario B ................................ .............................. 91 4 1 List of identified factors under each SWOT category ................................ ........ 121 4 2 Pair wise comparison matrix in questionnaire for the Weakness category ....... 121 4 3 Overall relative priorities of factors and SWOT Factors ................................ .... 122
10 LIST OF FIGURES Figure page 1 1 Red Sanders forests ................................ ................................ ........................... 30 1 2 Conceptual research framework ................................ ................................ ......... 31 2 1 Study design ................................ ................................ ................................ ....... 55 2 2 Treatmen t vs. Survival proportion ................................ ................................ ....... 55 2 3 Treatment vs. Height growth ................................ ................................ .............. 56 2 4 Treatments vs Root collar diameter (RCD) growth ................................ ............ 57 2 5 Treatment vs. Volume growth ................................ ................................ ............. 58 3 1 System boundary and reference system ................................ ............................ 92 3 2 Flow chart two stage dilute sulphuric acid process ................................ ............. 93 3 3 Process energy involved per liter in Ethanol production system ......................... 94 3 4 Et hanol production system impacts ................................ ................................ .... 94 3 5 Net impact changes Scenario A vs. Scenario B ................................ .............. 95 3 6 Cost components of ethanol production ................................ ............................. 96 3 7 Regression coefficients of total present value of costs. ................................ ...... 96 3 8 Prob ability distribution of discount rate ................................ ............................... 97 3 9 Probability distribution of biomass cost ................................ ............................... 98 3 10 Probability distribution of Ethanol cost ................................ ................................ 99 3 11 Sensitivity analysis ethanol production cost vs. Biomass cost ........................ 100 4 1 Red Sanders forests in Andhra Pradesh in India. ................................ ............. 123 4 2 Perceptions of the Administrators stakeholder group ................................ ....... 124 4 3 Perceptions of the Landowners stakeholder group ................................ ........... 125 4 4 Perceptions of the Traders stakeholder group ................................ .................. 126 4 5 Perceptions of the NGOs & Academia stakeholder group ................................ 127
11 4 6 Overall trends and SWOT category ................ 128 A 1 Red Sanders: a) Trees in urban area ; b) Plantation on public forests .............. 137 A 2 Twig of Red Sanders showing leaves and seeds ................................ ............. 137 A 3 Red Sanders plantation on public forests ................................ ......................... 138 A 4 Natural Red Sanders forest ................................ ................................ .............. 138 A 5 Red Sanders seedlings ................................ ................................ .................... 139 A 6 Mature Red Sanders tree bark ................................ ................................ ......... 139 A 7 Red Sanders heartwood (a) without grain (b) with wavy grain ......................... 140 A 8 Uses of Red Sanders wood ................................ ................................ .............. 140 A 9 Illegal RS wood seized by the local forest department ................................ ..... 141 A 10 Illegal RS wood seized at port in a shipping container ................................ ..... 141 A 11 Musical instrument parts seized by authorities while in transport ..................... 142
12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy RESTORATION OF RED SANDERS (Pterocarpus santalinus L.) FORESTS FOR ECOLOGICAL AND ECONOMIC BENEFITS By Sidhanand Kukre ty May 2011 Chair: Janaki R. R. Alavalapati Cochair: Shibu Jose Major: Forest Resources and Conservation Pterocarpus santalinus L (Fabaceae) commonly known as Red S anders (RS) is an endangered tropical timber tree species endemic to southern India. O verexploitation, anthropogenic disturbances and dr o ught are known to have degraded th is landscape to such an extent that restorative measures are needed to halt further decline This study aims at 1) determining approaches to improve regeneration and growth of RS; 2) conducting environmental and economic feasibility of utilizing understory grass, Cymbopogon coloratus for ethanol production; and 3) assessing stakeholders perceptions about sustainable R S trade. Field trials involving treatments Prescribed fire (PB), in combination with disking (DPB), singling (SPB), and singling+ disking (SDPB) and control (NT)) designed to ameliorate microsite conditions revealed improv ed seedling survival and growth in young regeneration T all and larger stump size seedlings experienced better height and volume growth with SPB and SDPB treatments respectively. For root collar diameter growth, t all seedlings and seedlings with smaller stump size showed better results with SDPB treatment The number of coppice shoots did not have any impact on growth but s eedlings with fewer coppice shoots showed
13 better survival with DPB treatment Even though the relation ship between treatment, survival and growth was identified, no s ingle treatment showed uniform ly better growth with all type of seedlings. Life cycle a ssessment and benefit cost analysis were used to ascertain environmental and socio economic feasibility of using Cymbopogon coloratus grass for ethanol. T he net energy value of 16.86 MJ L 1 ethanol and a net energy ratio of 5.15 indicated the potential of this biomass for ethanol The production and use of one liter ethanol as E10 motor fuel is expected to reduce net emissions by 0.74 kg ( CO 2 equivalent ) compared to the status quo situation of gasoline use and open burning of grasses. The reduc tion of emissions will reduce human health (cancer), global wa rming and smog potential by 29%, 9 % and 8 % respectively. W ith a delivered feedstock cost of $ 25.34 per ton ne dry grass biomass the estimated cost of ethanol at $0.42 L 1 is found to be competitive with the ethanol produced from other feedstocks Improved production efficiency of ethanol showed 16% mean increase in emissions. Finally, Analytic Hierarchy Process with Strength, Weaknesses, Opportunities and Threats ( AHP SWOT) techniques were used to assess perceptions of key stakeholder groups towards development of sustainable RS wood trade The negative perceptions (weaknesses and threats) for the administrator landowner trader and for the NGO s & academia stakeholder groups were found to be 291%, 63% 59% and 12% more than the positive perceptions (strengths and opportunities). These r esults indicate that in the existing set up, stakeholders participatio n would be limited in RS wood trade and if such a trade is to succeed, both external and internal issues will need to be addressed.
14 This study showed that improved restoration techniques, production of et hanol from understory grass, addressing the concern s and harnessing opportunities relating to RS trade can create a win win situation for the economy, environment and stability of rural communities A RS restoration c onsortium, consisting of representatives from state and national Government agencies, international Government s and aid agencies, non governmental organizations local communities, and the private sector may help in co llaboration and coordination of these activities.
15 CHAPTER 1 INTRODUCTION Tropical Forest Degradation in India Over the period of human dominance, many fragile and sensitive ecosystems have lost an innumerable number of species all over the world including India With a land base of three million square kilometer s and 1.3 b illion human population 75% of people in India directly or indirectly de pend on agriculture It is a large developing country known for its diverse forest ecosystems and mega biodiversit y and ranks 10 th amongst the most forested nations of the world (FAO, 200 5 ) It has 21 percent (69 million hectare s ) of its geographical area under forest and tree cover (FSI, 2009 ). With ports about 10% of the wo d iversity making it the seventh richest biodiversity country in the world. There is heavy poverty driven dependence on natural resources for Non Timber Forest Products (NTFP) including fuelw ood and fodder. It is estimated that 68% of the rural population of India is dependent on fuelwood for their energy requirements (Lal, 1989). Unrestricted grazing and w ildfires are yet other cause s of forest degradation in India. G razing and browsing cause damage to forests and compact soil which makes regeneration difficult. W ildfires destroy most of young regeneration besides causing immense environmental and ecological damage. Most wildfires are set by graz i ers for allowing a fresh flush of grasses to reappear but some are accidental as well. Due to t hese diverse problems acute degradation and fragmentation are threatening the forest diversity and conservation efforts. The State of Andhra Pradesh (AP), the 5th largest state of India has 23.2 % (6.38 million hectare) under forests. It is considered as one of the rich biodiversity states in
16 India due to its strategic location, geographical variation and supports a large number of endemic and endangered flora and fauna. With 26613 villages 55.4 million rural population (2001 census), and 60.08 million livestock (2007 (P) census) forests in AP are experienc ing immense pressure (APFD, 2010 ) T o overcome the biotic pressure led degradation AP launched joint forest management (JFM) during 1992 and encourage d community participat ion in management of forest s to restore the degraded forests especially in fringe areas Since 2002, Community forest m anagement (CFM) was introduced as an advancement over the joint forest management ( JFM ) with devolution of more powers to the forest protection committees in manage ment 1 The community is encouraged to protect and restore forests under their control while receiving benefits and earning livelihood. Ecological Restoration Ecological restoration i s an intentional activity that initiates or accelerates the recovery o f an ecosystem with respect to its health, integrity and sustainability ( Society of Ecological Re storation SER 2004) Restoration of damaged or degraded ecosystems is a challenging task. In most developing countries, where the communities share an inextricable and intricate relationship with, and dependence on forests for their livelihoods, restoration of degraded forests becomes even more challenging (Sayer et al 2004). Restoration of endangered species is more delicate in nature as it often inv olves dealing with the remnant population. Wherever possible, restoration by assisted natural regeneration has been one of the main choice s of practitioners. This is especially helpful for landscape level restoration due to low costs and ease of 1 Village level forest protection committees in Andhra Pradesh are known as Vana Samrakshana Samithies (VSS). For more details please see http://forest.ap.nic.in/JFM%20CFM/JFMINDEX.htm
17 management The guiding principle of assisted natural regeneration involves removal of regeneration. In developing countries the degradation is often due to socio economic reason s, and thus there is an identified need to ensure cooperation of rural communities living near the forest fringe areas and having a vested interest in success of the re storation (Lamb and Tomlinson, 1994). The Red Sanders (RS Pterocarpus santalinus L. ; Fabaceae ) forests of southern AP are one such forest in need of restoration. The Red Sanders Forests RS is a medium sized, dry deciduous tree endemic to southern India (Anuradha and Pullaiah, 1999). It i s found distributed in a geo botanically restricted area of 2000 km 2 between 13 0 30 15 0 0 N latitude and 78 0 45 79 0 E longitu de (Raju et al, 1999). Within this natural distribution, RS grows in approximately 500 km 2 of fragmented forest landscape of A P Karnataka and Tamil Nad u ( Jain and Rao, 1983; Molur and Walker, 1995; Oldfield et al. 1998). The total growing stock of RS in AP forests (Figure 1 1) have been estimated at 118 000 m 3 (A PFD, 2009). The tree prefers 80 c m to 120 c m rainfall, perfect drainage and elevations ranging from 200 m to 900 m (Ahmed and Nayar, 1984). RS forests are also home to o the r rare flora like Shorea tumbaggiana, Shorea talura, Syzygium alternifolium, Terminalia pallida, Cycas beddomie, Decalepis hamiltonii and fauna like Loris tardigradus Cal odactylodes aureus Elephas maximus Panthera pardus Melursus ursinus Cuon alpines and Viverricula indica RS is renowned for its characteristic timber of exquisite color, wavy grain, beauty and superlative technical qualities and ranks among the fines t luxur ies in Japan (Reddy, 1990) In its natural distribution it occurs almost exclusively in geological formations of Quartzites (80%) and Shales (20%) (Na garaju and Raju, 1998). It is a known
18 accumulator of Strontium and Cadmium (Raju and Rao, 1988). The photographs of R S trees, forest and wood types, seedlings and its uses are shown in Annexure A. Importance to Society RS wood was traditionally used in India for medicinal purposes, agriculture equipment, building timber and for carving religious souvenir figurines and statues (Ramakrishna, 1962). Outside India, the earliest historical record of RS wood appeared in China do cumented by Cui, Bao of Jin Dynasty (265 340 A.D.) in "Comments on Ancient and Modern Items". In this text, this wood is regarded as a rare and precious lignum as expensive as gold. The earliest modern recorded history of RS wood trade can be traced back to 1681 when the British East India Company transported it to England for dyeing purpose s (Reddy, 1972). During the latter half of the eighteenth century and most of the nineteenth century, RS wood was used in railway sleepers (Reddy, 1972). From the earl y twentieth century (1915 1920), RS wood trade for making musical instruments in Japan started for which only highest quality wavy grained RS wood is used (Kiyono, 2005). In China, RS wood is mostly used in making musical instruments and for producing Ming and Qing dynasty imitation furniture. In Ayurveda (the traditional Indian system of medicine) RS finds mention fo r its wound healing properties for external ap plication in treating inflammation, as an astringent and tonic for treating headache s skin di seases, fever, boils, scorpion sting s and for improv ing sight ( Biswas, 2003 ; Chopra et al 1956; Kirtikar and Basu, 1983 used in coloring pharmaceutical preparations and foodstuffs (Manjunatha, 2006). The heart wood contains Isoflav one Glucosides (Krishnaveni and Rao, 2000) and anti tumor lignans, e.g., Savinin and Calocedrin (Rao, 2001). Ethanol extract of stem bark was reported to possess anti hyperglyc emic activity (Rao et al, 2001). Narayan et al, (2005)
19 established antiulcer pro perties of ethanol extract of RS against ulcers produced by Ibuprofen and Misoprostol. Manjunatha (2006) reported the antibacterial activity of leaves and stem bark of RS against both Gram positive and Gram negative organisms. Even though RS is one of the most important species in AP a literature review did not reveal any peer reviewed study on its autecology and restoration. Unique RS Wood Qualities Other than variation in color of RS heartwood due to the presence of dye, the presenc e or absence of wavy g rains imparts unique characteristics to the RS heartwood (henceforth wood) The reasons for development of wavy grain in the RS wood are not known and such trees cannot be identified based on morphological differences. There could be some genetic basis for this character (Kedarnath and Rawat 197 6). Ramakrishna (1962) described the occurrence of wavy grain RS wood as rare, However, it is this wavy grain RS wood which attracts a maximum market premium For instance, the R S wood price ranged from US$ 6,870 9,160 per metric tonne in 2002 in global timber markets (M ulliken and Crofton, 2008). The finished RS wood products such as carved statues and furniture fetch even greater price. Due to the morphological variations in ind ividual trees low germination and slow growth, research needs to focus on improv ing production of good quality, genetically superior RS crop s. Increased thre at to RS forests, prompted the Government of India (GoI) and G overnment of Andhra Pradesh (GoAP) to restrict commercial harvest of RS in 1956 (Ramakrishna, 1962) and to impose a total ban on harvest from public forests in 1980 (Working plan Rajampet, 2008). Based on a proposal from India in 1995, RS was included in t he Convention on International Trade in Endangered Species ( CITES )
20 endangered plant list 2 wherein it is classified in Appendix II of CITES, as endangered alteration (IUCN, 2010) Under 2009 CITES rules, other than RS logs, woo d chips and powder, extracts also need GoI and GoAP clearance s for trade. RS was similarly a rees (Oldfield et al, 1998) and in the 1994 IUCN Red List du e to its endemic status, small range, fragmented populations and continuing decline (IUCN 2010 ). Due to inclusion of RS into CITES, G overnment and CITES clearances became mandatory for expor ting RS wood from the country. Even though, RS wood has an assur ed market and the demand is always high, there is low adoption of the species by landowners and private entrepreneur s for cultivation. Except for a few scattered trees planted by the farmers on field bunds, c ultivation at a commercial scale is not k nown to exist on private lands. Factors that may explain the for low adoption of RS as a plantation species include small landholdings, poor market information, lack of extension support, administrative delays in trade and competition from other commercial crops. The local forest department (Andhra Pradesh Forest Department APFD) established plantations in 4099 ha during 1960 1975 in Chittoor, Kadapa and Kurnool districts of AP which have not been c ommercially exploited (APFD 2010 ). Although plantation pro duced wood is often considered to be inferior in quality as compared to the wild variety 3 the local forest department has taken up promotion and the establishment of nurseries and plantations. 2 Annotat ion # 7, Chips, Powder and Timber of R ed Sanders is not permitted in trade. 3 Plantation wood has pale color and low density as compared to the wild variety and commands less premium.
21 RS Wood Trade Commercial exploitation of RS was restricted in 1956 (Ramakrishna, 1962) an d stopped in 1980 (Working plan Rajampet 2008 ). T he present Government policy is restricted to con servation and protection of RS trees on public lands but not on promoting RS plantation on private lands The endangered nature of the species and restricted trade on one hand and the ban on harvest owing to the endemic nature of the species and slow growth on the other have resulted in a demand supply gap. Lack of private participation in RS cultivation on private lands has further aggravated the situation. The short supply of RS wood in international market s is also the main cause of illegal trade. It is also important to note that international demand of RS wood far exceeds domestic demand and thus the challenge of RS survival has its root in its international trade. The challenge before the conservationists, resource managers, Government and international community is the management and regulation of RS wood trade. The demand supply gap is prompting illegal harvest and encouraging smuggling of RS wood. According to Ramakrishna (1962) illegal removal of mature RS trees for its high value has always been a cause of concern and as eviden t from the records of seizure, the situation seems to have worsened with passage of time. CITES (20 07) reports indicate seizure of 947 t of RS wood between 2003 and 2005. Furthermore, supply of illegal wood from non source countries like Nepal, Myanmar, Singapore, and Taiwan suggests that RS wood goes through a variety of markets befor e it reache s its final destinations Besides this, the local forest department alone seized 3067 t of RS w ood between 2001 and 2007 (APFD, 2010) The only legal RS wood that enters the trade is the wood seized by the APFD and later sold in open auction The annual
22 intern ational demand is estimated to be 3000 t (Mulliken and Crofton, 2008) which is far less than the potential supply. Challenges of Restoration The goal of restoration is to reestablish and retain the resilience of degraded forests to achieve sustainable mana gement and to provide a broad range of ecosystem services. It also requires intimate knowledge about the species or assemblage of species. Even though RS is one of the most important species of the southern Indian plateau, hardly any research has gone into its ecology, growth and factors responsible for its restoration. Knowledge of this indigenous species in a reference ecosystem and potential threats to the health and integrity of the restored ecosystem need to be studied and documented. Several factors t hat pose challenge s to RS conservation include slow tree growth, excessive biotic pressures, poor socioeconomic condition of local communities, and illegal removals. Development of a suitable environment for scientific enquiry and long term commitment of r esources together with a comprehensive policy towards achieving restoration of RS forests should be a step in th e right direction. Some of the important challenges are discussed below Ecological Aspects of RS Forests The landscape level RS restoration involves habitats, populations, species, genes along with all ecological interactions, processes and some of the traditional human practices that are associated with the ecosystem. For example, domestic cattle grazing and forest fires set by graziers cause severe damage to the RS seedlings (Ramakrishna, 1962). Additionally, the reduced number of mature seeding trees (mostly due to illegal logging), poor seed germination (30% 40%), and low survival rate of seedlings (due to recurring wildfires and grazing) are adversely impacting regeneration
23 of RS in forests. For example, as per the total tree enumeration data collected by the local forest department in 2006, 85% of RS trees in forests were below 75 cm and less than 1% we re above 100 cm girth at breast height (APFD, 2006). The present RS forest is mostly comprised of coppice growth. The severity of threat to this species highlights the necessity of immediate steps needed to ensure i n situ conservation of this species. The commercial exploitation of RS trees in the past without commensurate replenishment has often been blamed for deterioration of the RS ecosystem (Ahmad and Nayar, 1984). If the trend of degradation is allowed to continue, fragmentation and subsequent ext inction of RS population may become imminent. Serious steps to restore its shrinking extent, fragmentation and degrading habitat with simultaneous policy reforms are needed to save the species from extinction. The existing degraded RS forests have thus rea ched a stage where restoration has become critical in the interest of ecology, environment, and society. Other th a n smuggling, recurring wildfires in RS forests are a lso a cause of concern. In RS forests, Boda grass (BG) ( Cymbopogon coloratus ) is abundant as an understory associate. In the past local communities utilized this resource as thatch material for huts which has considerably reduced due to advent of modern day housing. As a result the grasses remain on the forest floor as fuel load. Most wildfi res are of anthropogenic origin; accidental or deliberate, as the graz i ers set fire to allow a fresh flush of grass to appear. BG serves as fodder when young but not when mature. Recurring wildfires damage most of the RS regeneration, besides causing other ecolog ical and environmental problems It is therefore important to find ways and means to control wildfires. Linking the use of BG by identifying an income generating activity
24 might help in its removal and in the process control wildfire or reduce their intensity to some extent. U nrestricted grazing prevalent in RS forests add s to the problem of degradation by adversely impacting the seedling regeneration and establishment. R S is an excellent fodder and survival of young RS regeneration su ffers on account of this activity. Socioeconomic Status of Surrounding Communities: The s ocioeconomic condition of the communities surrounding RS ecosystem plays a key role in RS production, management, and marketing. Associated with this is the fact that causes of degradation are often intermingled with the socio economic factors, and thus there is an identified need to ensure the cooperation of rural communities actually living in the area and having a vested interest in success of the restoration progra m (Lamb and Tomlinson, 1994). Unscrupulous smugglers take advantage of the poverty prevailing in the forest fringe villages and lure people to illegally harvest RS trees from public forests Other factors like domestic cattle grazing, recurrent wildfires e tc., are also intricately related to the socioeconomic status of local population s For example, families with poor socioeconomic background raise local variety cattle which can be grazed in forests. At the end of the grazing season, they deliberately set fire to forest to allow a fresh flush of grasses for grazing. The control of such disturbances is necessary for the ongoing restoration and long term viability of the restored landscape as their continued presence undermines the restoration activity. Adop tion of RS by Private L andowners Adoption of RS plantations by local farmers for commerce is another issue that deserves deeper understanding. Unlike many other species the RS wood is mostly exported to other countries. Many factors like small land holding s, administrative
25 complexities, and institutional regulations on harvest and marketing may be acting as barriers in adoption of RS for commercial purpose Further, the local communities get very little user benefits from public forests. D irect economic benefits and/ or indirect incentives including any environmental and social services are important to the local communities, and unless available, their involvement in protection is not likely to be sustained (Sayer et al., 2004) RS wood has been in trade since ancient times and excess demand still drives the markets. Currently, most of the RS wood in market is channelized through illegal involvement in management on public fo rests are either limited or zero. A way to meet the existing demand might include raising RS plantations on private and communal lands by encouraging landowners. However, administrative procedures for subsequent harvest and marketi ng under state and CITES rules are cumbersome, lengthy and time consuming These factors are known to cause frustration among private landowners and communities and they move away from raising RS trees. A review of rules framed by the APFD for removal of RS wood from private lands indicates that the harvest of RS trees is possible only after a number of revenue and forest officials have inspected the trees and wood on several occasions. The entire process from start to finish is so complex that instead of seeking p ermission, the fa rmers sell their standing trees at throwaway prices to the middlemen, who in turn manag e permissions and the markets. Since the farmers gain little from the sale, they are reluctant to take up planting of RS trees. As stated in earlier sections, o ther reas ons that add to this problem are theft, competition from other commercial crops, poor market information, lack of extension
26 and insurance mechanism s and lack of incentives for raising RS plantations. Remedial measures may require formulating a long term p olicy for developing RS markets by involving multiple stakeholders on one hand and reforming th e existing regulatory policies on the other For success of RS ecosystem restoration identifying the bottlenecks and creating incentives for stakeholders is impo rtant. It is assumed that successful adoption of RS by landowners will help in reducing the demand supply gap thereby curtailing the illegal removal from public forests. Problem Statement Review of the available literature reveals that despite the uniquen ess of RS forests, there is dearth of research on aut ecology restoration and policies pertaining to this species. The only literature available on this species refers to the presence of alkaloids, dyes and their medicinal properties. The above discussion indicates the problems associated with this unique landscape and it is evident that suitable steps to restore its habitat must be taken to ensure its survival. Other than protection of the landscape from wildfires and grazing, identification of treatments to improve survival of young regeneration will help in improving the density of RS in the wild Since RS is a protected species, local communities do not get any appreciable benefits fro m these forests. The lack of direct benefits on one hand and the increased regulatory trade policies on the other have further distanced the communities from this species. For In order to ensure involvement of communities and other stakeholders in active protection, it is necessary to incorporate their perception s in the formulation of restoration and trade related policies. Further, in order to improve community ownership, the use valu e of RS forests for communities must be increased Improving livelihood opportunities for the
27 rural commu nities would help in achieving this goal For example B odagrass (BG) which is abundant in RS forests can be used for energy, paper making and for ext r action of aromatic oils The removal of BG can help in creat ing a source of livelihood for rural communities on one hand and reduce instances of wildfire on the other An integrated, all encompassing restoration plan backed by research is required to save the RS landscape. T here exists a n identified knowledge gap and this study is an attempt to fill this gap and to develop an integrated strategy for restoration of degraded RS forests. The conceptual research framework for the study is shown in Figure 1 2. The framework essentially includes 1) identification and prioritization of restoration areas; 2) identification and evaluation of measures for improving species density; 3) involvement of stakeholders in restoration planning and policy formulation and 4) improving the use value of the existing forests by identifying inc ome generating opportunities. In this study the areas closer to the villages, controlled by th e forest protection committee s a re considered for restoration and hence identification of speci fic areas and agenc ies for restoration was excluded. The conceptual framework indicates the different component s of restoration stra tegy which can be constantly updated based on feedback received and observations in the field Without integration of differ ent components at landscape level it is difficult to achieve restoration goals. For instance, mere increase in species density may not result in long term success if the community does not support the protection of area from fire or grazing. On the other h and, the communities are unwilling to participate if there is no net gain for them in the process. The specific components of the conceptual framework researched in this study are presented as study objectives.
28 Study Objective s The overall objective of thi s study is to develop an integrated strategy for restoration of RS forests and seek answers to some of the fundamental questions that would help the policymakers, resource administrators and managers local communities, and private landowners to make decis ions relating RS T he main objectives are : a) To evaluate the impacts of silvicultural treatments on survival and growth of advance RS regeneration in the degraded forests b) To conduct the environmental and economic evaluation of the use of BG for bioenergy by conversion into ethanol. c) To assess stakeholder developing a sustainable RS wood trade. Organization of the Dissertation This dissertation is or ganized in to five chapters. Chapter 2 evaluates t he strategies for successful restoration by using biophysical treatment s with an aim to improve growth and surviv al of young RS regeneration A set of five cost effective treatments were applied to RS regeneration for a period of two years to study their effect on survival and growth. Chapter 3 explores the value addition to RS forests by assessing the energetic and economic feasi bility of using BG for bioenergy Both life cycle analysis and benefit cost analysis are applied to achieve this task. The s takeholders perceptions towards development of sustainable RS wood trade are discussed in C hapter 4 For this purpose SWOT AHP (Strengths, Weaknes ses, Oppo rtunities and Threats Analytic Hierarchy Process ) technique s were used to analyze perceptions of four key stakeholder groups (administrat ors, landowners, traders, and NGO & academia) towards sustainable RS wood trade development in the southern Indian state of Andhra
29 Pradesh. A brief summary, conclusions policy implications, limitations of the study, and directions for future research are provided in C hapter 5
30 Figure 1 1 Red Sanders forests (Modified from Forest Survey of India 2009)
31 Figure 1 2 Conceptual r esearch f ramework RESTORATION OF DEGRADED RED SANDERS FORESTS Historical Extent of RS Distribution Identify Areas for Restoration Predictive Distribution Modeling Prioritize Restoration Areas Identify Restoration Agency Formulate Sustainable RS Wood Trade Policies Perceptions Improve Growth & Survival of RS Seedlings Field Trials Identify Best Suited Biophysical Treatment Value Addition to RS Forests Identify Feasible Project Create Livelihood Opportunities Process Design Material and Energy Balance Benefit Cost Analysis Capital and Capital Cost Estimates Life Cycle Assessment Economic / Environmental Risks & Benefits Improve Community Linkages
32 CHAPTER 2 RESTORATION OF THE ENDANGERED Pterocarpus santalinus L. ENHANCING SURVIVAL AND GROWTH OF ADVANCED REGENERATION USING SILVICULTURAL TOOLS Restoration of Endangered Species Restoration of endangered species and ecosystems in a tropical context is often challenging due to limited biological information on the species. Formulating successful restoration strategies involves understanding species specific seed and seedling biolog y and ecology with respect to environmental stressors and disturbances (Holl et al., 2000; Zelder, 2005). Regeneration of trees primarily depends upon their ability to provide sufficient seeds, seed germination, seedling survival and subsequent establishme nt in the environment prevailing specific to the geographic location. The factors responsible for establishment of new recruits vary among species, site conditions, and climatic factors among others and the identification of the limiting factors is necessa ry for improving conservation of endangered species (Kennard et al., 2002). Further, the relative importance of these factors in controlling and assisting regeneration and establishment is also important for formulating effective restoration strat egies. Wi th this knowledge, it may be possible to manipulate the microsite characteristics to prevent further decline of endangered taxa (Guariguata and Pinard, 1998). Unfortunately, the relative contribution of silvicultural treatments on microsite amelioration is unknown for many endangered tree species (Turnbull et al., 2000). The Red Sanders (RS; Pterocarpus santalinus L., Fabaceae) of southern In dia is one such species. RS forests are part of the tropical dry deciduous forests that are ecologically complex on account of seasonality in rainfall, temperature, and number of flora and fauna existing therein. Among others, the geobotany and the highly variable
33 precipitation interspersed with frequent dry spells are the main features of these forests. Literature revi ew reveals the scarcity of studies on establishment, survival and growth of seedlings in tropical dry forests in general (Khurana and Singh, 2001). Vieira and Scariot (2006) noted four to five times more studies for natural regeneration in tropical rainfor ests as compared to tropical dry forests and only 3% of restoration studies focused on tropical dry forest restoration. Several factors have been shown to be particularly significant for regeneration and restoration of tropical dry forests e.g., vegetation cover and light conditions (Auspurger, 1984; Negi, 1996), inter and intra specific competition (Harcombe, 1987), humidity (Howe, 1990), soil compaction (Unger and Cassel, 1991), soil nutrients (Aide and Cavelier, 1994), microsite or the micro topography ( Shibata and Nakashizuka, 1995), fire (Tyler, 1995; Hofmann, 1996), grazing (Khurana and Singh, 2001), drought (Bunker and Carson, 2005), invasive species (Awanyo, 2010) and anthropogenic disturbances (Khurana and Singh, 2001). Even though RS is one of the most important tree species in southern India, very little information on autecology of this species is available. To our knowledge there are no peer reviewed studies dealing with restoration of RS forests and this study is an attempt to fill the knowledge gap. Red Sanders The Species Endemic to southern India, RS forests are distributed between 13 30' 15 0' N latitude and 78 45' 79 39' E longitude (Raju et al., 1999). Within its natural distribution, it grows on approximately 500 km 2 of fragment ed forest landscape of Andhra Pradesh (AP), Karnataka and Tamil Nadu states (Jain and Rao, 1983; Molur and Walker, 1995; Oldfield et al., 1998). The total growing stock of RS in AP forests have been estimated at 118,000 m 3 (Andhra Pradesh Forest Department (APFD),
34 2009). The tree prefers 800 mm to 1200 mm rainfall, perfect drainage and elevations ranging from 200 m to 900 m (Ahmed and Nayar, 1984). It occurs almost exclusively on Quartzites (80%) and Shales (20%) (Nagaraju and Raju, 1998) and is a known acc umulator of Strontium and Cadmium (Raju and Rao, 1988). RS heartwood (henceforth wood) is renowned for its characteristic color, wavy grain, and superlative technical qualities as well as for its medicinal properties (Reddy, 1990). The wood has strong dema nd in Japan and China where it is primarily used for making expensive furniture and musical instruments. A demand supply gap, that exists due to ban on tree felling on public forests and lack of supply from private lands, has resulted in an increase in ill egal logging in public forests. The International Union for Conservation of Nature (IUCN) identifies illegal logging and habitat loss originating from severe biotic pressures as the biggest thre at to RS and has placed it on the endangered list (IUCN, 2010) The location of the RS forests is shown in Figure 1 1. Challenges of R estoration A slow growing tree, RS starts developing heartwood at somewhere between 15 and 20 years (Raju et al., 1999). Since the very beginning, demand for wood has been a continued threat to its existence and commercial exploitation without any commensurate restoration in the past has often been blamed for its present degraded state (Ahmad and Nayar, 1984). Additionally, poor seed germination (30% 40%), prolong ed droughts, recurren t wildfires and grazing are considered responsible for regeneration failure (Working Plan Kadapa, 2004). If the trend of degradation is not controlled, extinction of remnant fragmented RS populations may become imminent and
35 thus there is an identified need for developing a cost effective and practical restoration strategy 4 Restoration Strategies restoration strategies of practitioners due to its low cost and ease of implementation involved, this is especially suited to landscape level restoration. As part of ass isted natural regeneration studies effect of treatments on seedling recruitment and establishment has been studied for weeding (Parrotta and Knowles, 1999), removal of dead foliage (Marrero Gomez et al., 2000), canopy opening (Li, 2003), interplanting N f ixing crop (Carpenter et al., 2004), irrigation (Bunker and Carson, 2005), erosion control (Snchez Coronado et al., 2007), vegetation and grass control (Bouffard et al., 2007; Gunaratne et al., 2010), understory vegetation control (Dupuy and Chazdon, 2008 ), and tilling (Casselman et al., 2006; Gunaratne et al., 2010). M ost of these studies focus on eliminating the bottlenecks in regeneration and / or ameliorating the site conditions for the seedling establishment. Like most tropical dry forest species, re sprouting after disturbance is common in RS and tending these seedlings may be a good way to restore the RS forests. Several studies have identified resprouting in dry tropical forests as a shortcut to recovery as it eliminates more vulnerable stages like seed predation, seed desiccation and seedling survival (Kennard et al., 2002; Vieira and Scariot, 2006). Similarly, disking (tillage) can 4 For more details on challenges of restoration please see C hapter 1.
36 ameliorate the detrimental effects of compaction and improve seedling growth. Prescribed fire is yet another tool that can be used to control the fuel load to suppress wildfires. The local forest department provides restoration support to RS on public forests by removing weed growth and multiple shoots, climber cutting, and pruning. However, the impacts of these activitie s on survival and growth have not been studied. The objectives of this study are to evaluate the impacts of silvicultural treatments designed to ameliorate growth limiting factors on seedling s in the degraded RS forests. It is hypothesized that silvicultural treatments would have positive impact on survival and establishment of young RS seedlings. Further, the treatment that assures maximum moisture retention and fire protection would be best suited for restoration of degraded RS forests. Since e xpensive restoration strategies are difficult to im plement at a landscape level, a set of simple and cost effective treatments that reduce intra and inter specific competition, and improve moisture and nutrient availability, when used either alone or in co mbination were studied Material and Methods Study Site The study site is located at 14 31' 28.42" N and 79 0' 29.63" E, and 205 m elevation near Kadapa town of Andhra Pradesh, India. A d egraded forest area with good number of RS advance regeneration was selected for setting up the experiment The history of the area indicated fire damage during the previous summer which was evident by the presence of dead stems along with resprouted coppice shoots. These forests had been clear cut in the past and were open for firewood gathering, grazing, and non timber forest product collection by local communities. The low biodiversity of the study site is reflected in Shannon Weiner diversity index (1.37 1.89) and the
37 0.46). The site thus r epresents a typical degraded forest that has suffered on account of varied abiotic and biotic disturbances. As per t he total tree enumeration data collected by the local forest department in 2006, 85% of RS trees in forests were below 75 cm and less than 1 % of were about 100 cm girth at breast height (APFD, 2006). This indicated the present state of RS forests. The most prevalent tree species on the study site were Anogeissus latifolia, Chloroxylon swietenia, Dolichondron crispa, Grewia rotundifolia, Hardwi ckia binata, Lannea coromandelica, Terminalia pallid a and Zizyphus xylophyrus The understory vegetation besides Cymbopogon coloratus includes Azima tetracantha, Canthium parviflorum, Cynodon dactylon, Eragrostis viscosa, Hemidesmus indicus Heteropogon co ntortus, Hybanthus enneaspermus, Phyllanthus madarasapatensis, Rynchosia capitata, Mucuna pruriens, Sehima nervosum, and Trema orientalis, The mean rainfall of the area, 696 mm, is well below the state average of 925 mm. The monsoons are generally errati c with late start s and interspersed with long gaps, making survival of planted and naturally germinating seeds in forests difficult. Kadapa district generally experiences hot summer s and mild winters. Summer temperature reaches as high as 47 C whereas the lowest winter temperature is around 13 C. Mean monthly maximum and minimum temperatures (average of 1997 to 2007) were 41.2 C in April and 17.5 C in January, which resemble that of a tropical dry semiarid climate (Thornwaite, 1931; Feddema, 2005). RS Forests are tropical dry forests classified as 5A/C3 forest type (Champion and Seth, 1936). The soils of the RS landscape are generally Lixisols with shallow depth and mostly with very little humus (FAO, 2003). A n many places even the thin layer of topso il h as
38 been washed away and the sub soil is exposed. The soils of the study area were highly porous and leached, slightly acidic (pH = 5.9 6.6) with electrical conductivity ranging from 0.06 0.21 mmho s cm 1 Soil analysis revealed phosphorus (7.50 42.50 kg ha 1 ), potassium (150.00 258.00 kg ha 1 ), zinc (0.75 1 ), manganese (1.50 34.00 1 ), iron (37.00 1 ) and copper (0.15 1 ). Study Design This study used a completely randomized block design comprising of four blocks situated 60 m 250 m apart (Figure 2 1) The blocks of size 30 m x 50 m were selected so as to include sufficient number s of seedlings required for the experiment. In August 2008, each block was subdivided into 15 plots (10 m x 10 m) and basel ine data for experimental seedlings (height, root collar diameter (RCD), number of coppice shoots) w ere collected. Additionally, plot level data for other neighboring seedlings (numbers, average height, average RCD), trees (average height, average diameter at breast height (DBH), crown cover) and soil were collected. Treatment means of variables and covariates derived from baseline seedling data are shown in Table 2 1. RS trees are known to seed sporadically every y ear, but gregariously in every three to f ive years. P oor regeneration due to irregular seed production on account of greater climatic variability is more common in dry forests (Bullock, 1995; Sagar and Singh, 2003). Therefore, to account for the short term varia bility amongst the seedlings, recru its up to 3 year old were considered in this study. T o avoid inclusion of already established seedlings in the experiment b ased on local expertise, RCD of 3 cm was assumed as cut off level for seedling selection. In each plot 17 to 25 RS seedlings were se lected and a permanent identification number was allotted to each seedling. The vegetation with
39 diameter above 10 cm at 1.32 m was classified as trees. The following five treatments were used: 1. Prescribed Burn (PB): in this treatment, the understory vegetation (grasses and weeds) in the whole plot were removed annually once. Owing to the presence of essential oils, C. coloratus grass burns even when it is green. Therefore, the tall grass was clip ped and removed from the plot and the remaining vegetation was prescribed burnt during January. This strategy was followed to avoid exposure of seedlings to excess temperature. This is the least expensive treatment. 2. Disking with Prescribed Burn (DPB): in this treatment, soil in 50 cm radius area around the young RS regeneration was disked to a depth of 15 cm and all the weeds / vegetation removed. The understory vegetation present in between the seedlings in the plot was also removed by PB treatment. The d isked area allows percolation of water apart from protecting the seedlings from fire. 3. Singling with Prescribed Burn (SPB): in this treatment, before providing PB treatment, all the weaker and leaning multiple shoots of RS seedlings were removed leaving one leading shoot to grow. This treatment ensures more nutrients for the leading shoot and also improves the quality of the crop. 4. Singling plus Disking with Prescribed Burn (SDPB): in this combination treatment, singling was followed by disking and lastly, pr escribed burn was carried out. This treatment ensures increased protection, moisture availability and improved quality of seedlings in the forests. However, due to inclusion of many components this is also the most expensive treatment. 5. No treatment (NT): control plots. The treatments, replicated thrice within a block, were randomly allocated to the plots. They were replicated so as to capture the variability in survival and growth due to factors other than the treatment such as overstory density, seedling density, and soil. Different treatment operations were applied as per calendar of operations practiced by the local forest department. Prescribed burn was carried out annually once in January (2009 and 2010). Disking and singling was done once in 2008 (Nov ember), twice in 2009, (July and November) and finally onc e in 2010 (July). Final data were collected in August 2010 after two years of treatments Overall, 1449 RS seedlings were marked for the experiment and additional data for 48 trees and 536 other ne ighboring seedlings
40 p resent in the experiment area were recorded. For s oil analysis, five samples collected at 0 15 cm depth from all corners and center of each plot were mixed for a plot sample Soil analysis was done at the Biotechnology and Tree Impro vement (BIOTRIM) Center, Tirupati. The new emerging RS regeneration was ignored during the final data collection. The entire experiment area was protected from wildfires by creating a 10 m wide fire line, maintained throughout the study. The plots were not protected from wild animals, but protection from domestic cattle grazing was provided by banning such activities in the study area. Statistical Analyses For studying the effec ts of treatment on survival, logistic regression with random effects (PROC GLIM MIX; SAS version 9.2, SAS Institute Inc., Cary, North Carolina, USA) was used The p roportion of surviving seedlings in a plot (i.e. survived / total) was taken as the response variable to quantify and assess effect of treatments on survival. Since this st udy was conducted in a natural setting, other plot level variables (mentioned in section s earlier ) and their interactions with treatment were also included in analysis as covariates. Blocks were treated as random and the treatments as fixed effect. Followi ng Crawley (2002) the backward elimination approach was used to determine the best model. The model was run with main effects of treatment, covariates as well as two way interactions between treatments and covariates. The interactions between covariates were not included. Non significant covariates (P > 0.05) and th eir interactions were then sequentially removed retaining all relevant effects. A significant interaction demonstrates that the survival changes disproportionately between treatments as the covariates changes in value therefore, if interaction terms were not significant, only main effect means were compared. When the logistic regression found
41 significant interactions with treatment / plot level covariates, orthogonal contrast between the treatments at two or more values of covariates were used to determine the nature of interaction The categories were defined for RCD, height, and number of coppice shoots (Table 2 2). The m ean comparison s were adjustment with significance set at P < 0.05 for all comparisons. To quantify and assess the effects of treatments and their interaction with different covariates on growth a linear mixed model with covariates (PROC MIXED; SAS version 9.2, SAS Institute Inc., Cary, North Carolina, USA) was used. The procedures indicated by Brandeis et al., (20 00) were followed. The initial volume (Vol 0 ) and final Volume (Vol 1 ) for each experimental seedling was derived by assuming the leading shoot a right angular circular cone. Volume Where, RCD is the root diameter collar and Ht is the height of the plant in cm. The volume calculations were used to calculate the absolute volume growth D_Vol for each experimental seedling for analyses. D_Vol = Vol 1 Vol 0 A s imilar method was used to calculate the absolute height growth ( D_Ht ), an d RCD growth ( D_RCD ). The RCD was found to be highly correlated with height 0.001) and the average number of trees in each plot was highly correlated with diameter at breast height (DBH) of trees and with height growt h = 0.82, P < 0.001). As expected, DBH and height of trees and RCD and height of other seedlings were highly correlated. The variables which contributed in lowering the AIC value of the model the most were us ed
42 in selecting the final model. For the growth study D_Ht D_RCD and D_VOL were used as responses for analyzing effect of various treatments and covariates. E xploratory analyses revealed non normal distribution of residual errors which was tackled by log transformation of the response variables. The final adjusted mean of different factors were back transformed to the original scale for presentation. The e xploratory analysis of soil data did not reveal any significant differences amongst the samples and he nce were excluded from the study. Results Survival The effect of treatments on survival was significant ( P = 0.008 Table 2 3). The seedlings with DPB, SDPB and PB treatments indicated 96%, 94% and 87% survivals respectively. Survival of seedlings was also dependent on number of coppice shoots (N_Cop) on the seedlings, and on height of other seedlings (Ht_Others) present in the plot. The interaction of treatment and height of other seedlings also significantly influenced the survival. M ean comparison correct ed by Bonferroni indicated significant differences between NT DPB, DPB SPB and SPB SDPB pairs of treatments. However, survi val amongst seedlings with DPB and SDPB were not statistically different. The s eedlings with fewer coppice shoots showed statistically signifi cant higher survival as compared to seedlings with higher (P = 0.04 6 ) and medium (P = 0.050) number of coppice sho ots. The mean survivals amongst seedlings in the stump size and height categories were not statistically different. The survival results and effect of number of coppice shoots is shown in Figure 2 2
43 Growth Height growth The treatment effect on height growth of seedlings was significant ( P < 0.001 Table 2 3). Relative to control, the SPB, SDPB and DPB treatments resulted in 42%, 31% and 22% more height growth (Table 2 4). The mean height growth comparisons with control indicated significantly higher results with SPB ( P < 0.001) and SDPB (P = 0.019) with DPB treatments. However, none of these three treatments were significantly different from one another. Height growth was also influenced by num ber of coppice shoots, initial RCD, average number of trees and other seedlings in the plot, and their height. Interactions of treatment with number of trees in the plot, and with block were also significant. Mean compari sons of absolute growth for all th ree stump size categories were statistically significant (P < 0.001) The se edlings with larger stumps experience d more absolute growth than medium and small stump size seedlings in general and specifically with the SPB and PB treatments. The mean differen ces by initial height categories were also statistically significant (P < 0.001) and taller seedlings accrued relatively more absolute height growth, compared to medium and smaller seedling s The number of coppice shoots did not have any clear effect on th e height growth. The results of height growth are shown in Figure 2 3 Root Collar Diameter growth The effect of treatments on RCD growth was highly significant ( P < 0.001, Table 2 3). RCD growth was influenced by initial height and number of trees in the plot. Initial height RCD and their interaction with treatment were also significant. All treatments resulted in positive RCD growth compared to control. Relative to control seedlings with
44 SDPB, DPB and SPB treatments resulted in 85%, 79% and 67% more statistically highly significant RCD growth respectively (Table 2 4). However, the growth means of these treatments were statistically not different from one another. Mean RCD growth comparisons for medium small and large small stump size categories were statistically significant (P < 0.001) However, the growth accrued by large medium RCD categories was not different The s mall stump sized seedlings show ed more absolute growth as compared to medium and high categories for all treatments. Mean difference s for absolute RCD growth, for all the height categories were statistically highly significant (P < 0.001) and tall seedlings gained more absolute growth for all the treatments. The RCD growth for seedlings with large number of coppice shoots was significa ntly different from seedlings with medium (P < 0.030) and fewer (P < 0.001) number of coppice shoots. The RCD growth of seedlings with medium and fewer coppice shoots was not different. The results of RCD growth are shown in Figure 2 4 Volume growth The e ffect of treatments on volume growth was significant ( P < 0.001, Table 2 3). All the treatments resulted in positive and significantly different mean volume growth as compared to NT (Table 2 4). Seedlings with SDPB, DPB and SPB g ained 97%, 87% and 71% resp ectively more volume growth as compared to control (Table 2 4). However, mean volume growth induced by these three treatments were not significantly different. Volume growth was influenced by initial height of seedlings, number of coppice shoots on seedlin gs and DBH of other trees. There was significant initial height and treatment interaction.
45 The mean differences for the thr ee stump size and height categories were statistically significant (P < 0.001) The tall and large stump size seedlings accrued high est volume growth with PB treatment. The volume growth in seedlings with many coppice shoots was significantly different from those with fewer number of coppice shoots (P = 0.02 3 ) The volume growth results are shown in Figure 2 5 Discussion Effect of Treatments on S urvival s The results of this study show that silvicultural treatments can ameliorate the microsite conditions leading to better survival of RS seedlings. The overall survival ranged from 81% to 96 % which is relatively high compared to the t ropical dry forest conditions. Analysis of the rainfall data revealed 21% more rainfall in the district during the study period. Availability of moisture is known to improve seedling survival in tropical dry forest seedlings (Gerhardt, 1996a). Better survi val across all the treatment and seedling categories can be attributed to better weather conditions prevailing during the study period. Long term studies interspersed with adverse seasonal conditions may be needed to capture the effects of different treatm ents on survival in the long run. The mean su rvivals for the treatments with disking (DPB, SDPB) were higher relative to the similar treatments without disking (PB, DPB), which was expected due to improved percolation and moisture availability for a longer duration. However, only the SDPB DPB survi val difference was significant. The effect of disking on seedling survival in tropical dr y forest in general and on RS is not well documented. Long term studies may be required to understand the effect of disking on survival. However, in Mediterranean environments, similar positive effects of disking on forest species by use of medium tillage have been recorded by Fonseca et al. (2011). Survival amongst seedlings with
46 fewer coppice shoots was significantly different from that of seedlings with medium and high number of coppice shoots. RS is a vigorous coppicer and every disturbance to seedlings results in resprouting (Mulliken and Crofton, 2008, R amakrishna, 1962). Due to the bushy nature of the sprouts, mortality due to intraspecific competition, grazing and fire in the initial years is severe. Therefore, it is not surprising that seedling s with fewer coppice shoots showed highest survival. Howeve r, removal of excess coppice shoots by singling did not improve survival. For instance the survival of seedlings with SPB and SDPB treatments is not significantly different from seedlings with PB and DPB treatment respectively. This indicates that nutrient availability was not a limiting factor for the younger RS seedlings with higher number of coppice shoots. In this study initial height and initial RCD did not have any impact on survival. This is contrary to the finding s of other studies wherein taller seedlings with higher RCD were shown to have better survival (Gerhardt, 1996a) and can be attributed to better survival of smaller seedlings due to better rainfall conditions prevailing during the study period. The only other significant factor that had a negative impact on survival was height of other seedlings in the plot. Although shade may assist in seedling emergence and early establishment, it is known to adversely impact the growth and survival during and after establishment (McLaren and McDonald, 2 0 03; Marod et al., 2004). In this study, height and RCD of other competing seedlings were negatively correlated with survival, which is consistent with the effect of competition and the strong light demanding behavior of RS seedlings. Since higher average R CD and height of other seedlings in the plot is a measure of increased competition for the younger seedlings, removal of competing seedlings of other species might help in improving survival. Improvement in survival and
47 growth by reducing number of competi ng seedlings has been noticed in several species like Pinus palustris (Demers et al., 2000; Knapp, 2008), Pinus ponderosa (Kolb and Robberecht, 1996) Pinus lambertiana and Abies concolor (Plamboeck et al., 2008). Effect of Treatments on Seedling Growth The study results suggest that it is possible to enhance RS seedlings growth (and hence establishment) with the help of silvicultural interventions ( Table 2 5 ) Taller seedlings accrued better absolute growth compared to smaller seedlings due to their well developed root system. Well developed root system is known to help in survival of species during establishment phase. For instance, the Pinus palustris seedlings remain in grass stage for a long time during which the seedlings strengthen the root system ( Jose et al., 2003). Only after the root system is sufficiently strong does the height growth take place. Similar survival and height growth results for tall seedlings were reported for Swietenia macrophylla by Gerhardt, (1996 a, 1996b ). In cas e of RCD growt h, contrary to this study hypothesis, smaller stump size seedling s gained more absolute RCD growth as compared to medium and higher categories. One possible explanation to this could be the preferable resource allocation to root growth by the seedlings at younger age as a strategy to overcome adverse seasonal conditions and thus resulting in higher RCD growth Higher allocation to root system has been reported for woody resprouting species facing adverse seasonal conditions (Schutz et al, 2009; Paula and Pa usas 2010). It was hypothesized that seedlings with fewer coppice shoots and seedling s with singling treatment would show better survival and gain more growth. This is due to the fact that t he coppice shoots are known to draw moisture and nutrients through the pre established root system, which improves survival of coppice shoots during dry periods
48 (Gerhardt, 1993; Kennard et al., 2002). However, in this study although the seedlings with fewe r coppice shoots showed better survival, they did not show hi gher growth as compared to seedlings with medium and higher number of coppice shoots This can be attributed to two reasons; first, it is possible that nutrient and moisture availability was not a limiting factor for the site and second, such differences might become obvious with increased duration of stud y. The discussion in earlier sections points towards availability of higher nutrient and moisture availability due to better rainfall during the period of study. On the other hand, seedlings responded well to singling treatment and showed better growth than the non singling treatments in all cases. For instance seedlings with SPB treatment showed 7%, 3%, and 6% more height, RCD and vo lume growth r espectively than seedlings with PB treatment. Similarly seedlings with SDPB treatment showed 26%, 32% and 52% more height, RCD and vo lume growth as compared to seedlings with DPB treatment. Resprouting after disturbances is more common in seasonally dry t ropical forests (Ewel, 1980; Kennard et al., 2002, Vieira and Scariot, 2006). However use of coppice shoots in forest restoration is poorly studied (Bunker and Carson, 2005 ). These results indicate that singling of multiple coppice shoots can help in estab lishing and the restoration of RS forests. The growth in height was influenced by the number and height of other seedlings and presence of trees in the plot. The presence of trees in the plot also influenced RCD growth. The number of trees and height of o ther seedlings in the plot were negatively correlated to increment in height and RCD and hence reduced tree cover and lower competition was considered good for RS seedling growth and establishment. This is consistent with a strong light demanding nature of RS and also in agreement with other
49 studies having similar results (Honu and Dang, 2002). On the other hand, height growth was positively correlated with number of other seedlings indicating that surrounding seedlings induce height growth due to increased competition for sunlight. The s ummary of growth and treatments shown in table 4 5 indicates no single effective treatment for different class es of seedlings. There were no significant difference s among the growth promoted by DPB, SPB and SDPB treatments. Disking has been shown to positively influence on seedling growth due to increased porosity and water holding capacity (Carlson et al., 2006, Godf roid et al., 2007). In this study disking has positive e ffect on survival as well as on RCD and volume growth Relative to PB the seedlings with DPB treatment record ed higher height (8%), RCD (42%) and volume (65%) growth. Similarly relative to SPB treatment, SDPB treatments gained higher RCD (8%) and Volume (15%) growth. Amongst the higher growth, only the RCD and volume growth in seedlings with DPB pair ware statistically different from PB. D isking improves soil aeration, decreases bulk density and improves infiltration (Espinoza, 2004) and should be used on a site specific basis keeping in view the root growth impeding characteristics of soil. In case the physical properties of soil do not adversely affect root growth, disking may not be necessary. Such decisions are important fro m the economic point of view as well. Further research for ascertaining the effect of disking on seedling growth in different soils types is needed to confirm this finding. In this study, the PB treatment was applied by clipping and removing the excess ve getation followed by burning of the leftover. Such strategies have been successfully used in restoring the Pinus palustris overstory (Johnson and Gjerstad, 2006), in
50 controlling Festuca arundinacea (Tall Fescue) (Madison et al., 2001) and Phalaris arundinacea (Reed canary grass) (Adams and Galatowitsch, 2006) in southeast United States. Since the grasses are fast growing, physical removal by cutting or by herbicide treatment prior to burn season can be used to avoid fuel buildup. In normal course of forest management prescribed burn is applied without removal of excess ground fuel. These facts should be kept in mind while replicating the study. Dwarf and smaller stump size seedlings with fewer coppice shoots accrued maximum relative g rowth in height RCD and volume This indicates that in terms of relative growth, older seedlings do not have any advantage over the younger ones and hence, wherever the density of RS seedlings is a limiting factor, smaller seedlings should be preferred for silvicultural treatment. This study is limited by time and number of treatments involved. Further long term research is necessary to improve, refine and formulate cost effective restoration strategies for this species. Summary and Prescriptions This study provides impo rtant insights on the interaction of silvicultural treatments with ecological factors to influence seedling survival and establishment and can be used to draw recommendations for restoration of RS or similar endangered species Tall and larger stump size s eedlings experienced better height and volume growth with SPB and SDPB treatments respectively. Tall seedlings and seedlings with smaller stump size showed better root collar diameter growth with SDPB treatment. The number of coppice shoots did not have an y impact on growth but seedlings with fewer coppice shoots showed better survival with DPB treatment. These results indicate that in case seedling density is not limited preferential treatment of taller seedlings may result in better establishment However if the seedling density is limited, preferential treatment to
51 younger seedlings may result in better growth RS seedlings with many coppice shoots gained more growth with singling indicating its usefulness in improving growth. Removal of coppice shoots i s good not only for survival and growth, but also for future quality of the RS forests as it improves the quality of trees in the stand Although seedlings showed better growth with disking the results are also influenced by other factors like rainfall and bulk density. The endangered and endemic nature of RS demands that the species be protected and conserved at the landscape level. In this study e ven though a relation between survival, growth and treatments is found, i t is difficult to pick the most effective treatment for no single treatment is effective in promoting significantly better growth and survival Hence, restoration activity will have to be tailored to this landscape at different levels of management based o n seedling density, biotic and abiotic factors, and soil type. Besides available infrastructure, budget and other logistic limitations must also be taken into account to select the best possible combination of tools for restoration of each unit.
52 Table 2 1 Description of variables (means SE) by treatment for see dlings used in survival study (N = 126 3 ) Variables Label (Plot means) NT (n = 23 6 ) PB (n = 24 8 ) SPB (n = 27 6 ) DPB (n = 23 3 ) SDPB (n = 270) All seedlings (n = 126 3 ) N_Cop Number of coppice shoots 2.37 0.16 3.01 0.18 2.92 0.24 3.09 0.13 2.78 0.23 2.83 0.08 RCD0 (cm) Initial r oot c ollar diameter 0.97 0.09 1.05 0.06 0.97 0.10 1.08 0.09 0.93 0.09 0.99 0.03 Ht0 (cm) Initial height 63.96 6.54 72.00 4.51 66.17 5.58 72.81 6.37 67.12 5.31 68.44 2.50 N_Trees Number of t rees in p lot 0.50 0.26 0.50 0.23 1.00 0.30 0.75 0.18 1.25 0.33 0.80 0.12 DBH_Trees (cm) DBH of trees in plot 4.89 2.14 5.21 2.34 7.90 2.06 8.83 1.98 9.09 1.97 7.18 0.93 Ht_Trees (m) Ht of trees in plot 2.28 1.03 2.56 1.12 4.10 1.14 4.07 0.94 4.47 0.97 3.49 0.46 N_Others Number of other seedlings 6.50 1.17 8.17 1.06 8.33 0.91 11.17 1.70 10.50 2.15 8.93 0.67 RCD_Others (cm) RCD of other seedlings 6.72 0.69 6.06 0.63 6.63 0.94 5.51 0.68 7.19 0.93 6.19 0.28 HT_Others (cm) Ht. of other seedlings 3.31 0.23 3.13 0.26 3.66 0.38 2.81 0.22 3.81 0.43 3.34 0.14 Density (%) Crown density in the plot 51.87 4.39 42.43 4.73 47.22 5.52 43.15 5.19 49.80 3.35 46.89 2.08 NT = no treatment, PB = prescribed burn, DPB = disking with prescribed burn, SPB = singling with prescribed burn, and SDPB = singling plus disking with prescribed burn.
53 Table 2 2 Description of categorical variables for study of treatment effects Continuous Variable Categorical Variable Survival Study Growth Study Low Medium High Low Medium High RCD0 St_Size (cm) < 0. 84 0. 84 1 15 > 1 15 < 0. 68 0. 68 1 30 > 1 40 Ht0 Ini_Ht (cm) < 58 58 79 > 79 < 50 50 100 > 100 N_Cop Copp (Numbers) < 2.44 2.44 3.12 > 3.12 < 3 3 4 > 4 RCD0 = i nitial root collar diameter, Ht0 = initial height, N_Cop = n umbe r of coppice shoots, St_size = s tump size category Ini_Ht = initial height category, and Copp = coppice category Table 2 3 Results for survival and growth study analyses Source # Degrees of Freedom Variable (P >F) Survival D_Ht D_RCD D_VOL Block 3 0.009 < 0.001 0.003 N_Cop 1 0.056 < 0.001 0.007 RCD0 1 < 0.001 Ht0 1 < 0.001 < 0.001 N_Trees 1 < 0.001 0.006 0.084 N_Others 1 0.002 Ht_Others 1 0.039 0.042 0.321 Trt 4 0.008 < 0.001 < 0.001 < 0.001 Ht0*Trt 4 < 0.001 0.027 Ht_Others*Trt 4 0.007 N_Trees*Trt 4 0.003 Block*Trt (15 ) 12 0.3 97 0.004 0.237 0.255 # O nly the variables in the final model are displayed. The blank ( ) denotes absence from the model. Degrees of freedom for the survival study only. Survival variables are plot level averages. D_Ht = Height growth, D_RCD = root collar diameter growth, D_VOL = Volume growth, N_Cop = number of coppice shoots, RCD0 = initial root collar diameter, Ht0 = initial height, N_Trees = number of trees in the plot, DBH_trees = diameter at breast height for trees, N_Others = number of other seedlings in the plot, Ht_Others = height of other seedlings, and Trt = Treatment factor.
54 Table 2 4 R esults for different treatments (mean SE) Treatments Survival proportion D_Ht (cm) D_RCD (cm) D_Vol (cm 3 ) NT 0.82 0.04 14.65 0.73 0.37 0.02 21.43 1.49 PB 0.87 0.03 16.59 0.73 0.46 0.02 24.23 1.63 DPB 0.96 0.02 17.92 0.76 0.66 0.03 40.03 2.66 SPB 0.81 0.04 20.87 1.09 0.61 0.03 36.73 2.62 SDPB 0.94 0.02 19.21 0.86 0.68 0.03 42.30 2.82 D_Ht = Height growth, D_RCD = R oot collar diameter growth, D_VOL = Volume growth, NT = no treatment, PB = prescribed burn, DPB = disking with prescribed burn, SPB = singling with prescribed burn, and SDPB = singling plus disking with prescribed burn. Table 2 5 Survival and growth by seedling characteristics and favored treatments Item Stump size Initial height Number of coppice Favored T reatment s* Height Growth Large Tall SPB, SDPB RCD G rowth Small Tall Many SDPB, DPB Volume Growth Large Tall Many SDPB, DPB Survival Few DPB, SDPB, *First and second best results shown, DPB = disking with prescribed burn, SPB = singling with prescribed burn, and SDPB = singling plus disking with prescribed burn.
55 Figure 2 1 Study design Figure 2 2 Treatment vs. Survival proportion and effect of number of coppice shoots on survival M eans not sharing the same letter are significantly different. Mean correction by Bonferroni (P= 0.05)
56 Figure 2 3 Treatment vs. Height growth (Columns) and effect of (a) stump size (b) Initial height (c) Number of coppice shoots (Lines) on height growth Means not sharing the same letter are significantly different. Mean correction by Bonferroni (P = 0.05)
57 Figure 2 4 Treatments vs Root collar diameter (RCD) growth (columns) and effect of (a) stump size (b) Initial height (c) Number of coppice shoots (Lines) on RCD growth Means not sharing the same letter are significantly different. Me an correction by Bonferroni ( P = 0.05)
58 Figure 2 5 Treatment vs. Volume growth (Columns) and effect of (a) stump size (b) Initial height (c) Number of coppice shoots (Lines) on volume growth T reatment means not sharing the same letter are significantly different. Mean correction by Bonferroni (P = 0.05)
59 CHAPTER 3 ENERGETIC, ENVIRONME NTAL AND ECONOMIC EV ALUATION OF ETHANOL PRODUCTION FROM FORE ST GRASS BIOMASS IN INDIA A LIFE CYCLE ASSESSMENT APPROACH I ndia Lignocellulosic Ethanol Scenario and depends heavily on oil imports to meet its domestic demand. To reduce its dependence on fossil fuels India has launched research and development of alternate renewable fuel resources. As a part of the fossil fuel reduction strategy, 5% ethanol blend gasoline was made mandatory in 11 states and three union territories during 2004, with a proposed increase to 10% during 2011 12 (Planning Commission, 2003). In 2008, with the total gasoline demand of 15.23 billion liters, the corresponding blending targe ts of 5% and 10% were equivalent to 0.77 and 1.53 billion liters of ethanol respectively. India produces 2.7 billion liters of ethanol from molasses, a byproduct of the sugar industry. The liquor and chemical industry accounts for substantial consumption a nd the surplus (AIDA, 2006; Sukumaran and Pandey, 2009; Sukumaran et al., 2010). Production of ethanol from lignocellulosic biomass and its use as blended gasoline fuel for transport has been recognized as one of the promising routes of reducing gasoline consumption. More recently, Godavari Sugar Mills 5 and Praj Industries 6 have set up pilot plant for production of ethanol from in Maharashtra and Karnataka states of India. The former uses only su garcane bagasse as biomass, the latter uses corn stover and sugarcane bagasse both. Keeping in view the huge population and associated 5 http://www.biofuelsdigest.com/blog2/2008/09/01/ 6 http://www.biofuelsdigest.com/blog2/2009/02/11/
60 large demand for food it will be prudent for India to explore feedstocks that do not have food or feed value to augment its ethanol production. Use of agricultural residues is an option, but the actual availability of these residues for ethanol production is not assured due t o their consumption as fodder, fuel, manure, and by the paper industry (Pandey et al., 2009). Therefore, in order to meet future demand for ethanol, wherever possible, new resources need to be identified, e valuated and developed at local and regional leve ls Use of forest biomass is one such area which has not yet been explored for its availability and environmental and economic feasibility of its conversion into ethanol. This is primarily because forest policy and legislation in India do es not allow comm ercial use of public forests by private enterprise and the role of Government owned corporations 7 in commercial activity is very limited The future policy with respect to raising bioenergy crops on forestland notwithstanding, certain lignocellulosic forest resources like forest grasses are available to the communities for free collection. Although forest biomass itself may be cheap, the c osts of its collection and processing may be relatively high as compared to dedicated bioenergy crops. Furthermore, in order to be a viable energy source, the ethanol derived must be environmentally benign and economically competitive relative to the ethan ol produced from other feedstocks available in India. Thus the selection of feedstock for lignocellulosic ethanol technology itself needs careful analysis 7 Forest Development Corporations in India are quasi government organization s esta blished by the government as a public sector company registered under the Indian Companies Act in 1975. For example AP Forest Development Corporation (APFDC). For more information please see http: //apfdc.apts.gov.in/
61 Bodagrass (BG; Cymbopogon coloratus ) is one such abundantly available resource in the Red Sanders f orests of southern Andhra Pradesh (AP), India. The conventional use of this grass as thatch material was discontinued by communities on the advent of permanent housing. Lying unused, it becomes fuel load for wildfires causing mortality in young forest rege neration. As such, removal of BG from forests may be beneficial for restoration of Red Sanders forests. Considering the potential of Red Sanders forests of AP in meeting the supply of BG on a small scal e, this case study evaluates the environmental and eco nomic suitability of its utilization for ethanol L ife cycle assessment (LCA) and benefit cost analysis are applied to achieve the task for assessing systems including pro ducts or processes. In ISO 14040, LCA is defined It has been extensively used for comparison of ene rgy and greenhouse gas balances of several biofuels with conventional fossil fuels. In the Indian context, fuel ethanol production from molasses was studied by Prakash et al. (1998) and from sugarcane bagasse by Kadam (2000). Literature review did not reve al studies on the use of forest grass biomass for ethanol in India. This case study is an attempt to evaluate the use of BG for ethanol as a transportation fuel on a small scale. The LCA process and inventory are discussed in the next section. The third se ction deals with the economics of ethanol production and in the fourth section sensitivity, uncertainty and risk are discussed. The LCA impacts and interpretations are discussed sequentially in the fifth section. The sixth section details the results and d iscussions and the last section concludes the study.
62 Methods This LCA is structured in accordance with International Organization for Standardization (ISO) standard # 14044: 2006 guidelines (ISO 2006), which requires four phases: (1) definition of th e assessment goals and scope, (2) inventory of all material and energy inputs and outputs from the system, (3) assessment of environmental impacts associated with the inventoried system inputs and outputs, and (4) interpretation of the impacts according to the defined goal and scope. The goal and scope stage is vital in determining the system boundaries and objectives of the LCA. The system boundaries are the limits placed on data collection for the study and can influence the outcome of the LCA. This first stage also specifies the functional unit (FU) of the LCA. The choice of correct functional unit is important for meaningful comparisons. One of the main purposes for a functional unit is to provide a reference to which the input and output data are normal ized. Inventorying the inputs (e.g., raw materials and energy) and outputs (e.g., products, byproducts, waste and emissions) comprises the second stage of an LCA. The impact assessment can be defined as the cterize and assess the effects of the divided into three parts: classification, characterization and valuation. The classification stage links the inputs and outputs distinguished during the inventory process to corresponding environmental impacts. The characterization stage quantifies impacts and determines the effect of inputs and outputs on the impact categories. In the final stage; valuation weighs impacts giving r elative importance to each category so that a single index indicating environmental performance can be calculated (Allen and Shonnard, 2002). Interpretation or improvement analysis is the last stage of an LCA. It
63 involves identification of significant envi ronmental issues, evaluation of these issues and drawing conclusions. These phases are described in more detail below with specific reference to this study. Goal and Scope The goals of this LCA are to assess, evaluate, quantify and compare the net energeti c and environmental flows associated with conversion of BG feedstock into ethanol over its entire life cycle including its subsequent use as E10 (10% ethanol and 90% leaded gasoline) motor fuel. The energetic efficiency as net energy gain and net energy ratio were calculated. Emissions and associated environmental impacts were evaluated for categories of (1) global warming potential, (2) acidification, (3) smog formation, (4) human h ealth (cancer) and (5) water use. The net environmental burdens caused by production of ethanol and its use as E10 on energy equivalent basis in vehicles (S cenario B) were compared with the reference case of open burning of BG in forests and use of gasolin e as motor fuel (Scenario A). The economic benefits accrued to the communities and cost of ethanol were also estimated. Although, ethanol production technologies ha ve developed significantly in recent years, India still lacks mature technologies (Sukumaran et al., 2010). In this study the two stage dilute acid technology developed by National Renewal Energy Laboratory (NREL) was considered for ethanol production (Aden et al., 2002). The two stage dilute sulphuric acid process is well established though no t the most cost effective technology for production of ethanol. In terms of geographical scope, this study is restricted to areas with Red Sanders forests in southern AP, India where BG is abundant. However, this study can be replicated for all such fores ts where removal of grass biomass is beneficial for the health of forests.
64 System Boundaries A broad based system begins with raw material acquisition and continues through the production process, consumption and final disposition. The system that was ana lyzed included harvest of BG, transport of feedstock, production of ethanol, blending, transportation of E10 and finally its use as motor fuel. Unlike developed countries, India still relies on manual labor for harvest of agriculture crops, so harvest of B G was assumed to have been carried out manually. Inclusion of labor energy is a controversial issue in LCA and the methods of energy estimation differ to a great extent. In this study the labor energy is a major component for harvest ing biomass. Hence it h as been included in addition to the primary energy (electricity and diesel) inputs in the conversion process. However, as is the case for most other studies, inclusion of t he labor component elsewhere for conversion and transport has been avoided. The embo died energy of machinery, other chemicals and materials used were a lso included. Embodied energy is the amount of energy (MJ) necessary to produce an unit of a material. I nfrastructure, production related raw materials, energy, emissions, and impacts were assessed to get a complete understanding of the benefits or drawbacks of both systems that are being compared. Figure 3 1 shows the general system boundaries for this study. Functional Unit The functional unit for this study is the production of one liter of ethanol. The derived ethanol is used for blended gasoline E10 which replaces the gasoline on energy equivalent basis. The net emissions arising from ethanol production and its use as motor fuel as E10 (Scenario B) was compared with the status quo sit uation of using gasoline as motor fuel and burning of BG in forests (Scenario A).
65 Assumptions In addition to the scope and system boundaries defined above, several assumptions about the system parameters are necessary to facilitate data collection and calculations. The key assumptions of this study are as follows: The conversion process, productivity, equipment costs, and other parameter values taken from the NREL and other literature sources applicable to the Indian scenario. The emissions for open bu rning of BG along with other forest forage in forests are considered equal to the burning of dry tropical forests. The electricity generated at the mill is considered a coproduct. The emissions avoided were calculated as replacement of coal in electricity generation. The BG biomass is a self propagating resource and there is no loss of productivity on account of harvest. The energy involved in on site production of enzyme and yeast were considered negligible and hence excluded from the study. For economic analysis, costs of all capital goods and services are constant during the span of the study. The value of electricity sold to the grid was not included in economic analysis. Inventory The life cycle inventory (LCI) calculations were performed using SimaP ro 7.1 (Pre Consultants, Amersfoort, The Netherlands). However, rather than using existing databases in SimaPro, all data process modules were created and customized using Excel This was done to maintain a better level of transparency and specificity to the system of interest as well as to achieve consistent data quality. Primary data pertaining to BG biomass weight, moisture content, labor requirements, and labor costs were col lected directly from a study area situated in Chittoor district of AP state where this grass is abundant. Secondary data related to production processes, chemicals and
66 materials used, emissions, and costs are derived from secondary sources along with surro gate data. The data used and sources are summarized in Table 3 1. Biomass Characteristics and H arvest For the purpose of biomass estimation four plots of 4m*4m were laid randomly in the Red Sanders forests. The harvests of grass per person day data were c ollected by calculating average harvest from 10 person days harvest It was estimated that 5.24 persondays per dry tonne of BG were required. From the sample plots, it was estimated that 13.71 t ha 1 mature BG can be collected. There was 31% moisture cont ent which indicated 9.46 t ha 1 dry weight. T o make up for the moisture content, 1.31 t of green BG was harvested per tonne dry BG biomass. A tonne of green feedstock cost at the loading point from where it will be t ransported to the ethanol mill was estima ted to be INR 983.00 ($21.83) Labor energy for harvest of BG was based on the Life Style Support Energy (LSSE) method recommended by Odum (1983) and used by Nguyen et al. (2007) its cost by the average energy to monetary unit ratio or energy intensity of the economy. Using energy intensity data NREGA, 2009), the energy value of labor in AP was derived as 5.77 MJ h 1 With an 8 hour workday, total energy required for harvest of one tonne dry BG biomass was estimated to be 242 MJ. T he energy input in this category was then segregated into fossil (51%) and non fossil (49%) energy items, based on primary energy generation by fuel s ources in AP (APGenco, 2010). The data on chemical composition of grass w ere provided by Kumarappa handmade paper institute, Jaipur, India. The total growing stock of BG in AP were estimated by the local forest department at 0.9 million tonne annually (APF D, 2010).
67 Transportation There are two transportation steps within the system boundary. First it is assumed that the transportation of feedstock is done by trucks. It was assumed that the average distance from the forest boundary to production mill was 12 0 km. After production, it is assum ed that the reformulated gasoline (E10) is transported 80 km average distance before it is utilized The distances from the forest boundary to the mill and then to a point of utilization were calculated based on averages. The fuel consumption for biomass transport to produce one functional unit was calculated based on the data by Khatiwada and Silveira, (2009) in Nepal. The energy used in feedstock and ethanol transport was estimated at 0.55 MJ L 1 and 0.10 MJ L 1 respectively. Ethanol P roduction The two stage dilute sulphuric acid process for conversion of lignocellulosic biomass to ethanol requires pretreatment to hydrolyze the hemicelluloses and cellulose by dilute sulphuric acid to produce sugars. The pretre ated biomass is then processed through a solid liquid separator so that the liquid stream can be detoxified by adding lime to neutralize the sulfuric acid. The sulfuric acid is removed to avoid the subsequent enzymatic hydrolysis and fermentation reactions Following detoxification, the liquid stream is mixed back with the solid stream from prehydrolysis and processed in a series of SSCF (simultaneous saccharification and co fermentation) reactors. A portion of the sugar and solid stream from prehydrolysis is used to produce enzymes on site, which are added with enzymes into reactors to produce beer. Following the SSCF reactors, the biomass beer is processed through distillation and dehydration to produce the final ethanol. During the process gypsum and CO 2 are produced and lignin is left as residue. Lignin is used for onsite production of steam and electricity tha t can be used by the
68 plant. Any surplus electricity can be supplied to the grid. The flow diagram of the process is shown in Figure 3 2. The chem ical composition of BG is 76.6% h olocellulose, 16% lignin, 6.78% a sh and 0.62% extractives. The literature review indicates use of different hydrolysis efficiencies by various studies. The NREL two stage dilute sulphuric acid study (Aden et al., 2002) usi ng corn stover as biomass, has shown hydrolysis efficiency of cellulose and hemicelluloses at 90% and fermentation efficiencies of glucose at 90% and x ylose (2008) used average conversion efficienc ies for cellulose and hemicelluloses of 76% and 51% respectively and subseque nt fermentation efficiency for Glucose at 75% and Xylose at 50% This study used the latter ( lower ) conversion efficiencies for production of ethanol and the former study efficien cies were used for sensitivity analysis. Using conservative conversion efficiency, the production of ethanol was estimated at 359 L t 1 t 1 ) of dry BG biomass, which is comparatively less than the theoretically achievable yield (USDOE, 2010). The reasons for low output can be attributed to the old technology being assumed for this study. At lower efficiencies the energy balance of the production process is essentially breakeven or even positive, indicating more energy production than use. Coughlin and Fridley (2008) sho wed that at higher recovery and fermentation efficiencies, the quantities of undigested cellulose in lignin are reduced and thus lesser energy is recovered in the production system. This necessitates import of additional energy from the grid thus increasin g the input energy and sometimes resulting in net negative energy balance. For cellulosic ethanol production to be energetically viable, it
69 may be more important to focus on the processing energy consumption than to improve the efficiencies of the biochemi cal processes involved. Energy P roduction The main energy requirements at the plant are process energy for running the mill, heat required for production of steam and energy required for wastewater treatment. There is considerable debate on the energy requ irements of the ethanol mill. S ome studies have shown the entire energy requirement of the mill being offset by the residual biomass (Shee han, 2004; Schmer et al., 2007), but other studies challenge d these findings (Pimentel and Patzek, 2005; Coughlin and Fridley, 2008). For this study, assuming 60% water utilization and 80% boiler efficiency, the energy required for steam generation was calculated and the total requirement was estimated at 3.3 MJ kg 1 of water. The energy used was higher than 2.6 MJ kg 1 e nthalpy of steam required at approximately 150 psi and 160C which can be attributed to lower process efficiency. Although higher value products such as organic acids, phenols and vanillin by products from lignin can significantly enhance the competitivene ss of ethanol technolo gy (Hamelinck et al., 2005), this study considers lignin only for power generation. The steam generated by burning lignin was able to completely offset the energy needed at the mill for heating besides electricity generation. Assuming 40% c onversion efficiency net surplus electricity of 0.16 MJ per liter ethanol was generated, which was assumed to be sol d to the grid. The wastewater can be treated by using 1.49 MJ m 3 of energy (Tchobanoglous et al., 2003). Since no data for the BG bi omass handling at mill were available, surrogate data for bagasse handling by Quintero et al. (2008), 1.50 MJ kg 1 were used. Following Maciejewska et al. (2006) 90 MJ t 1 energy consumption was assumed for BG milling
70 Allocation of Energy and Material U s e Gypsum is the only major byproduct produced at the mill. The allocation of energy for gypsum produced was done by assuming its displacement from other sources on an economic basis. Based on the market prices of fuel Ethanol ($0.38 kg 1 ) and Gypsum ($0.01 3 kg 1 ) in India the allocation and environmental burdens were distributed between Ethanol (97%) and Gypsum (3%). The allocation of energy produced by lignin was done by using the system expansion method. For calculation of embodied energy t otal life of the ethanol mill was taken as 15 years and the capacity of the mill was assumed to be 25 million gallons per year (25MGPY). In absence of actual data for construction material for the ethanol mill surrogate data from the Hill et al. (2006) study were used. The construction steel required for the production of ethanol from one tonne BG biomass was normalized with the proportion of total steel in the lifetime of the mill. Allocation for stainless steel and concrete were made on similar lines. Use of lignin as a source of energy displaces emissions from coal at the mill causing net reduction in greenhouse gas (GHG) emissions. Reduced wildfire emissions on account of removal of BG from forests and for subsequent use as E10 for displacing gasoline as motor fuel were also considered as avoided emission for calculating net impact. BG is a self propagating resource in the Red Sanders forests and hence no inputs were considered in the production process. Since there are no actual m ills using BG biomass as feedstock, surrogate data for chemicals and material required for production of ethanol from sugarcane from Kadam (2000) w ere used. While doing so, appropriate modifications for total ethanol and lignin produced were made using the composition of BG.
71 Net E nergy In LCA, the energy efficiency of bioenergy process or biofuels is expressed in terms of Net Energy Value (NEV or Energy balance) and the Net Energy Ratio (NER). Farrell et al. (2006) showed that net energy in LCA is a poorly defined parameter and that there were no consistent definitions. In many cases it has been calculated as energy content of ethanol minus fossil energy used to produce ethanol while in other cases it has been reported as the energy in ethanol and co pro ducts less the energy in the inputs In this study, net energy value (NEV) of ethanol is calculated as the difference between the energy content of ethanol produced from biomass and total energy inputs required in the entire fuel production cycle. As disc ussed in earlier section s the labor energy required for harvest of feedstock has also been added towards input energy. A negative NEV suggests that more energy is required to produce the bioenergy than the amount of energy that can be used for fuel (a net energy loss), whereas a positive NEV is an estimate of the energy gained for fuel use in the production process (a net energy gain). NEV= E e Ei Where E e is the energy content (Lower heating value) of ethanol and Ei is the total energy required to produce it. The net energy ratio (NER) is defined as the ratio of energy content in the ethanol to the total fossil energy required to produce it. NER = Ee / Enr Where, Enr is the energy content of fossil fuel used. An NER < 1 suggests a net energy loss, whereas an NER >1 suggests a net energy gain A high positive NEV and NER provide motivation to opt for the new fuels.
72 In order to c alculate the total energy inputs the quantity of direct energy inputs was multiplied by th e energy content (MJ L 1 ) of the fuel source and summed with the embodied energy inputs to give the total energy input per stage. The embodied energy of each component used was calculated by following values given in Hill et al. (2006) In order to allocat e the equipment use per functional unit produced, the total embodied energy of the equipment was multiplied by the quotient of the hours of use per functional unit and the lifetime (hours) of the machine. For energy consumed or displaced in the form of ele ctricity, the composition of the AP electric grid was considered. The calorific value of the ethanol produced was taken as 21.13 MJ L 1 Emissions The total greenhouse gas emissions associated with the utilization of one tonne BG input was calculated in Si maPro 7.1 using TRACI (Tools for the Reduction and Assessment of Chemical and other environmental Impacts) impact factors (Bare, 2003). Emissions associated with transportation of feedstock, ethanol production, electricity usage at the mill, and ethanol tr ansportation were accounted for. Since there were no inputs associated with feedstock production and harvest, no emissions were envisaged. No net CO 2 produced by combustion of lignin and wastes for production of steam was considered towards total emissions carbon that is part of the natural carbon cycle which does not contribute to atmospheric concentrations of carbon dioxide. For comparison, ethanol production and end use of ethanol as E10 emissions (Scenario B) were compared with the status quo situation i.e. BG burning and gasoline use on energy equivalent basis (Scenario A). Removal of BG from forests result s in avoided emissions (due to forest burning) which were calculated using open forest burn
73 data for dry tropical forests provided by Venkataraman et al. (2006). For scenario comparisons, the CO 2 and other emissions caused due to use of lignin in boilers for production of steam and electricity were added but the avoided emissions assuming displacement of coal for the same activity were reduced. Further, CO 2 released from renewable sources was differentiated from CO 2 released from fossil fuels. The production and combustion of E10 fuel, was assumed to occur in AP. Since, t here are no commercial biomass deri ved ethanol plants in India, and emission data fo r gasoline and E10 from India were not readily available, tailpipe emission data for gasoline and E10 reported by Kadam (2000) were used for calculating emiss ions from both the scenarios. Lifecycle Impact A ssessments Life Cycle Impact Assessment (LCIA) was performed to connect the mass and energy input and emission output results from the LCI to broader indicators or categories of impact to the environment or human health. The SimaPro 7.1 software supports t he performance of LCA and includes numerous methods to carry out impact assessments. In order to translate the emissions to the impact categories considered, the LCIA TRACI 2 v 3.00 developed by the U.S. Environmental Protection Agency was used (Bare, 200 3). For this LCIA, categories were selected based on their relevance to 2 equivalent), acidification (moles H+ equivalent), smog formation (kg NOx equivalent), human he alth imp acts (carcinogenic benzene equivalent) and water use. A brief description of these impacts is given in Appendix A
74 Interpretation LCI and LCIA results were interpreted based on the stated goal and scope of the study to compare the net energy balance, an d environmental burdens associated with the use of BG for ethanol production with other studies in India and elsewhere. Additionally, net GHG changes between Scenario A and Scenario B were interpreted using the environmental and human health indicators men tioned above. Economics of Ethanol P roduction Cost of E thanol In order to assess the economic viability of ethanol produced from BG feedstock, the cost of production per unit of ethanol was calculated. Cost data pertaining to mill construction, salaries, delivered biomass feedstock, fuel, water, chemicals, and disposal of ash were included in the analysis. Since there is no actual mill running, surrogate data for mill construction, operating costs and wages used in Aden et al. (2002) were used. The quantit y and cost adopted for each input per liter of ethanol produced are given in Table 3 2. All costs and benefits were scaled to 25 MGPY capacities by using the following scaling formula proposed by Aden et al. (2002), New cost = Original Cost*(New Size / Or iginal Size) exp where, the value of scaling exponent ( exp = 0.6) used by NREL (McAloon et al., 2000) was adopted. The annual costs for feedstock, fuel, water, chemicals, and ash disposal were calculated based on prevailing market rates in India and the amounts of each input necessary per year to meet the plant capacity of 25 MGPY. All the inputs were used to calculate the net present value (NPV) of the project over the 15 year life of the plant. The NPV was calculated with the following formula:
75 Where, B and C are benefits and costs incurred in year t and r is the discount rate. Following Short et al. (1995) a real discount rate of 10% was used in the study. Although the discount rate varies according to the risk taking and investment motive of the investor, Short et al. (1995) argued that in the absence of statistical data on discount rates, 10% discount rate should be taken for projects with risks similar to renewable energy investments. The breakeven cost o f production per unit of ethanol was c omputed by using the Solver add in tool in Excel Socioeconomic B enefits Collection of BG from forests entails significant benefits for the rural communities living in the forest fringe villages. Due to complexity involved in calculating the ecological and socioeconomic benefits, only economic benefits were calculated. Fo r this purpose biomass harvest to the mill capacity and the prevail ing labor rate in AP were us ed. Besides, the removal of BG from forests improves survival of precious Red Sanders regeneration by reducing risk of wildfires. Red Sanders is an endemic and endangered tree species in need of protection. The quantification of social, environmental services and ecological benefits accrued is a subject of future research. Sensitivity, U ncertainty and R isk A nalysis Sensitivity, uncertainty and risk analyses were performed on the assumptions and determine the effect of change in specifically defined values of a variable on the final outcome. Uncertainty and risk analysis it eratively calculates the value of the desired variable based on a distribution for the input variables defined by the analyst. A
76 triangular distributi on of costs was assumed for each of these components and the minimum, maximum and most likely, values of parameters were defined for each input variable ( Campbell and Brown, 2003 ). Using a Monte Carlo simulation with 10000 iterations on the input parameter s analyses were performed to determine the sensitivity with any given outcome, respectively. This was achieved by using the @Risk (version 5.7, Palisade Corporation, NY ) add in tool for Excel The sensitivity values, returned by @Risk depict normalized variations of the regression coefficients. It uses two methods: multivariate stepwise regression and rank order correlation for calculating sensitivity analysis. For t he sensitivity analysis of emissions, the low, medium and maximum values of ethanol production were taken as 358.80 L, 409.44 L, and 480.70 L. The lower and most likely values were based on the lower and higher conversion efficiencies suggested by Coughlin and Fridley, (2008), whereas theoretical production value (USDOE, 2010) was taken as the highest value. For the economic study the most likely values were the deterministic base case values arrived at in the study and the minimum and maximum values were s et at 10% below and 25% 35% above the base case values, respectively (Table 3 3). With these inputs and their given distributions, a Monte Carlo simulation with 10,000 iterations was run to give the mean emissions, cost and probability distribution for t he unit cost of delivered feedstock and for production of ethanol. Results and D iscussions The results for the energy balance, environmental emissions and impacts associated with ethanol production from BG are presented. Subsequently, the net
77 change in emi ssions and environmental impacts between the two scenarios are presented. The net differences between the two scenarios are presented in the form of a change from Scenario A to Scenario B. Energy B alance Ethanol production used 26.21 MJ L 1 of energy which was close to 27 MJ L 1 shown by other studies (Pimentel and Patzek, 2005; Sheehan et al., 2004; Farrell et al., 2006; Schmer et al., 2007). There was surplus 0 16 MJ electricity produced per liter of ethanol produced. The s team generation stage consumed 81% of total energy required in ethanol production. The net energy involved in the system is shown in Figure 3 3 and Table 3 4 Considering the energy content of anhydrous fuel ethanol as 21.13 MJ L 1 (lower heating value) the NEV was 16.86 MJ L 1 The high NEV indicates that very little energy was being utilized in the ethanol production system. This is not a su rprising outcome since there is no energy involved in raising the biomass crop. Further, the use of lignin displaced the fossil fuel s for conversion process thereby reducing net fossil fuel use. After taking into account all forms of nonrenewable energies (including embodied and nonrenewable component of labor energy) the NER was 5.15 indicating that the production of ethanol from BG i s an energy efficient process. A number of life cycle assessment (LCA) studies have studied NEV and NER of ethanol production from various lignocellulosic feedstocks. However, the geographical and temporal scope, choice of system boundaries, feedstock, all ocation assumptions, and methods of production make it difficult to conduct a fair comparison of NEV and NER. Reviews by Blottnitz and Curran, (2007) and Farrell et al (2006), also confirms this contention. The energy balance of the production system alon g with renewable and non renewable sources is shown in Table 3 4.
78 Emissions The total greenhouse gas emissions (CO 2 equivalent) for diverting one tonne of BG w ere 1.96 t which translates to 4.78 kg L 1 ethanol produc ed. Use of l ignin for generating steam at the mill displaced coal and greenhouse gas emissions equivalent to 2.07 kg L 1 ethanol produced. With the avoided emissions the net emissions in etha nol production were reduced to 2 71 kg L 1 The production phase contributed almost all the GHG emissio ns. Most of these emissions were in the form of CO 2 (99%), the remaining being methane. Since most fossil based CO 2 was replaced by lignin the CO 2 produced can be traced to the fermentation process and hence biogenic in origin. The details of emissions are provided in Table 3 5. Impacts and I nterpretation s Ethanol P roduction The emissions identifi ed resulted in global warming (7.34 kg CO 2 Equiva lent), human health cancer (5.54 kg Benzene equiv alent) and acidification (0.16 kg H+ equivalents) impacts. There were negligible smog impacts identified. The conversion sys tem was water intensive with 10 L water utilized per liter of ethanol produced. The overall impacts were very low as compared to other production processes since the BG feedstock grows naturally in forests, and no fertilizers are used as in the production of bioenergy crops. The details of emissions and impacts are shown in Figure 3 4. T he use of lignin in production of ethanol resulted in avoided emissions which further lowered the emissions. LCI A of Gasol in e Use and Grass B urning v s E thanol P roduction and E 10 Use The net change of emissions between scenario A and B are shown in Table 3 6. A positive value indicates the percentage by which the values for ethanol with E10 (Scenario B) were lower than the gasoline plus BG burning (Scenario A). CO 2 emissions
79 arising due to burning of biomass have been accounted for separately. The net impa cts in scenario A and B due to change in emissions are shown in Figure 3 5. Net e missions scenario A vs scenario B Scenario B favored reduction in emissions of SO x by 67% and CO by 41% which are quite significant. Fossil origin CO 2 de creased by 29 % in Scenario B whereas the biomass based CO 2 increased by 34 % due to the replacement of coal by lignin in the mill. Simila rly lead and benzene emissions we re reduced by 6% and 7% respectively in Scenario B, which can be attributed to displacement of gasoline b y ethanol in E10. On the other hand N 2 O and NO x emissions were lower for Scenario A by 93 % and 3 % respectively. The huge increase in N 2 O emissions in Scenario B were caused by the lignin based emissions in the boiler. The total quantity of CH 4 was higher i n Scenario B by 422 %. In this study CH 4 is a byproduct of ethanol production and waste treatment system and its use for heat generation has not been envisaged due to overall small quantity Net i mpacts scenario A vs scenario B Global war ming potential w as reduced by 9 % in Scenario B due to net reduction of emissions. Even though the proportion of CH 4 and N 2 O increased in Scenario B, their impact was more than offset by reduction in fossil based CO 2 The acidification potentials in both the scenarios were almost similar. The decrease caused by SO 2 in Scenario B was balanced by the increase in NO x Further the air acidification indicator values are offset by emissions arising from lignin burning. There was an 8% reduction in smog causing potential in Scenario B. This was primarily due to reduction of volatile organic carbon compounds released due to forest fires. Smog due to wildfire is known to cause respiratory ailments and hence its decrease is reflected in decrease of human
80 health cancer impacts as well. There was 29% reduction of human health cancer causing chemicals in scenario B. This can be primarily attributed to the reduction of CO 2 caused due to the substitution of gasoline by E10 and tha t of coal by lignin. This is an expected outcome since Scenario A has higher emissions due to use of higher quantity of gasoline and also due to open burning of forest grasses. The water use in scenario B has a potential of causing water stress due to its high water demand. Other than off site water demand during gasoline refining, and evaporation losses during forest fires there were no local water demands. Since the methodologies for water impact are not yet fully developed, the water use is mentioned for caution. Economics of Ethanol Production Cost of E thanol The unit cost of ethanol was estimated at $1.60 per gallon ($0.42 L 1 ) using the mean delivered feedstock cost of $25.34 per dry tonne. Based on the lower energy content of ethanol relative to gasol ine, the cost of an energy equivalent liter (EEL) and gallon (EEG) of ethanol were calculated to be $0.63 and $2.37 respectively. The cost of project investment at 47% of total costs was the largest single factor in the unit cost of ethanol production. The cost of the biomass feedstock, fixed operating costs, and ammonia contributed 21%, 13% and 7% respectively to the unit cost of ethanol production (Figure 3 6). Electricity costs were completely offset due to the combustion of lignin, a byproduct of the ac id hydrolysis. Socioeconomic I mpact The success of BG ethanol is linked with ecological benefits to Red Sanders and with community welfare. The annual production of 25 MGPY of ethan ol would require harvest of 0. 33 million tonne of BG which accounts for ha rvest of only 0.04 % of total
81 grass available. Assuming full capacity production, the local communities would g ain annual revenue worth $7.30 m illion by way of harvest labor. There would be additional employment creation for transportation and for mill oper ations as well. Along with production of ethanol and gypsum the mill will also produce surplus energy that can be sold to the grid. The BG supply can also be augmented with other locally available agricultural residues like rice straw and husk, wheat straw corn stover and sugarcane bagasse produced in the district. The environmental and ecological benefits of ethanol from BG include control of wildfires and improvement in the ecology of the endang ered Red Sanders forests. Sensitivity, Uncertainty and R isk A nalysis The sensitivity analysis performed to c heck the effects of increased production of ethanol on emissions indicated a corresponding increase in emissions and impacts. The i mproved production efficiency indicated a mean increase in emissions at 16%, with the lower bound of 90% distributio n situated at 6% and the higher bound at 27 % for all the impact categories studied. The r esults of sensitivity simulations run over the lifetime of the mill show that discount rate and bio mass feedst ock delivered cost are the most influential variable s towards the present value cost followed by the other major cost components of fixed operating costs, ammonia, diesel and sulphuric acid. The regression values of the variables influencing the present va lue of costs are shown in Figure 3 7. The magnitude of regression coefficients is expressed as standardized beta coefficients which imply a change in the output by one standard deviation from its mean in response to a corresponding change of one standard d eviation in an input variable (Gujrati, 2004) For instance, a change in biomass cost by one standard deviation will cause a change of 0.52 standard deviations in the total present value costs. The impact
82 of discount rate is large st amongst the variables, refl ecting that a low discount rate (interest on capital) would be essential for viability of an ethanol mill using BG as feedstock. Sensitivity simulation for the discount rate for the lifetime of the mill indicated a mean value of 10% with lower and upp er bound of 90% confidence interval at 8.35% and 11.63%. This indicates that if the discount rate goes beyond 11.63% the cost of ethanol could also increase. The probability distribution of discount rate is shown in Figure 3 8. The simulation performed det ermined the mean delivery price of feedstock to be $26.67 t 1 The bounds of the central 90% were correspondingly determined to be $23.98 t 1 and $30.16 t 1 (Figure 3 9). The unit cost of ethanol production modeled under each of the three values showed the mean at $0.46 L 1 and the lower and upper bounds of the 90% probability distribution at $0.40 L 1 and $0.53 L 1 centered on the mean (Figure 3 10). The cost of production compares well with the cost of ethanol for blending at $0.48 L 1 in India (IEP, 2008). The 2010 cost of ethanol for blending have been further revised to $0.60 L 1 which makes lignocellulosic ethanol production more competitive. Simulation results of cost of biomass versus the cost of ethanol production in dicated a li near relationship: Y = 0.431x + 0.431 (R 2 = 1 ) shown in Figure 3 1 1, where Y is the cost of production and x is the percent change in the cost of biomass. These results indicate that the production cos t of ethanol increases by $ 0.04 on average for every 10 % increase in cost of biomass within the probability distribution of biomass with a lower bound of $22 and a higher bound of $31 per dry tonne. The sale of surplus electricity to the grid has not been included towards economic analysis as this surplus de pends largely on the facility design. The cost of production will be further reduce d if the sale of electricity is also taken into account. The range of
83 values for the unit costs reflects the uncertainty of the final delivered feedstock cost and also the c ost of production. The results of Monte Carlo simulations run to check uncertainty and risk associated with the net present value (NPV) analysis indicat ed positive NPV throughout the simulations. This indicates that production of ethanol from BG is a viabl e project within the uncertainty limits indicated in this study. This study is limited by the use of data from sources other than Indian scenario. The identification and development of a pilot study will help in improving results of the study. Summary, C onclusions and Policy Implications Life cycle assessment o f ethanol production from BG indicated 26.21 MJ L 1 energy consumption The net energy value of 16.86 MJ L 1 and a net energy ratio of 5.15 indicated lower energy inputs relative to energy output. The production system also generated surplus electricity of 0.16 MJ L 1 that could be sold to the grid. The conversion of biomass to ethanol and use as motor fuel (E10) had lower environmental burdens and impacts as compared to the status quo situation o f use of gasoline as motor fuel and open burning of BG There was a net reduction of 0.74 kg of GHG emissions (CO 2 equivalent) per liter ethanol produced and used as E10 for motor fuel as compared t o gasoline use and allowing the biomass to burn. The reduc ed emissions resulted in alleviation of human health (cancer), global w arming, and smog potentials by 29%, 9 % and 8% respectively compared to status quo situation The cost of ethanol production based on 2008 rates $0.42 L 1 was competitive with the 2008 fuel ethanol price of $0.48 L 1 in India. The cost of production was sensitive to the biomass cost which accounted for 21% of the total production costs. With a feedstock cost range of $23.79 t 1 to $30.01 t 1 the corresponding costs of production was $0. 40 L 1 to $0.53 L 1 The scope of this study was limited to environmental impacts and economic evaluation.
84 The evaluation of social and economic impacts by using Life Cycle Costing (LCC) and Social Life Cycle Assessment (SLCA) would improve this study. The results of this study indicate energetic, ecological and economic benefits of utilizing Bodagrass for ethanol production in so uthern A P Since Bodagrass is a self propagating resource, biomass production does not involve any land use change and inputs. Besides contribut ing to energy security it will also save valuable foreign exchange and improve self reliance in motor fuels. The disposal of this grass from forests will also help in reducing wildfires, thereby improving restoration of Red Sanders forest s. Its free availability contributes to poverty alleviation on one hand and ecological be nefits on the other. Since local communities are already dependent on these forests, small scale community based projects have greater chances of success. Further, th ese endeavors can be built around area specific biomass availability. Since India cannot afford the diversion of arable lands for biofuels, identification of low input biofuels systems such as this will have far reaching implications. Additionally, researc h and development for multiple lignocellulosic feedstock based bioethanol production and suitable policy initiatives for involvement of private entrepreneurs and communities will help in achieving self sufficiency in the long run. All ethanol produced in I ndia is currently from molasses. With an increase in blending targets, sourcing ethanol from other lignocellulosic resources will be essential. Since the technological advancement in converting lignocellulosic resources to ethanol is a challenging task, In dia will have to implement integrated research and development to overcome this challenge. Since large scale initiatives require land use change, small scale, regional or local initiatives for identification and development of feedstocks would
85 be a better option. Further, there is uncertainty in availability of a particular biomass for the entire year, bioethanol technology with multiple feedstock use would be better for the Indian scenario. Involvement of local communities and private entrepreneurs in this field would help in local initiatives.
86 Table 3 1 Sources of data used in the study Item Data required / Process Data sources Biomass Green weight per tonne Data collected in Chittoor district of Andhra Pradesh State, India Dry weight per tonne Labor required per tonne Energy Harvest labor (LSSE*) Odum, 1983 Energy intensity India EIA, 2008 Transport Feedstock Khatiwada and Silveira, 2009 Biomass Milling Maciejewska et al., 2006 Biomass handling Quintero et al., 2008 Steam generation Enthalpy of steam; Kadam,2000 Wastewater treatment Tchobanoglous et al., 2003 Embodied Energy Hill et al., 2006; SimaPro Transport ethanol Khatiwada and Silveira, 2010 Emissions Lignin combustion SimaPro Fuel combustions SimaPro Vehicular emissions (gasoline) Kadam, 2000 Vehicular emissions (E10) Kadam, 2000 Biomass burning in forests Venkatraman et al., 2006 Miscellaneous Hydrolysis efficiency Coughlin and Fridley, 2008 Fermentation efficiency Coughlin and Fridley, 2008 Electricity Mix AP Power generation corporation Material required (Mill) Hill et al., 2006 Material required (Chemicals) Kadam, 2000 Labor wage rates NREGA, 2009 Life style support energy
87 Table 3 2 Inputs costs and data sources Items Unit Quantities L 1 EtOH Cost ($/unit) Data Sources / Reference Biomass Dry kg 3.53 0.025 Data Collection Hydrated lime kg 0.05 0.08 http://www.icis.com/StaticPages/k o.htm#L NH 3 kg 0.09 0.37 http://www.icis.com/staticpages/a e.htm Diesel kg 0.01 0.89 http://petroleum.nic.in/petstat.pdf H 2 SO 4 kg 0.17 0.08 http://www.icis.com/StaticPages/p s.htm#S Ash Disposal kg 0.33 0.02 Aden et al. 2002 Water kg 12.68 0.0004 Aden et al., 2002 Gypsum produced kg 1.31 0.02 http://www.alibaba.com/showroom/raw gypsum.html Discount rate % 10.00 Short et al. 1995 Initial project investments L 0. 10 Aden et al. 2002 Fixed Operating Costs L 0. 05 Aden et al. 2002 Salaries L 0.0 1 Modified from Aden et al. 2002
88 Table 3 3 Risk and sensitivity analysis values Risk element Unit Minimum Most Likely Maximum Discount rate % 0.08 0.10 0.13 Ethanol produced M L 85.17 94.64 118.29 Ethanol Cost $ L 1 0.38 0.42 0.57 Biomass Cost $ L 1 0.023 0.025 0.032 Ethanol sale M $ 29.81 39.75 49.68 Gypsum sale M $ 0.37 0.41 0.45 Benefits (Ethanol + Gypsum) sale M $ 30.18 40.16 50.14 Dry Biomass M $ 7.62 8.47 9.32 Hydrated lime M $ 0.30 0.33 0.37 Water M $ 0.43 0.48 0.52 NH 3 M $ 2.75 3.05 3.36 Diesel M $ 1.07 1.19 1.49 H 2 SO 4 M $ 1.13 1.26 1.38 Ash Disposal M $ 0.58 0.64 0.70 Initial Project Investments M $ 125.59 139.54 153.49 Salaries M $ 0.80 0.89 0.98 Fixed Operating Costs M $ 3.36 4.48 5.60 Costs M $ 143.62 160.33 177.21
89 Table 3 4 Energy balance of ethanol production system Activity Total Energy used (MJ L 1 ) EtOH Fossil energy Inputs (MJ L 1 ) EtOH Renewable energy inputs (MJ L 1 ) EtOH Harvest (Human labor) 0.67 0.34 0.33 Diesel (Feedstock transport) 0.55 0.55 Process energy 4.44 4.44 Steam generation 20.06 20.06 Waste water treatment 0.01 0.01 Diesel at mill 0.39 0.39 Diesel (EtOH transport) 0.10 0.10 Steam produced (Lignin) 20.06 20.06 Electricity produced 4 .6 0 4 .6 0 Embodied energy of material Diesel 0.95 0.95 Lime 0.05 0.05 Water 0.06 0.06 NH 3 1.75 1.75 H 2 SO 4 0.02 0.02 Steel 0.01 0.01 Concrete 0.00 0.00 Sum 4.40 4.23 0 18 Allocation to EtOH (97%) 0.97 Energy Content EtOH 21.13 NEV 16.86 NER 5.15
90 Table 3 5 Net emissions in ethanol production Emission type Emissions in EtOH production (kg t 1 ) biomass Avoided emissions Heat production (kg t 1 ) Avoided emissions Electricity Production (kg t 1 ) Total avoided emissions (kg t 1 ) biomass Net Emissions (kg t 1 ) biomass Net Emissions (kg L 1 ) Ethanol VOC 0.08 0.00 0.00 0.00 0.08 0.0002 CO 0.90 0.65 0.09 0.74 0.16 0.0005 NO x 1.35 1.42 0.20 1.62 0.27 0.0008 SO 2 0.04 8.04 1.15 9.19 9.14 0.0255 CH 4 23.66 0.00 0.00 0.00 23.66 0.0659 N 2 O 0.11 0.00 0.00 0.00 0.11 0.0003 CO 2 (Biogenic) 1929.60 0.00 0.00 0.00 1929.60 5.3779 CO 2 (Fossil) 0.00 645.06 92.39 737.45 737.45 2.0553 PM 0.23 8.51 1.22 9.73 9.50 0.0265 Hg 0.00 4.96E 05 7.11E 06 5.67E 05 0.00 0.0000 Total 1955.98 663.67 95.05 758.72 1197.26 3.3368
91 Table 3 6 Emissions Scenario A vs Scenario B ** ( g kg 1 Biomass) Emission Type Scenario A Scenario B Difference Percent Difference Favored Scenario Emission Tropical Forest Fire a Gasoline Emissions b Burning + Gasoline Emissions EEL_ E10 Emissions c EtOH Production Emissions d EEL_E10 + EtOH Production Emissions VOC 7.40 38.00 45.40 39.00 0.08 39.08 6.32 13.93 B CO 83.00 198.00 281.00 164.31 0.93 165.24 115.76 41.19 B NO x 3.40 31.00 34.40 33.90 1.39 35.30 0.90 2.60 A SO x 0.60 0.26 0.86 0.24 0.04 0.28 0.58 67.16 B CH 4 3.30 1.50 4.80 1.39 23.66 25.06 20.26 422.00 A N 2 O 0.00 0.13 0.13 0.14 0.12 0.25 0.12 93.41 A CO 2 (Biogenic) 1591.00 0.00 1591.00 208.20 1929.60 2137.80 546.80 34.37 A CO 2 (Fossil) 0.00 2775.00 2775.00 2687.21 720.47 1966.75 808.25 29.13 B Lead 0.00 0.20 0.20 0.19 0.00 0.19 0.01 6.41 B Benzene 0.00 1.53 1.53 1.42 0.00 1.42 0.11 6.88 B Benzo(a)pyrene 0.00 0.00005 0.00005 0.00 0.00 0.00 0.00 6.41 B Ethanol 0.00 0.00 0.01 0.40 0.00 0.40 0.39 NM A NH 3 1.20 0.00 1.20 0.00 0.00 0.00 1.20 100.00 B *A= Gasoline use + Open burning of Bodagrass, ** B= EtOH production + E10 use ; a Venkatraman et al., 2006; b, c Kadam, 2000; d This study, EEL_E10 = Energy Equivalent Liter of E10; NM = Not meaningful
92 Figure 3 1. System boundary and reference system
93 Figure 3 2 Flow chart two stage dilute sulphuric acid process
94 Figure 3 3 Process e nergy involved per liter in E thanol production system Figure 3 4 Ethanol production system impacts (g L 1 EtOH)
95 Figure 3 5 Net impact changes Scenario A (Gasoline use + Open burning of Bodagrass) v s. Scenario B (EtOH production + E10 use)
96 Figure 3 6 Cost components of ethanol production Figure 3 7 Regression coefficients of total present value of costs.
97 Figure 3 8 Probability distribution of discount rate
98 Figure 3 9 Probability distribution of biomass cost
99 Figure 3 10 Probability distribution of Ethanol cost
100 Figure 3 11 Sensitivity analysis ethanol production cost v s. Biomass cost
101 CHAPTER 4 SANDERS WOOD MARKETING IN ANDHRA PRADESH, INDIA Red Sanders F orests Red Sanders (RS, Pterocarpus santalinus L .) is an endangered and valuable tropical tree species endemic to southern India. It is distributed in a geo botanically restricted area of 2,000 km 2 between 13 30' 15 0' N and 78 45' 79 39'E (Raju and Raju, 1999). The tree prefers 80 cm to 120 cm rainfall, perfect drainage, and elevations ranging from 200 m to 900 m (Ahmed and Nayar, 1984). Within its natural distribution, it grows in about 500 km 2 of fragmented forest landscape. RS wood is renowned for its characteristic color, wavy grain, and many other chemical properties (Reddy, 1972). The reasons for the development of wavy grain in the RS heartwood are not well known. However, it is this wavy grained dark color hea rtwood that fetches premium price in the international timber market. For instance, the RS wood price ranged from US$ 6,870 9,160/metric tonne in 2002 in global timber markets (Mulliken and Crofton, 2008). Note that finished RS wood products such as carved statues and furniture fetch an even greater price. RS wood was traditionally used in India for medicinal purposes, agriculture equipment, building timber, and carving religious souvenir figurines and statues (Ramakrishna, 1962). Outside India, the earlies t historical record of RS wood appeared in China as documented by Cui Bao of Jin Dynasty (265 340 A.D.) in "Comments on Ancient and Modern Items". In this text, RS wood is regarded as rare and precious like gold. The earliest recorded history of RS wood t rade can be traced back to 1681 when the British East India Company transported it to England for dyeing purpose (Reddy, 1972). During the latter half of the eighteenth century and most of the nineteenth
102 century, RS wood was used in railway sleepers (Reddy 1972). From the early twentieth century (1915 1920), RS wood trade for making musical instruments in Japan started for which only highest quality wavy grained RS wood is used (Kiyono, 2005). In China, RS wood is mostly used in making musical instruments and for producing Ming and Qing dynasty imitation furniture. In addition to its timber value, extracts obtained from RS wood have medicinal values. These extracts are used for healing wounds and boil burns reducing inflammation, treating headaches and fev ers, curing skin diseases, and improving sight (Chopra et al. 1956; Kirtikar and Basu 1983; Biswas and Mukherjee, 2003; Krishnaveni and Rao 2000; Rao et al., 2001). RS wood extracts have been reported to possess anti hyperglycemic (Rao et al., 2001), antiu lcer (Narayan et al., 2007), and antibacterial properties (Manjunatha, 2006). Furthermore, the dye obtained from RS wood, S antaline is used in coloring pharmaceutical preparations and foodstuffs (Manjunatha, 2006). Several factors that pose challenge s t o RS conservation include slow growth, excessive biotic pressures, poor socioeconomic condition of local communities, and illegal removals. Additionally, the reduced number of mature seeding trees (mostly due to illegal logging), poor seed germination (30% 40%), and low survival rate of seedlings (due to recurring wildfires and grazing) are adversely impacting regeneration of RS in forests. For example, as per the total tree enumeration data collected by the local forest department in 2006, 85% of RS tree s in forests were below 75 cm and less than 1% were abo ve 100 cm girth at breast height (APFD, 2006). This clearly shows the severity of threat to this species and highlights the necessity of immediate steps to ensure in situ conservation of this species. Increased threat s to RS forests prompted
103 the G overnment of India (GoI) and G overnment of Andhra Pradesh (GoAP) to restrict commercial harvest of RS in 1956 (Ramakrishna, 1962) and to impose a total ban on harvest from public forests in 1980 (Working plan Rajampet, 2008). Furthermore, to address the overharvest and habitat alteration threats, RS was placed on the CITES endangered list in 1995 (IUCN, 2010). As a result, GoI and CITES clearances became mandatory for exporting RS wood item s from the country. A ll the steps taken by GoI and GoAP indicate that the present policy focus is restricted to conservation and protection of RS forests but not on production of RS on private lands. Although APFD has raised RS plantations on public forestland s (APFD, 2009), mostly limited to raising a few scattered trees on field bunds. Continuous decline in the supply and ever increasing demand for RS wood has resulted in an increase in illegal harvest from public forests. For example, th e local forest department alone seized 3067 t of RS wood during 2001 2007 (APFD, 2009) which was sold by the GoAP to buyers across the globe. It should be noted that this is the only legal source of RS wood in the market and is not sufficient to meet the a nnual international demand of approximately 3000 t (Mulliken and Crofton, 2008). It can therefore be safely inferred that the management of RS wood trade holds a key for its conservation. Conservation of endangered species often includes a total or restric ted trade ban. These two practices are difficult to implement in practice especially in the context of a developing country where communities are directly dependent on forests for livelihood (Swanson, 1993). In the case of RS wood, promotion of RS growing on private lands can be a prudent choice for conservation of this endangered species. This approach will provide economic returns to the society at large and help in reducing pressure on public
104 forests by curtailing illegal logging. CITES (2001) emphasize s that sustainable trade of some species can itself contribute to its survival by providing value and therefore, economic incentives for its continued existence. It is envisioned that such a trade can employ people in developing countries who might otherwis e turn to more destru ctive practices like smuggling or shifting agriculture and thus pose a bigger threat to the species. However, the endemic and endangered nature of the species on one hand and the huge demand for this wood on the other, make the development of sustainable RS wood market a challenging task. In order to address this issue, a deeper understanding of intricacies involved with the RS wood trade is needed. This improved understanding can help us formulat e appropriate policies that could further help in bringing a balance between ecological, social, and economic aspects of RS wood trade. However, policy development is a composite process which frequently takes place in an unstable but rapidly changing context and is subject to unpredictab le internal and external factors s are critical for policy formulation. A deeper understanding of the s s will ensure improved cooperation and conflict resolution (Simmons and L ovegrove, 2005). In th is study the SWOT AHP (Strengths, Weaknesses, Oppo rtunities and Threats Analytic Hierarchy Process ) technique was used to analyze perceptions of four key stakeholders (administrators, landowners, traders, and NGO & academia) toward s sustainable RS wood trade development in the southern Indian state of Andhra Pradesh (AP) T his study was conducted in the state of A P, where most of the remnant RS forestlands reside (Figure 4 1). This study aims to understand critical areas of interven tion, and challenges and opportunities for establishing sustainable RS wood
105 trade in the region. The next section defines the SWOT AHP framework in detail. The third section discusses methods. In the fourth section results are presented and discussed. Conc lusions are drawn in the last section. SWOT AHP Framework SWOT analysis is a tool used for identification of key internal and external factors that are important to achieve the objective of an agen t SWOT analysis groups key pieces of information into tw o main categories internal and external. Internal factors are described as strengths and weaknesses as these are internal to the organization whereas external factors are opportunities and threats presented by the external environment to the organization ( Kotler, 1994; Weihrich, 1982). However, SWOT analysis does not measure each factor quantitatively, and hence the relative weightage of factors influencing the decision making process cannot be assessed (Pesonen et al., 2001). Another technique, AHP can be used for assessing relative weights of different factors present in a SWOT analysis. AHP is a multi criteria mathematical model for facilitating complex decision making process es (Saaty, 1977). It involves pair wise comparisons and relies on the judgments of experts to derive priority scales that measure intangibles in relative terms. From these pair wise comparisons, the relative priority value of each factor within each SWOT group can be computed using the eigen value method as explained below. The method ology adopted by Shrestha et al. (2004) can be used to place the information derived from pair wise comparisons of SWOT factors as a reciprocal matrix of weights, where the assigned relative weight enters into the matrix as an element a ij and its reciprocal goes to the opposite side of the main diagonal.
106 A = = (Eq. 4 1) Where, rows indicate the ratio of weights of each factor with r espect of all others. In the matrix, when i = j a i j = 1 i.e., the main diagonal element of matrix A. Thus, a i j values degree of preference of factor i over factor j When us ing AHP in group decision making the geometric mean has been suggested as a method of calculating the overall comparisons across all respondents (Forman and Peniwati, 1998; Saaty, 1982; Saaty and Alexander, 1989). Aczel and Saaty (1983) proved that the ge ometric mean is consistent and upholds the salient featu res of the AHP. When matrix A is multiplied by the transpose of vector of weights, w it gives t he resulting vector nw Aw = nw, where, w = ( w 1 w 2 n ) T ( Eq. 4 2) Equation 4 2 can be rewritten as, (A nI) w = 0 ( Eq. 4 3) consistency. Inconsiste responses in pair wise comparisons. Therefore, the matrix should be tested for consistency by using the following formula CI = ( max n ) / ( n 1), ( Eq. 4 4) CR = CI / RI, ( Eq. 4 5)
107 Where, CI is the consistency index, RI is the random index generated for a random matrix of order n and CR is the consistency ratio. As a rule of thumb, a CR value of 10% or less is considered as acceptable (Kagnas, 1994; Ananda and Herath, 2003; Shr estha et al., 2004; Masozera et al., 2006). Kurtilla et al. (2000) developed the hybrid method by using AHP in SWOT which enables a researcher to derive quantitative measures of importance of each factor under consideration under a given decision making sc enario. Since then, this hybrid method has been used for study of perceptions in a variety of natural resource management areas e.g., strategic natural resource management and planning (Pesonen et al., 2001), agroforestry (Shrestha et al., 2004), tourism m anagement (Kajanus et al., 2004), community based protected area management (Masozera et al., 2006), and forest biomass based bioenergy development (Dwivedi and Alavalapati, 2009 ). However, to the best of knowledge, this framework has not been used for eva luation of stakeholder perceptions pertaining to sustainable trade development of an endangered species so far, and this study is an attempt in this direction. Methods In this study four key stakeholder groups were identified based on their relevancy towards sustainable RS wood trade These include administrator s landowner s trader s and NGO s & academia. B rainstorming sessions were conducted with experts from each stakeholder group and expert opinions were solicited for suitable factors of each SWOT c ategor y The responses received were analyzed and grouped into different SWOT categories (Table 4 1). Second, a questionnaire for ascertaining relative priorities of identified stakeholder groups was developed for each SWOT factor listed under each SWOT ca tegor y A brief explanation of each factor was also included in the
108 questionnaire to ensure common understanding among respondents. In the questionnaire, pair wise comparisons of a factor present in a particular SWOT category were arranged against all othe r factors in the same SWOT category. This was repeated for all the factors present in a SWOT category. The respondents were asked to evaluate both the factors present in a pair wise comparison and then to mark order of importance of one factor over another based on their own understanding. Four priority categories i.e., equal, moderate, strong, and very strong were created and while calculating actual priorities, these factors were equated to 1, 3, 5, and 7, respectively. An example of a comparison table in cluded in the designed questionnaire is shown in Table 4 2. G roup interviews were conducted for the landowners and junior level forest officers 8 to solicit their feedback on the questionnaire at Rajampet, Kadapa, and Tirupati during June July, 2009. Before conducting group interviews, the objective and methodology was explained to all the participants to maintain uniformity. For each stakeholder group, p articipants for the firs t questionnaire wer e selected from three forest divisions (Ra jampet, Kadapa, and Tirupati). These forest divisions contain the highest proportion of total RS growing stock in the state 9 (APFD, 2010). For the administrator s stakeholder group, participants were selected from all hierarchy levels of the APF D based on their prior work experience in RS forests. For the landowner stakeholder group, participants were selected randomly from community forest villages and from all agriculture land ownership classes (marginal: < 1 ha, small: 1 2 ha, medium: 2 4 ha, and large: > 4 ha). With the help of local divisional forest offices the RS 8 9 The Total growing stock of Red Sanders have been estimated at 118,000 m 3 out of which 19%, 18%, and 9.5% is available in Rajampet, Kadapa, and Tirupati forest div isions, respectively
109 wood traders were contacted and invited to participate as trader stakeholder group in the survey. For the NGO s & academia stakeholder group, professors and lecturers in the life science department of Sri Venkateswara University at Tirupati and Yogi Vemana University at Kadapa were invited. Local NGOs working with forest communities were included in the survey. P air wise comp arisons were done by different stakeholder groups at individual and group levels. Thereafter, the participants were requested to discuss various pair wise comparisons listed in the questionnaire and to arrive at consensus with respect to each question. Onc e consensus was reached, the resource restricted to facilitating discussion, translation, and recording the outcome of discussion for each pair wise comparison. Individual qu estionnaires were delivered electronically to all other participants and responses were collected electronically Overall, 14 group interview s (4, 8, and 2 for administrator s landowner s and trader s stakeholder groups, respectively) were conducted and 30 individual questionnaires (15 and 15 for administrator s and NGO s & academia stakeholder groups, respectively) were delivered For questionnaires, 11 responses (8 and 3 for administrator s and NGO s & academia stakeholder group, respectively) we re received W ithin stakeholder groups, for each factor present in a particular SWOT category, individual responses for pair wise comparisons were consolidated using geometric mean s (Saaty, 2001) and processed using the SWOT AHP framework. In this study the SWOT groups and factors were weighted equally. For each stakeholder group and for each SWOT category, factors with the highest priority score were identified and used to develop a second set of questionnaires to
110 estimate overall priority of each factor with respect t o each other for each stakeholder group. These questionnaires were delivered to 30 participants (8, 8, 6, and 8 for administrator s landowner s trader s and NGO s & academia stakeholder groups, respectively) identified during the first round of questionnaire s in August, 2009. In return, 10 responses (4, 2, 2, and 2 for administrator s landowner s trader s and NGO s & academia stakeholder groups, respectively) we re rec eived Once again, using the above method, the responses were analyzed and final priorities of factors under each SWOT category for each stakeho lder group were derived. MS Excel 2007 was used to check for inconsistent preferences at the aggregate level. T hrou ghout the analyse s, consistency ratios were maintained below 10% for all pair wise comparisons. Results and Discussion The overall results of the study are summarized in Table 4 3. The factor priority scores demonstrate the relative importance of each factor within SWOT groups. The overall priority scores (obtained by adjusting factor priority scores by multiplying with SWOT group priority score) illustrate the relative importance of each factor across all SWOT categories. Following Masozera et al. (200 6), the combined overall priority values of strengths and opportunities categories can be interpreted as a positive perception whereas for weaknesses and threats categories it reflects negative perception. Administrator s Stakeholder Group This stakeholder group gave highest priority to the factor high price of RS wood (0.325) followed by the factor geographical advantage (0.306) under the strengths category (Figure 4 2). These preferences can be attributed to their intrinsic knowledge about the uniq ue RS habitat and commercial value of RS wood There is common
111 knowledge among this stakeholder group that RS grown outside its habitat does not have a good commercial value as the wood does not de velop requisite color and grain when compared to the wood grown within the habitat. Therefore, this stakeholder group feels that all those landowners who have lands within RS habitat have geographical advantage an d that they would benefit by planting RS for sustainable RS wood trade. Un der the opportunities category, this stakeholder group gave highest score to the factor improvement in economic standard of landowners and handicraft artisans (0.302) followed by the factor forest conservation and growth (0.291). These preferences were in line with the general mandate of the APFD which emphasizes alleviation of rural poverty while conserving forests and wildlife. Under weaknesses the category, the factor long gestation crop (+40 years) (0.368) received highest priority followed by th e factor small land holdings (0.311). RS is a long gestation crop and normal rotation in the wild is about 60 years (Working Plan Rajampet, 2008). With the help of modern tissue culture techniques and by selecting best genotypes the rotation can realisti cally be lowered to 40 years. However, even 40 years is a long period for the landowners especially when the majority of landowners have small land holdings in the region. Under the threats category, the factor increased risk of illegal removals due to r educed Government control (0.311) received highest priority followed by the factor change in market demand (0.246). During the past decade, there has been a steep increase in the demand for RS wood in China. Due to the demand supply gap and restricted tr ade policy under CITES, there has been a spurt in illegal removals which is reflected in seizure of illegal wood by all major protection agencies from all major ports in India (NMDT, 2007). This stakeholder group perceived that due to less Government
112 contr ol under the sustainable RS wood trade, there are high chances that illegal logging of RS wood might escalate. This clearly implies that this stakeholder group favors a proper monitoring and verification component in a new policy to ensure RS conservation and equity among trade beneficiaries. The second highest priority under the threats category i.e., change in market demand also has an impact on the RS protection as short supply may increase spurt in smuggling. Since the international RS wood demand i s met through legal and illegal channels, it is difficult to have a clear idea about the important obstacle in developing and planning sustainable RS wood trade. This conc ern is quite logical since expansion of RS wood trade is difficult without accurate demand supply information. Analysis of overall scores of SWOT categories across the groups revealed overwhelming priorities for the weaknesses (0.380) and threats (0.41 6) categories so much so that there were no strengths and opportunities factors in the top five factors based on their overall priority values across all SWOT categories. Since all the overall priority factors adds up to one, the combine d priority (0.7 96) of these factors can be interpreted as a negative perception towards feasibility of sustainable RS wood trade. This negative perception (80%) towards RS wood trade far outweighs a positive perception (strengths and opportunities). Landowner s Stakeholder Group The SWOT factor priorities under strengths category for this stakeholder group revealed high price of RS wood (0.353) as most important factor followed by the factor geographical advantage (0.251) (Figure 4 3). As in the case of the administrator stakeholder group, these priorities showed their awareness and knowledge about the
113 unique RS habitat and commercial value of RS wood. The factor improvement in economic standard of landowner and handicraft artisans (0.460) was given high est priority score within the opportunities category followed by the factor forest conservation and growth (0.259). These priorities were expected since success of sustainable RS wood trade is likely to benefit landowners and artisan members of the com munity. The factor small landholdings (0.360) received highest priority under the weaknesses category, followed by the factor long gestation crop (+40 years) (0.302). In rural Andhra Pradesh, subsistence agriculture is a dominant system and it is dif ficult for them to set aside land for tree planting. Tree planting is practiced on field bunds but if the landholdings are small, the farmers do not plant trees to avoid shade and to protect agriculture yield. Therefore, highest priority for the factor sm all landholdings reflects the general practice adopted by the farmers. The landowner stakeholder group gave highest priority to the factor competition from other commercial crops (0.472) under threats category. In the state, landowners generall y plant fast growing commercial species to get maximum benefits from small land holdings and this practice is a definite threat to RS wood trade which has a long gestation. Analysis of priorities across the group revealed that landowner stakeholder group perceived threats (0.398) as the main constraint to sustainable RS wood trade even though this group also saw opportunities (0.293) associated with sustainable RS wood trade. There is not a single factor belonging to the strengths category in the top five factors based on their overall priority values across all SWOT categories. This clearly indicates lack of confidence of this stakeholder group towards the concept of sustainable RS wood trade. Like the administrator stakeholder group, the combine d
114 score (62%) of a negative perception (weaknesses: 22% and threats: 40%) outweighs a positive perceptions (38%) (strengths: 9% and opportunities: 29%) for this stakeholder group as well. This indicates that that this stakeholder group would be hesi tant to participate in developing a sustainable RS wood trade. The h igh priorities of factors present under the threats and opportunities categories reflect that this stakeholder group perceives external factors (not controlled by local agencies) as t he main constraints towards development of sustainable RS wood trade. Trader s Stakeholder Group Under the strength category, this stakeholder group gave highest priority to the factor niche international market available (0.420) closely followed by the factor high price of RS wood (0.405) (Figure 4 perception can be attributed to the fact that niche market and high price are the two most important factors for practicing in RS wood trade. The factor fore st conservation and growth (0.492) received highest priority under the opportunities category. This clearly implies that this s takeholder group is interested in maintain ing a sustainable resource base which can be used for meeting their business needs f rom time to time. The factor lack of incentives (0.458) received highest priority under the weaknesses category followed by the factor administrative complexities (0.247). This indicates that this group perceives that incentive programs run by Government would definitely help in bringing private sector involvement in developing sustainable RS wood trade. International trade in RS wood is governed under CITES rules and as such the export of RS wood is ridden with administrative complexities right from harvesting to conversion, transportation, and export permissions from different departments. It can sometimes be a frustrating process for traders and as a result, relative high priority for the factor
115 administrative complexities under weaknesses cat egory is not a surprising outcome. Additionally, there is no extension, market information and insurance mechanism for the RS plantations as well as wood. Under threats category, this stakeholder group gave highest priority to the factor lack of coordinat ion amongst external agencies (0.603). The reasons for this priority can be traced to the fact that most state forest departments, various federal environmental agencies, and international agencies function in seclusion and there is limited flow of inform ation amongst them. As a result, this stakeholder group faces lot of uncertainty. Therefore, this clearly implies that a new policy should not only strive for a better coordination among different agencies involved with RS wood trade and but also emphasize on reducing unnecessary bureaucratic hurdles. Analysis of overall scores across the categories showed that this stakeholder group perceived weaknesses (0.518) as main constraint in developing sustainable RS wood trade. Interestingly, this stakeholder gr oup also perceived strengths (0.250) as important. Amongst the top five overall priority factors, three factors were under the weaknesses category and two were under the strengths category. However, the opportunities (0.137) and threats (0.095) c ategories were not considered important for sustainable RS wood trade by this stakeholder group. The combine d higher priority (61%) of a negative perception (weaknesses: 52% and threats: 9%) can be interpreted as reluctance of this group towards adoption o f RS wood trade. As per this stakeholder group, internal issues (controlled by the local agencies) were bigger constraints towards development of sustainable RS wood trade. NGO s & Academia Stakeholder Group The highest priority under the strengths catego ry was given to the factor high price of RS wood (0.439) followed by the factor niche international market available
116 (0.322) by this stakeholder group (Figure 4 5). Under the weaknesses category, factors long gestation crop (+40 year) (0.406) followed by the factor lack of incentives (0.243) were given top priorities. These priorities suggest that besides suitable incentive programs, research targeted at reducing the long gestation of RS would help in developing sustaina ble RS wood trade. Under the opportunities category, the factors improvement in public private relations (0.317) and forest conservation and growth (0.317) were given highest priorities. The highest priority within the threats category was given to the factor lack of coordination among external agencies (0.323) followed by the factor competition from other commercial agricultural crops (0.265). Analysis of overall priority scores for this stakeholder group reveals that this group perceives weak nesses (0.450) as the main constraints in developing sustainable RS wood trade. Interestingly, this group gave a priority value of 0.375 to the strengths category (0.375) which is the highest across all stakeholder groups. The higher overall priority fo r strengths category indicates that this stakeholder group perceives a distant future for sustainable RS wood trade provided the challenges associated with it are addressed suitably. It was noticed that among top five overall priority factors, three were u nder the weaknesses category and other two under the strengths category. Categories of opportunities (0.096) and threats (0.080) were not perceived as important. In terms of overall priorities, this stakeholder group was completely in agreement wit h the trader stakeholder group and felt that internal factors existing within the system need to be addressed for ensuring sustainable RS wood trade. The combine d priority of a negative perception s 53% (weaknesses: 45% and threats: 8%) can be interprete d as their reluctance to adopt RS wood trade which outweighs the positive
117 perceptions 47% (strengths: 37% and opportunities: 10%). These priorities suggest that despite a more positive outlook than other groups this stakeholder group perceived a remote po ssibility of sustainable RS wood trade in coming years. Overall Perceptions of All Stakeholder Groups The overall priorities (Figure 4 6) under the strengths category across all stakeholder groups indicated overwhelming acceptance of the factor high price of RS wood as a main factor except for the trader stakeholder group who selected the factor niche international market available as the most important. Under the weaknesses category, the administrator and NGO s & academia stakeholder grou ps perceived the factor long gestation crop (+40 years) as most important whereas the landowner and trader stakeholder groups perceived factors small landholdings and lack of incentives as the biggest hindrance in developing a sustainable RS wo od trade, respectively. Under the opportunities category, the administrator and landowner stakeholder groups gave highest p mprovement in economic standard of landowners and handicraft artisans whereas the trader stakeholder group gave highest priority value to the factor forest conservation and growth Under the same SWOT category, the NGO s & academia stakeholder group gave highest priority to the factor improving public private relationship This implies t hat there was no clear consensus among stakeholder groups regarding opportunities category. There was no clear consensus among stakeholder groups regarding threats category as well. The administrator stakeholder group perceived the factor increas ed risk of illegal removals due to reduced Government control as the biggest threat whereas the landowner stakeholder group perceived that the factor small landholdings might cause significant bottlenecks in the development of sustainable RS wood tra de. The
118 trader and NGO s & academia stakeholder groups perceived the factor lack of coordination amongst external agencies as a significant threat that could jeopardize development of sustainable RS wood trade. The administrator and landowner s takeholder groups felt that external issues may cause more hindrance to the sustainable RS wood trade, which was opposite to the perception of the trader and NGO s & academia stakeholder groups who felt that internal issues were more important for the development of sustainable RS wood trade. Analysis of maximum overall priority values across all stakeholder groups for all SWOT categories revealed that maximum number of high s cores were for the weaknesses category. This implies that a majority of stakeholders perceived internal issues related to trade policy as hindrances in developing sustainable RS wood trade. Summary and conclusions RS, the flagship sp ecies of southern AP is currently facing extinction threat due to various reasons including illegal logging. Existing national mandates and international policies do not allow harvesting and trade of RS wood from public forests in any form. The only legal RS wood that enters in the market comes from the Government sources as the supply of RS wood from private lands is negligible. On the other hand there is a huge international demand for t his wood creating a wide demand supply gap. This imbalance in demand and supply is fuelin g illegal logging of RS trees from public forests and thus resulting in slow but steady damage to its survival chances. A c omplete ban on trade under CITES is one option to curtail illegal logging but success of this option is dependent on several extraneo us factors. Forests are open resources an d therefore, are difficult to protect by active policing alone. However, active community participation towards sustainable forest management could help in ensuring forest protection and
119 thus, higher benefits for co mmunities. In this regard, development of a policy emphasizing sustainable RS wood trade could be an appropriate step as this might help in reducing illegal logging of RS especially from public forestlands, ensuring higher benefits to local communities, an d bringing a sense of forest stewardship among public agencies, private sector, and local communities. For successful formulation of such a policy, it is imperative to understand perceptions of key stakeholder groups as their opinions can help in identifyi ng critical area s of interventions that a new sustainable RS wood trade policy should address. In this study perceptions of key stakeholder groups were analyzed with an assumption that these perceptions may be helpful for policymakers in critically review ing the existing rules and policies and in revising current policies or formulating new ones. This study shows a compelling presence of negative perceptions against the development of sustainable RS wood trade R esults indicate that in the existing set up, stakeholders participation would be limited in RS wood trade and if such a trade is to succeed, both external and internal issues will need to be addressed. The administrators for RS wood trade feel that RS wood trade, which is hitherto highly restricted due to internal (state and federal) and external (CITES) controls, may need opening up which in turn might result in increased smuggling. Analyse s also show prevalence of negative perception s which, being internal, can be addressed more easily. Further, if the public forests are to be protected, the policymakers may have to formulate a trade policy and create a suitable investment environment For instance, creation of a mechanism for accounting of RS trees planted by individual farmers and subsequent single window administrative clearance required for harvest should reduce
120 administrative delays. Similarly, extension facilities leading to availability of RS seedlings, field support and trainings for getting maximum yields and assurance for minimum suppo rt price would also encourage the landowners to adopt RS as a favored plantation species. I n centives for raising RS plantation s on private lands in the form of lower land taxes, and creating insurance mechanism against theft of RS trees may help in better adoption of RS For example in the United States, the federal crop insurance corporation (FCIC) promotes economic stability of agriculture through a system of crop insurance. A similar mechanism can be developed to reduce the uncertainty and risks associ ated with RS trees planted by individual farmers. Overall success of sustainable RS wood trade would require a balanced approach for control, sustainable management, and consumptive uses of RS derived from cropping on private lands. Further research by inc orporating participation from policymakers and higher administration would refine these findings. Research on improving user benefits for communities from RS forests and in developing and identifying genetically superior varieties of RS will certainly help in reducing the gestation period and therefore, will also result in better adoption of this species.
121 Table 4 1 List of identified factors under each SWOT categor y Helpful in achieving objectives Hindrance in achieving objectives Internal factors Strengths Weaknesses S1: Geographical advantage S2: High price of RS wood S3: Technical know how S4: Niche international market available W1: Administrative complexities W2: Long gestation crop (+40 years) W3: Lack of incentives W4: Small land holdings External factors Opportunities Threats O1: Feedstock for handicraft market O2: Improvement in economic standard of landowners and handicraft artisans O3: Improveme nt in public private relations O4: Forest conservation and growth T1: Competition fr om commercial agriculture crops T2: Increased risk of illegal removals due to reduced Government control T3: Change in market demand T4: Lack of coordination among external agencies Table 4 2 Pair wise comparison matrix in questionnaire for the Weakness category Please fill in only one The selected box should reflect your degree of priority towards one factor over the other. Pair wise comparisons for the SWOT category: WEAKNESSES A Very strong Strong Moderate Equal Moderate Strong Very strong B Administrative complexities Long gestation crop (+40 years) Administrative complexities Lack of Incentives Administrative complexities Small land holdings Long gestation crop (+40 years) Lack of Incentives Long gestation crop (+40 years) Small land holdings Lack of Incentives Small land holdings
122 Table 4 3 Overall relative priorities of factors and SWOT Factors SWOT categories and factors Factor priorities for stakeholder groups Overall priorities for stakeholder groups Administrator Landowner Trader NGO & Academia Administrator Landowner Trader NGO & Academia Strengths 0.1213 0.0866 0.2499 0.3750 S1: Geographical advantage 0.3055 0.2512 0.0956 0.0606 0.0370 0.0217 0.0239 0.0227 S2: High price of RS wood 0.3253 0.3533 0.4045 0.4386 0.0394 0.0306 0.1011 0.1645 S3: Technical know how 0.1339 0.1980 0.0802 0.1786 0.0162 0.0171 0.0201 0.0670 S4: Niche international market available 0.2354 0.1975 0.4197 0.3221 0.0285 0.0171 0.1049 0.1208 Weaknesses 0.3802 0.2224 0.5184 0.4497 W1: Administrative complexities 0.1730 0.1880 0.2466 0.1801 0.0658 0.0418 0.1278 0.0810 W2: Long gestation crop (+40 years) 0.3676 0.3023 0.1270 0.4061 0.1398 0.0672 0.0658 0.1826 W3: Lack of Incentives 0.1488 0.1500 0.4581 0.2428 0.0566 0.0334 0.2375 0.1092 W4: Small land holdings 0.3105 0.3596 0.1683 0.1710 0.1181 0.0800 0.0873 0.0769 Opportunities 0.0823 0.2931 0.1366 0.0958 O1: Feedstock for handicraft market 0.1910 0.1276 0.1494 0.1426 0.0157 0.0374 0.0204 0.0137 O2: Improv. economic standard 0.3018 0.4590 0.1652 0.2258 0.0248 0.1345 0.0226 0.0216 O3: Improvement in public pvt. relations 0.2167 0.1541 0.1938 0.3170 0.0178 0.0452 0.0265 0.0304 O4: Forest conservation and growth 0.2906 0.2593 0.4917 0.3146 0.0239 0.0760 0.0671 0.0301 Threats 0.4162 0.3979 0.0951 0.0795 T1: Com. from com. agri. crops 0.2288 0.4724 0.1165 0.2645 0.0952 0.1880 0.0111 0.0210 T2: Increased risk of illegal removals 0.3114 0.2266 0.1912 0.2152 0.1296 0.0902 0.0182 0.0171 T3: Change in market demand 0.2461 0.1130 0.0892 0.1975 0.1024 0.0450 0.0085 0.0157 T4: Lack of coord. among ext. agencies 0.2137 0.1880 0.6031 0.3228 0.0890 0.0748 0.0574 0.0257 Numbers with are highest factor priority values within a SWOT category for a particular stakeholder group. Numbers in italics are the overall priority values for a SWOT category for a particular stakeholder group. Five highest overall priority values for each stakeholder group are underli ned for comparison across all SWOT groups. Improv. economic standard = Improvement in economic standard of landowners and handicraft artisans, Improvement in public pvt. relations = Improvement in public private relations, Com. from com. agri. Crops = Competition from commercial agriculture crops, Increased risk of illegal removals = Increased risk of illegal removals due to reduced Government control, Lack of coord among ext. agencies = Lack of coordination among external agencies
123 Figure 4 1 Red Sanders forests in Andhra Pradesh in India. (Source: Geometics cell, APFD, Hyderabad, 2007)
124 Figure 4 2 Perceptions of the Administrators stakeholder group
125 Figure 4 3 Perceptions of the Landowners stakeholder group
126 Figure 4 4 Perceptions of the Traders stakeholder group
127 Figure 4 5 Perceptions of the NGOs & Academia stakeholder group
128 Figure 4 6. Overall trends and SWOT category
129 CHAPTER 5 CONCLUSIONS POLICY IMPLICATION S AND FUTURE RESEARC H Conclusi ons In developing countries there often exists a deep relationship between culture and ecosystems. Associated with this is the fact that causes of degradation often involve socio economic factors. This research is an attempt to investigate the strategies f or successful restoration of Red Sanders, an endemic and endangered species in the Southern Andhra Pradesh, India It lays emphasis on the fact that the physical restoration of species cannot succeed without involvement of stakeholders in restoration plann ing and process. Other than designing an integrated restoration strategy, restoration may require revisiting the existing policies and / or formulating newer ones. Further, restoration efforts should also be integrated with environment al and socio economic benefits f or local communities. The major conclusions that can be drawn from this study are listed by chapter and followed by policy implications, limitations, and scope for future research. Chapter 1 details the history of Red Sander s (RS) forests through a comprehensive literature search and by drawing on the experience in these forests. The main conclusions arising from the review include: 1. Past o verexploitation, continued anthropogenic disturbances and dr o ugh t are the main causes of degradation of Red Sanders landscape. 2. Over the passage of time, the local communities have distanced themselves from these forests and currently there is no sense of ownership. 3. Unrestricted grazing, recurrent fires and illegal logg ing are reducing density of Red Sanders in the wild.
130 4. The demand supply gap that exists due to huge international demand for Red Sanders wood and lack of supply from public forests is fueling illegal logging. 5. participation in cultivation and trade of wood is very limited 6. Restoration strategies duly integrating biophysical methods at a landscape level, review of existing policies in the lig ht of stakeholders preferences, and strategies for socioeconomic and environmental benefits for local communities would be helpful. The result s in Chapter 2 provide important insights on the interactions of silvicultural treatments with ecological factors to influence seedling survival and establishment. These findings can be used to draw recommendations for restoration of RS or similar endangered forests. The se include: 1. The silvicultural treatments can ameliorate site conditions to improve survival and gr owth of young regeneration. 2. The seedlings with fewer coppice shoots showed consistently better and statistically significant survival across all the treatments H ence treatments involving removal of coppice shoots and preferential treatment of seedlings w ith fewer coppice shoots would be helpful in improving seedlings density. 3. The RS seedlings with higher number of coppice shoots showed better growth when excess shoots are removed by singling. Removal of coppice shoots is not only good for survival and gr owth, but also for future quality of the RS forests. 4. Treatments with disking have significant ly higher survival and growth advantage over the treatments without disking however, the treatment with disking were not
131 the best suited in all cases. This indicates that disking is useful only when the soils are highly compacted, e.g. areas exposed to high domestic cattle grazing. Soil analysis may be helpful in making the appropriate treatment decisions. 5. The tall seedlings showed better height, RCD an d volume growth in term s of absolute growth. Hence, in case seedling density is not limited, taller seedlings should be preferentially treated for successful restoration. 6. The younger RS seedlings showed better relative growth compared to the older seedlin gs This suggests that if the seedling density is limited, preferential treatment to the younger seedlings woul d help in successful restoration. Chapter 3 is devoted to a socioeconomic and environmental assessment of Bodagrass ( BG ) utilization for biofuel s Bodagrass is abundantly available in Red Sanders forests and currently has no apparent use. 1. The results indicate energetic, ecological and economic benefits of utilizing Bodagrass for ethanol production in southern Andhra Pradesh (AP) Since the grass g rows naturally in the forests, biomass production does not result in land use change and does not require inputs. Its free availability to local communities will therefore address the issue of poverty alleviation on one hand and ecological benefits on the other. Besides, it will contribute to energy security. 2. Life cycle assessment results of conversion of B G into ethanol involves consumption of 26.21 MJ L 1 .The net energy value of 16 86 MJ L 1 and a net energy ratio of 5.15 indicate lower energy inputs rel ative to the energy produced.
132 3. The entire process of conversion of biomass to ethanol and use as motor fuel (E10) has lower environmental burdens and impacts as compared to the status quo situation of use of gasoline as motor fuel and open burning of Bodag rass. 4. The cost of ethanol production based on 2008 rates was $0.42 L 1 which was competitive with the 2008 fuel ethanol price of $0.48 L 1 in India. 5. The cost of ethanol production is sensitive to biomass cost which accounts for 21% of the unit cost Wit h a feedstock cost range of $23.79 t 1 to $30.01 t 1 the corresponding cost of production was $0.40 L 1 to $0.53 L 1 6. The disposal of this grass from forests has potential to reduce wildfires, thereby improving restoration of Red Sanders forests. Chapter 4 sustainable R S wood trade. Using a SWOT AHP framework, the development of RS wood trade as a strategy for reducing illegal logging and improving economic wellbeing of farmers was evalua ted in this study. R esults indicate that: 1. The negative perceptions outweigh the positive ones towards the development of sustainable RS wood trade Since the negative perceptions are internal to any organization they can be easily addressed by suitable rev iew In the existing conditions very few stakeholders are expected to participate in RS wood trade 2. The administrators for RS wood trade feel that this trade, which is hitherto highly restricted due to internal (state and federal) and external (CITES) co ntrols, may need opening up which in turn might result in increased smuggling. 3. The landowners perceive that the competition from other commercial crops is one of the most important factors in the adoption of RS growing.
133 4. The traders and NGOs & academia per ceive that the existing Government policy is hindering RS wood trade and feel that suitable modification s are essential for promoting RS trade. T he strengths of the RS market can be better exploited if its weaknesses are addressed 5. For the development of sustainable trade, both external and internal issues will need to be addressed. The pr omotion of RS trade may require review of existing trade policy, and / or formulating new ones including creating a suitable investment environment. 6. Incentives for the farmers willing to raise RS plantations on their lands backed by extension, market support and insurance might improve private landowners involvement in trade. Ecological Benefits This study deals with three most important aspects related to RS; first, restoration of RS at species level ; second, value addition to RS forests and third, policy issues related to RS wood markets Besides, the additional benefits from and of RS forests have also been identified. For instance, conversion of Bodagrass collected from RS forests has potential of assisting regeneration providing energy security and reducing wildfires RS is the flagship species of the forests of southern AP. Therefore it is assumed that an attempt to save RS would have positive effect on overall e cology and environment as well. For example, protection and restoration of RS may have positive impact on other endangered flora like Cycas beddomie, Decalepis hamiltonii and fauna like Loris tardigradus found in RS forests. RS is the dominant tree species of the forests of southern AP. Even though, i n the mineral rich soils of these areas, RS has thrived for many centuries, m any of its unique properties are still unexplored. For example, RS
134 wood is a known source alkaloids and medicinal properties of some of these have been documented. However, there are still many which been not been studied for their medicinal properties yet Similarly, the Strontium and Cadmium accumulat ion properties of RS have not been further researched. Th is species is known to exist in pure patches where it cannot be replaced and in mixed forests where the loss of this species may initiate a loss of ecosystem function and dynamics Form the results of the study and discussion s it can be safely concluded that restoration of RS in these forests is of great ecolog ical importance. Policy Implications The endangered and endemic nature of RS demands that the species be protected and conserved at the landscape level. Hence, restoration activity may have to be tailored to this land scape at different levels of management based on seedling density, biotic and abiotic factors, and soil type. In addition, infrastructure, budget and other logistic limitations may be taken into account in select ing the best possible combination of silvicu ltural tools for restoration. Future RS management and restoration policies formulated by involv ing local stakeholders may help in long term survival of this species. Sinc e local communities are dependent on RS forests, identification of value small scale community based projects may help in improving community ownership of RS forests. For instance, conversion of Bodagrass from RS forests into Ethanol may help in improving opportunities for enhancing rural livelihoods. Further, such endeavors can be built around area specific biomass availability. Since India cannot afford the diversion of arable lands for biofuels, identification of low input biofuels such as utilization of BG would have far reaching implications for the economy and environment The positi ve energy ratio of conversion of Bodagrass into ethanol can be seen as a motivation for
135 adoption. Besides providing energy security it has a potential of enhancing socioeconomic wellbeing of rural communities The Government of India and / or Government of AP may consider installing a pilot scale plant for production of ethanol plant using Bodagrass as feedstock to ascertain feasibility of such an endeavor. Overall success of sustainable RS wood trade wou ld require a balanced approach of administrative cont rol and incentives for adoption of RS plantations on private land s. For instance, by introducing tax relief on income from RS plantations insurance mechanism against theft of RS trees and wood, and by creating buy back guarantee mechanism s for the RS wood the Government can create a positive atmosphere to improve local farmers participation Such policy initiatives will help the protection of public forests and also result in economic benefits for farmers and handicraft artisans. A RS restoration consorti um, consisting of representatives from state and national Government agencies, international Government and aid agencies, non governmental organization s, local communities, and the private sector representatives must be formed to ensure collaboration and c oordination of these activities. Limitations of the Study and Scope of Future Research The findings of the effect of silvicultur al treatments on young RS regeneration can be further improved by conducting long term research to refine and formulate cost eff ective restoration strategies for this species. The addition of evaluation of social and economic impacts by using Life Cycle Costing (LCC) and Social Life Cycle Assessment (SLCA) can improve t he scope of life cycle assessment study perc eption evaluation can be further improved by incorporating more participation from policymakers and higher administration. Further research on improving user benefits for communities from RS forests and in developing and identifying genetically superior
136 va rieties of RS will certainly help in reducing the gestation period and therefore, will also result in better adoption of this species.
137 APPENDIX A RED SANDERS SELECTED PHOTOGRAPHS Figure A 1 Red Sanders: a) Trees in urban area ; b) Plantation on public forests Figure A 2 Twig of Red Sanders showing leaves and seeds
138 Figure A 3 Red Sanders plantation on public forests Figure A 4 Natural Red Sanders forest with Bodagrass (Cymbopogon coloratus) understory with fire scarring on tree trunks
139 Figure A 5 Red Sanders seedlings: a) with post fire died back shoots; 2) Regenerated multiple coppice shoots Figure A 6 Mature Red Sanders tree bark
140 Figure A 7 Red Sanders heartwood (a) without grain (b) with wavy grain Figure A 8 Uses of Red Sanders wood: a) carved museum article b) musical instruments c) and d) imitation furniture e ) carved statues
141 Figure A 9 Illegal RS wood seized by the local forest department Figure A 10 Illegal RS wood seized at port in a shipping container (Source: D irectorate of revenue intelligence, India)
142 Figure A 11 Musical instrument parts seized by authorities while in transport ( source: Directorate of revenue intelligence, India )
143 APPENDIX B SUPPLEMENT TO CHAPTE R 3 D ESCRIPTION OF IMPA CT CATEGORIES Global Warming Potential climate caused by the buildup of greenhouse gases that trap heat from the reflected sunlight that would have otherwise passed out global warming potentials, a midpoint metric proposed by the International Panel on Climate Change (IPCC, 2007) for the calculation of the potency of greenhouse gases relative to CO 2 The 100 year time horizons recomme nded by the IPCC were adopted within TRACI. The final sum, known as the global warming index, indicates the potential contribution to global warming and is calculated as: Global Where, mi is the emission (in kilograms) of subs tance i and GWPi is the global climate ch ange potential of substance i. Acidification Potential Acidification is a phenomenon resulting from processes that increase the acidity (hydrogen ion concentration, [H + ]) of water and soil systems. Changes in the alkalinity of lakes, related to their acid neutralizing capacity, are used as a diagnostic for freshwater systems analogous to the use of H + budgets in terrestrial watersheds. Acid deposition also has deleterious (corrosive) effects on buildings, monuments and historical artifacts. The resulting acidification characterization factors are expressed in H + mole equivalent deposition per kilogram of emission and are dependent on the specific emission. Characterization factors take account of expected differenc es in total deposition as a result of the pollutant release location.
144 Smog Potential Nitrogen oxides (NO x ) and volatile organic compounds (VOCs) are emitted into the atmosphere from many natural and anthropogenic processes. In the atmosphere, these substances enter a complex network of photochemical reactions induced by ultraviolet light (UV light) from the sun. These reactions lead to the formation of ozone (O3), per oxy acetyl nitrate (PAN), per oxy benzoyl nitrate (PBN), and a number of other subs tances in the troposphere. The photochemical smog compounds degrade many materials and are toxic to humans, animals, and plants. The smog can be observed as a reddish brown cast in the air above many cities. In general, characterization factors estimate th e smog formation potential of the release of chemicals in terms of NO x The approach to smog characterization analysis for VOCs and NO x in TRACI incorporates the relative influence of individual VOCs on smog formation, and relative influence of NO x concent rations versus average VOC mixture on smog formation. Human Health The cancer and non cancer human health impacts measure the potential of a chemical released into the environment to cause a variety of specific human cancer and non cancer effects, respect ively (Bare et al., 2003). The relative toxicological concern of an emission in the context of human health is currently calculated in TRACI based on human toxicity potentials (HTPs) (Hertwich et al., 2001). The HTP is an indicator used to compare the rela tive importance of toxic emission in situations where a site specific risk assessment would be too expensive or data on the release sites is not always available (Hertwich et al., 2001).
145 Water Use Water use has generally been tracked in simple mass or volu me terms in life cycle inventory without subsequent characterization analysis that would weight different usage flows to take into account important differences among source types and usage locations. Rather than trying to capture the addition of water po llutants into the environment, this impact category is structured to capture the significant use of water in areas of low availability. As an impact assessment methodology for water use is not incorporated within TRACI, keeping in view the low water availa bility in the study area the volume of water use in ethanol production was calculated in this study
146 LIST OF REFERENCES Aczel, J., Saaty, T. L., 1983. Procedures for synthesizing ratio judgments, J. Math. Psychol. 27, 93 102. Adams, C. R., Galatowitsch, S. M., 2006. Increasing the effectiveness of Reed canary grass (Phalaris arundinacea L.) control in wet meadow restorations. Restor. Ecol.14, 441 451. Aden A ., Ruth, M., Ibsen, K., Jechura, J., Neeves, K., Sheehan, J., Wallace, B. 2002. Lig nocellulosic biomass to ethanol process design and economics utilizing co current dilute acid prehydrolysis and enzymatic hydrolysis for corn stover. Technical Report NREL/TP 510 32438. National Renewable Energy Laboratory. Golden, Colorado. Aguiar, M. R., Soriano, A., Sala, O. E., 1992. Competition and facilitation in the recruitment of seedlings in Patagonian steppe, Funct. Ecol. 6, 66 70. Ahmed M., Nayar M. P., 1984. Red Sanders tree (Pterocarpus santalinus Linn. F.) on the verge of depletion, Bull. B ot. Surv. India, 26 (1 & 2), 142 143. AIDA, 2006. All India Distillers Association. www.aidaindia.org (accessed: December 12th, 2010). Aide, T. M., Cavelier, J., 1994. Barriers to lowland tropical forest restoration in the Sierra Nevada de Santa Marta, Colombia. Restor. Ecol. 2, 219 229. Ahmed, M., Nayar, M. P., 1984. Red Sanders tree (Pterocarpus santalinus Linn. F.) on the verge of depletion. Bull Bot. Surv. In dia. 26(1 2), 142 143. Alle n, D. T., Shonnard, D. R. 2002. Green engineering: environmentally conscious design of chemical processes. Prentice Hall PTR. Upper Saddle River, NJ. Ananda, J., Herath, G., 2003. The use of analytic hierarchy process to incorporate stakeholder preferences into regional forest planning. Forest Policy Econ 5, 13 26. Anuradha, M., Pullaiah, T. 1999, Propagation studies of red sanders (Pterocarpus sanalinus L.f.) in vitro as endangered taxon of Andhra Pradesh, India Taiwani a, 44:311 24 APFD, 2010. Andhra Pradesh Forest Department, Boda grass expression of interest http://forest.ap.nic.in/Prod%20Boda%20Grass.htm (accessed: January 4th, 2010) APFD, 2010 Andhra Prade sh Forest Department (APFD), 2010 facts and figures, http://forest.ap.nic.in/ (accessed: September 24th, 2010)
147 APFD, 2010. Andhra Pradesh Forest Department, Inventory Report, http://forest.ap.nic.in/inventory.htm (accessed: 25th March, 2010). APGenco, 2010. Andhra Pradesh State Power Generation Company, Power generation performance http://www.apg enco.gov.in/inner.asp?frm=glance (accessed December 28 th 2010). Augspurger, C. K., 1984. Seedling Survival of tropical tree species: interactions of dispersal distance, light gaps, and pathogens. Ecology. 65, 1705 1712. Awanyo, L., Attuah, E. M., McCar ron, M., 2011. Rehabilitation of forest savannas in Ghana: The impacts of land use, shade, and invasive species on tree recruitment. Appl. Geogr. 31, 181 190. Bare, J., Norris, G., Pennington, D., McKone, T., 2003. The Tool for the Reduction and Assessmen t of Chemical and other Environmenta l Impacts. J. Ind. Ecol. 6(3 4), 49 78. Biswas, T. K., Mukherjee, B., 2003. Plant medicine of Indian origin for wound healing activity: a review. IJLEW. 2(1), 25 39. B lottnitz, V. H., Curran, M. A., 2007. A review of ass essments conducted on bio ethanol as a transportation fuel from a net energy, greenhouse gas, and environmental life cycle perspective. J. Cleaner Prod. 15, 607 619. Bouffard, D., Nolet, P., Del agrange, S., Lorenzetti, F., Yamasaki, S. 2007. V egetation control treatments to favor naturally regenerated Betula alleghaniensis Saplings Following Seed Tree Cut: Sapling Monitoring Two Years after Treatment. Restor. Ecol. 15, 679 687. Brandeis, T. J., Newton, M., Cole, E., 2001. A comparison of overstory den sity measures for describing understory conifer growth.For. Ecol. Manage. 152(1 3), 149 157. Brander, J. A., Taylor, M. S., 1997. International Trade between Consumer and Conservationist Countries, Resour Energy Econ. 19(4), 267 298. Brugha, R., Varvasovsz ky, Z., 2000. Stakeholder Analysis: a review, Public Health Plan. 15(3), 239 246. Bu llock, S. H., Mooney, H. A. Medina. E. 1995. Seasonally dry tropical forests. Cambridge University Press, New York. Bunker, D. E., Carson, W. P., 2005. Drought stress an d tropical forest woody seedlings: effect on community structure and composition. J. Ecol. 93, 794 806. Campbell, H., Brown, R., 2003. Benefit cost analysis: financial and economic appraisal using spreadsheets. Cambridge University Press, UK
148 Carlson, C. A., Fox, T. R., Colbert, S. R., Kelting, D. L., Allen, H. L., Albaugh, T. J., 2006. Growth and survival of Pinus taeda in response to surface and subsurface tillage in the southeastern United States. For. Ecol. Manage. 234, 209 217 Carpenter, F. L., Nichol s, J. D., Pratt, R. T., Young, K. C., 2004. Methods of facilitating reforestation of tropical degraded land with the native timber tree, Terminalia amazonia For. Ecol. Manage. 202, 281 291. Casselman, C. N., Fox, T. R., Burger, J. A., Jones, A. T., Galbr aith, J. M., 2006. Effects of silvicultural treatments on survival and growth of trees planted on reclaimed mine lands in the Appalachians. For. Ecol. Manage. 223, 403 414. CITES, 2001. CITES and wood product trade What you should know brochure; http://www.fws.gov/international/DMA_DSA/CITES/pdf/Cites_eng.pdf (accessed: May 14th 2010). CITES, 2007. Review of Significant Trade in specimens of Appendix II species PC17 Doc 8.4 http://www.cites.org/eng/com/PC/17/E PC17 08 04.pdf (accessed: May 14th 2010). Champion, H. G. and Seth, S. K., 1968 A revised survey of the forest types of India. Govt. India Press. Chopra, R. N., Nayar, S. L., Chopra, I. C.,1956. Glossary of Indian Medical Plants. New Delhi: Council for Scientific and Industrial Research Publications. Crawley, M. J., 2002. Statistical computing: An introduction to data analysis using S plus, John Wiley and Sons, Hoboken, New Jersey. Coughlin, K., Fridley, D., 2008. Physical Energy Accounting in California: A Case Study of Cellulosic Ethanol Production, Ernest Orlando Lawrence Berkeley National Laboratory, Berkeley, CA (US) http://www.osti.gov/energycitations/product.biblio.jsp?osti_id=953684 (accessed: March 15th, 2009) Demers, C., Long, A., Minogue, P., 2011 Longleaf Pine regeneration publication #SS FOR 13 http://edis.ifas.ufl.edu/fr064 (accessed: January 20th, 2011) DRI, 2005. Directorate of Revenue Intelligence, Annual report, 2005 06. www.dri.nic.in/DRI/ANNUAL_REPORT_2005 06.pdf (a ccessed: March 28th 2010. Duncan, R. S., Chapman, C. A., 2003. Consequences of plantation harvest during tropical forest restoration in Uganda. For. Ecol. Manage. 173, 235 250. Dupuy, J. M., Chazdon, R. L., 2008. Interacting effects of canopy gap, understory vegetation and leaf litter on tree seedling recruitment and composition in tropical secondary forests. For. Ecol. Manage. 255, 3716 3725.
149 Dwivedi, P., Alavalapati, J. R. R., 2009. eptions on forest biomass based bioenergy development in the southern US, Energ. Policy 37, 1999 2007. EIA, (2008) http://www.eia.doe.gov/pub/international/iealf/tablee1g.xls (ac cessed: December 28th, 2010 ) Espinoza, J. A., 2004. Site selection, site preparation and weed control for Gmelina arbore a in western Venezuela. Ne w F or. 28, 2 17 226. Ewel, J., 1980. Tropical succession: manifold routes to maturity. Biotropica 12, 2 7. FAO 2003 World soil resources, Food and agriculture organization of the United Nations, European commission joint research centre, International Soil Reference and Informati on Centre. http://www.fao.org/ag/agl/agll/wrb/wrbmaps/htm/soilres.htm (accessed: January 25th, 2011) FAO 2005 Global Forest Resources Assessment 2005: Progress towards sustainable forest management. FAO Forestry Paper 147. Food and Agriculture Organisation of the United Nations, Rome. ftp://ftp.fao.org/docrep/fao/008/A0400E/A0400E00.pdf ( accessed : December 19th, 2010) Farrell, A. E., Plevin, R. J., Turner, B. T., Jones, A. D., O'Hare, M., Ka mmen, D. M., 2006. Ethanol can contribute to energy and environmental goals. Science. 311, 506 508. Feddema, J., 2005. A revised Thornthwaite type global climate classification. Phys. Geogr. 26, 442 466. Forman, E., Peniwati, K., 1998. Aggregating individ ual judgments and priorities with the Analytic Hierarchy Process. Eur. J. Oper. Res. 108, 165 169. FSI 2009 India State of Forest Report, 200 9 ., Ministry of Environment and Forests Government of India, Dehra Dun, India. http://www.fsi.nic.in/sfr_2009.htm accessed : November 10 th 2010) Gerhardt, K., 1993. Tree seedling development in tropical dry abandoned pasture and secondary forest in Costa Rica. J. Veg. Sci. 4, 95 102. Gerhardt, K., 1996a. Effects of root competition and canopy openness on survival and growth of tree seedlings in a tropical seasonal dry forest. For. Ecol. Manage. 82, 33 48.
150 Gerhardt, K., 1996 b Germination and development of sown Mahogany ( Swietenia m acrophylla King) in secondary tropical dry forest habitats in Costa Rica. J. Trop. Ecol.12, 275 289. Godefroid, S., Monbaliu, D., Massant, W., Van der Aa, B., De Vos, B., Quivy, V., Koedam, N., 2007. Effects of soil mechanical treatments combined with bra mble and bracken control on the restoration of degraded understory in an ancient beech forest. Ann. For. Sci. 64, 321 331. Guariguata, M. R., Pinard, M. A. 1998. Ecological knowledge of regeneration from seed in neotropical forest trees: Implications for natural forest management. For. Ecol. Manage.112, 87 99. Guine, J. B., 2002. Handbook on life cycle assessment: Operational guide to the ISO standards. Kluwer Ac ademic, The Netherlands. Gujrati, D. 2004 Basic econometrics, 4 economyth edn. Tata McGraw Hill, New Delhi Gunaratne, A. M. T. A., Gunatilleke, C. V. S., Gunatilleke, I. A. U. N., Madawala Weerasinghe, H. M. S. P., Burslem, D. F. R. P., 2010. Barriers to tree seedling emergence on human induced grasslands in Sri Lanka. J. Appl. Ecol. 47, 157 165. Hamelinck, C. N., Hooijdonk, G. V., Faaij, Andr P. C., 2005. Ethanol from lignocellulosic biomass: techno economic performance i n short, middle and l ong term, Biomass Bioenerg., 28 ( 4 ) 384 410 Harcomb e, P. A., 1987. Tree Life Tables, Bioscience 37, 557 568. Hardwick, K., Healey, J., Elliott, S., Garwood, N., Anusarnsunthorn, V., 1997. Understanding and assisting natural regeneration processes in degraded seasonal evergreen forests in northern Thailand. For. Ecol. Manage. 99, 203 214. Hertwich, E. G., Mateles S. F., Pease, W. S., McKone T. E., 2001. Human toxicity potentials for life cycle assessment and toxics release inventory risk screening. Env iron. Toxicol. Chem. 20 (4):928 939. Hill, J., Nelson, E., Tilman, D., Polasky, S ., Tiffany, D., 2006. From the cover: e nvironmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. P Natl. Acad. Sci., 103, 11206 11210. Hoffmann, W. A., 1996. The effects of fire and cover on seedling establishment in a neotropical savanna. J. Ecol. 84, 383 393. Holl, K. D., Loik, M. E., Lin, E. H. V, Samuels, I. A., 2005. Tropical montane forest restoration in Costa Rica: Overcoming barriers to dispersal and establish ment, Restor. Ecol. 8 (4) 339 349.
151 Honu, Y. A. K., Dang, Q. L., 2002. Spatial distribution and species composition of tree seeds and seedlings under the canopy of the shrub, Chromolaena odorata Linn ., in Ghana. For. Ecol. Manage.164, 185 196. Howe, H. F., 1990. Survival and growth of juvenile Virola surinamensis in Panama: effects of herbivory and canopy closure. J. Trop. Ecol. 6, 259 280. ISO, 2006. Environmental Management Life cycle assessment Requirements and guidelines. International Standard ISO 14044:2006 (E). second edition, International Organization for Standar dization. Geneva. IPCC, 2007 Intergovernmental Panel on Climate Change, Climate change 2007: Mitigation. Contribution of working group iii to the fourth assessment report of the intergovernmental panel on climate change. IPCC. 8 p. IUCN, 2010. Internation al Union for Conservation of Nature Red list of threatened species. http://www.iucnredlist.org/apps/redlist/details/32104/0 (a ccessed: December 15 th 2010). Jain, S. K., Rao, R. R., (Eds.), 1983 An assessment of threatened plants of India. Proceedings of the seminar held at Dehra Dun, 14 to 17.09.1981. Botanical Survey of India, Dept. Environment, Howrah, India. Johnson, R., Gjerstad, D., 2006. Restoring the overstory of Longleaf pine ecosystems, in: Jose, S., Jokela, E. J., Miller, D. L. (Eds), The longleaf pine ecosystem: ecology, silviculture, and restoration, Springer, Jose, S., Merritt, S., Ramsey, C. L., 2003. Growth, nutrition, photosynthesis and transpiration responses of long leaf pine seedlings to light, water and nitrogen. For. Ecol. Manage.180, 335 344. Kadam, K. L., 2000 Environmental life cycle implications of using bagasse derived Ethanol as a gasoline oxygenate in Mumbai (Bombay), Technical report, NREL/TP 580 28705, N ational Renewable Energy Laboratory. Golden, Colorado. Kathiawada, D., Silvereira, S., 2009. Net energy balance of molasses based ethanol: The case of Nepal, Renew. Sust. Energ. Rev. 13, 2515 2524. Kajanus, M., Kangas, J., Kurttila, M., 2004. The use of value focused thinking and the gement, Tourism Manage. 25, 499 506. Kangas, A., 1994. An approach to public participation in strategic forest management planning, For. Ecol. Manage 75 88. Kedharnath, S., Rawat M. S., 1976. Studies on variation in fibre morphology in wavy grained and straight grained trees of Red Sanders. Indian For. 102. 441 6.
152 Kennard, D. K., Gould, K., Putz, F. E., Fredericksen, T. S ., Morales, F., 2002. Effect of disturbance intensity on regeneration mechanisms in a tropical dry forest. For. Ecol. Manage.162, 197 208. Kirtikar, K. R., Basu, B. D., 1983. Indian Medicinal Plants 1, Periodical Experts, New Delhi, India. Kiyono, H., 20 05. TRAFFIC Japan, http://www.cites.org/eng/com/PC/17/E PC17 08 04.pdf (accessed: March 5th, 2010). Khurana, E., Singh, J. S., 2001. Ecology of seed and seedling growth for conservation and restoration of tropical dry forest: a review. Environ. Conserv. 28, 39 52. Knapp, B. O., Wang, G. G., Walker, J. L., Cohen, S., 2006. Effects of site preparation treatments on early growth and survi val of planted longleaf pine (Pinus palustris Mill.) seedlings in North Carolina. For. Ecol. Manage. 226, 122 128. Kotler, P. (Ed.), 1994. Marketing Management: Analysis, Planning, Implementation and Control, 8th ed. Prentice Hall, Englewood Cliffs. NJ. US A. Koo, B., Wright, B. D., 1999. The role of biodiversity products as incentives for conserving biological diversity: some instructive examples, Sci. Total Environ. 240 (1 3 ), 21 3 0. Kurttila, M., Pesonen, M., Kangas, J., Kajanus, M., 2000. Utilizing the analytic hierarchy process (AHP) in SWOT analysis a hybrid method and its application to a fores t certification case. For. Polic y Econ. 1(1), 41 52. Kolb, P. F., Robberecht, R., 1 996. Pinus ponderosa seedling establishment and the influence of competition with the bunchgrass Agropyron spicatum Int. J. Plant Sci.157, 509 515. Kozlowski, T. T., 2002. Physiological ecology of natural regeneration of harvested and disturbed forest sta nds: implications for forest management. For. Ecol. Manage. 158, 195 221. Krishnaveni, K. S., Rao. S., 2000. A new triterpene from callus of Pterocarpus santalinus. Fitoter. 71, 10 3 Lal, J. B., 1 989. myth and reality, Natraj Publishers, Dehradun, India Lamb, D., Tomlinson, M., 1994. Forest rehabilitation in Asia Pacific region: Past lessons and present uncerta inties. J. Trop. For. Sci. 7(1), 157 170. Li, Q., Ma, K., 2003. Factors aff ecting establishment of Quercus liaotungensis Koidz. under mature mixed oak forest overstory and in shrubland. For. Ecol. Manage.176, 133 146.
153 Maciejewska, A., Veringa, H., Sanders, J., Peteves, S. D., 2006. Co firing of biomass with coal: constraints and role of biomass pre treatment, DG JRC Institute for Energy, http://www.techtp.com/Cofiring/Cofiring%20biomass%20with%20Coal.pdf (accessed : Jan 4 th 2011). Madison, L. A., Barnes, T. G., Sole, J. D., 2001. Effectiveness of Fire, Disking, and Herbicide to Renovate Tall Fescue Fields to Northern Bobwhite Habitat. Wildl. Soc. Bull. 29, 706 712. Manjunatha, B. K., 2006. Antibacterial Activities of Pterocarpus santalinus, Indian J. Pharm. Sci 68. (1), 115 116. Masozera, M. K., Alavalapati, J. R. R., Jacobson, S., Shrestha, R. K., 2006. Assessing the suitability of community based management for the Nyungwe Forest Reserve, Rwanda. Forest Policy Econ. 8, 206 216. Marod, D., Kutintara, U., Tanaka, H., Nakashizuka, T., 2004. Effects of drought and fire on seedling survival and growth under contrasting light conditions in a sea sonal tropical forest. J. Veg. Sci. 15, 691 700. Marrero Gmez, M.V., Ramn Arvalo, J., Baares Baud et, ., Carqu lamo, E., 2000. Study of the establishment of the endangered Echium acanthocarpum (Boraginaceae) in the Canary Islands. Biol. Conserv. 94, 183 190. McAloon, A., Taylor, F., Yee, W., Ibsen, K., Wooley, R., 2000. Determining the Cost of Prod ucing Ethanol from Corn Starch and Lignocellulosic Feedstocks, Technical report, NREL/TP 580 28893, National Renewable Energy Laboratory, Golden, Colorado McLaren, K. P., McDonald, M. A., 2003. The effects of moisture and shade on seed germination and seedling survival in a tropical dry forest in Jamaica. For. Ecol. Manage.183, 61 75. Molur, S., Ved, D. K., Tandon, V., Namboodiri, N., Walker, S. (Eds.), 1995. Conservation assessment an d management plan (CAMP) for selected species of medicinal plants of southern India. Mulliken, T., Crofton, P., 2008. Review of the Status, Harvest, Trade and Management of Seven Asian CITES listed medicinal and aromatic plant species, http://www.cites.org/common/com/PC/17/X PC17 Inf 10.pdf (accessed: March 25th, 2010) NMDT, 2006. National Multi Disciplinary Team Minutes of the first meeting, New Delhi, India. http://sudiv.gov.in/PDFFiles \ What'sNew \ NewInc/MinuteDRI.pdf (Accessed March 15th 2010).
154 NREGA, 2010.National Rural Employment Guarantee Act Prevailing Minimum Wages & Schedule of Rates (SOR) for Agric ultural Labourers (INR / Day) http://nrega.nic.in/wages.pdf (accessed: January 4th, 2011) Nagaraju, A., Raju, K. K., 1998. Geobotany of Red Sanders ( Pterocarpus santalinus L.) A case study from the southeaste rn portion of Andhra Pradesh, Environ. Geol. 37 (4), 340 344. Narayan, S., Devi, R. S., Devi, C. S., 2007. Role of Pterocarpus santalinus against mitochondrial dysfunction and membrane lipid changes induced by ulcerogens in rat gastri c mucosa, Chem. Biol. Interact. 170, 67 75. Negi, A. S., Negi, G. C. S., Singh, S. P., 1996. Establishment and growth of Quercus floribunda seedlings after a mast year. J. Veg. Sci. 7, 559 564. Nguyen T. L. T., Ghe ewala S. H., Garivait, S., 2007. Full Cha in Energy Analysis of Fuel Ethanol from Cassava in Thailand. Environ. Sci. Technol. 14 (11), 4135 4142 Odum, H. T., 1983. Systems Ec ology. Wiley, New York, pp. 476 507. Oldfield, S., Lusty, C., Mackinven, A., 1998. The world list of threatened trees Worl d Conservation Press, Cambridge, UK. Pandey, A., Biswas, S., Sukumaran, R. K., Kaushik, N., 2009. Study on availability of Indian biomass resources for exploitation: a report based on a nationwide survey. TIFAC, New Delhi Parrotta, J. A., Knowles, O. H., 1999. Restoration of tropical moist forests on bauxite mined lands in the Brazilian Amazon. Restor. Ecol. 7, 103 116. Paula, S., Pausas, J. G., 2011. Root traits explain different foraging strategies between resprouting life histories Oecologia. 165, 321 331 Pesonen, M., Kurttila, M., Kangas, J., Kajanus, M., Heinonen, P., 2001. Assessing the Forest and Park Service, For. Sci. 47 (4), 534 541. Pimentel, D., Patzek, T. 2005. Et hanol production using corn, switchgrass, and wood: biodiesel production using soybean and s unflow er. Nat. Resour. Res., 14( 1 ) 65 76. Planning Commission, 2003. Report of the committee on development of biofuels. Planning Commission, Government of India. http://planningcommission.nic.in/reports/genrep/cmtt_bio.pdf (accessed : December 19th, 2010). Prakash, R., Henham, A., Bhat I. K., 1998. Net energy and gross pollution from bioethanol production in India. Fuel. 77( 14 ) 1629 1633.
155 Plamboeck, A. H., North, M., Dawson, T. E., 2008. Conifer seedling survival under closed canopy and manzanita patches in the Sierra Nevada. Madro 55, 191 201. Quintero, J. A., Montoya, M. I., Sanchez, O. J., Giraldo, O. H., Cardonaa, C. A., 2008.Fuel ethanol production from sugarcane and corn :Comparative analysis for a Colombian case, Energy 33, 385 399. Raju, K. K., Raju, A. N., 1999. Biogeochemical investiga tions in south eastern Andhra Pradesh, The distribution of rare earths, Thorium and Uranium in plants and soils. Environ Geol. 38(10), 1102 06. Raju K. K., Rao, J., 1988. Red sander an accumulator of strontium. Geobios. 15, 207 209 Raju K. K., Nagaraju A., Sudheer, A. S. 1999. Biogeochemistry of Red Sanders from southeastern part of Andhra Pradesh, south India. Fresenius Envir. Bull. 8, 204 212 Ramakrishna, A., 1962. The Red Sanders and its future. Indian For 3, 202 205. Rao, K. B ., Giri, R., Kesavulu, M. M., Apparao, C., 2001. Effect of oral administration of bark extracts of Pterocarpus santalinus on blood glucose level in experimental animals. J. Enthopharmacol. 74, 69 74. Reddy, C. V. K., 1972. Red S anders and its history of Ut ilization, Indian For. 589 93. Raju, K. K., Na garaju, A., 1999. Geobotany of Red S anders ( Pterocarpus santalinus ) A case study from the southeastern portion of Andhra Pradesh. Environmental Geology. 37, 340 344. Reddy K. K., Srivasuki K. P., 1990. Vegetative propagation of Red Sanders ( Pterocarpus santalinus Linn.) Indian For 116, 536 540 SER, 2004. Society for Ecological Restoration International Science & Policy Working Group. 2004. The SER International Primer on Ecological Restoration. www.ser. org & Tucson: Society for Ecological Restoration International. http://www.ser.org/pdf/primer3.pdf (accessed: April 10th, 2009) Saaty, T. L., 1977. A scaling method for priorities in hierarchical struc ture J. Math. Psychol. 15, 234 281. Saaty, T. L., 1982 Decision Making for Leaders; The Analytical Hierarchy Process for Decisions in a Complex World, Belmont, CA, Pittsburgh: RWS Publications Saaty, T. L., 1993. The analytic hierarchy process: a n overview. Cent. Eu rop. J. Oper. Res. Econ. 2, 119 137.
156 Saaty, T. L., 2001. The seven pillars of the analytical hierarchy process. In: Saaty, T.L., Vargas, L.G. (Eds.), Models, Methods, Concepts & Applications of Analytical Hierarchy Process. Kluwer Academic Publishe r, Dordrecht, Netherlands, pp. 27 46. Saaty, T. L., Alexander, J., 1989. Conflict Resolution: The Analytic Hierarchy Process, New York: Praeger. Sagar, R., Raghubanshi, A. S., Singh, J. S, 2003. Tree species composition, dispersion and diversity along a disturbance gradient in a dry tropical forest region of India, For. Ecol. Manage. 186, 61 71. Snchez Coronado, M. E., Coates, R., Castro Colina, L., de Buen, A. G., Paez Valencia, J., Bar radas, V. L., Huante, P., Orozco Segovia, A., 2007. Improving seed germination and seedling growth of Omphalea oleifera (Euphorbiaceae) for restoration projects in tropical rain forests. For. Ecol. Manage. 243, 144 155. Sayer, J., Chokkalingam, U., Poulsen, J., 2004. The restoration of forest biodiversity and ecological values. Fo r. Ecol. Manage. 201, 3 11. Schippmann, U., 2001 Medicinal Plants Significant Trade Study CITES project S 109, BfN Skripten, 39. Schmer, M. R., Vogel, K. P., Mitchell, R. B ., Perrin, R. K., 2008. Net energy of cellulosic ethanol from switchgrass. P Natl. Acad. Sci. 105, 464 469. Schutz, A. E. N., Bond, W. J., Cramer, M. D., 2009. Juggling carbon: allocation patterns e prone savanna. Oecologia. 160, 235 246. Sheehan, J., Aden, A., Paustian, K., Killian, L., Brenner, J., Walsh, M., Nelson, R. 2004. Energy and environmental aspects of using corn stover fo r fuel ethanol. J. Ind. Ecol. 7, 117 146. Shibata, M., Nakashizuka, T., 1995. Seed and seedling dem ography of four co occurring Carpinus species in a temperate deciduous forest. Ecology. 76, 1099 1108. Short, W., Packey, D., Holt, T., 1995. A manual for the economic evaluation of energy efficiency and renewable energy technologies. National Renewable E nergy Laboratory TP 462 5173. 120 p. Shrestha R. K., Alavalapati, J. R. R., Kalmbacher, R. S., 2004. Exploring the potential for silvopasture adoption in south central Florida: an application of SWOT AHP method. Agr. Syst. 81 (3), 185 199. Simmons, J., Lovegrove, I., 2005. Bridging the conceptual divide: lessons from stakeholder analysis. J. Organ. Change Manag. 18 (5), 495 513.
157 Sukumaran, R. K., Pandey, A., 2009. India c ountry report. In: Eisentraut, A. (Ed.), Potential for sustainable production of 2nd generation biofuels, IEA 2009, p. 26. Sukumaran, R. K., Surender, V. J., Sindhu, R., Binod, P., Janu, K. U., Sajna, K. V., 2010. Lignocellulosic ethanol in India: Prospects, challenges and feedstock availability. Bioresource technol. 101, 4826 4833. Swans on, T., Bolton, P., Manning, A., 1993. Regulating en dangered s pecies, Econ. Policy 8(16) 183 205. Tchobanoglous, G., Burton, F. L., Stensel, H. D., 2003. Wastewater engineering, treatm ent and reuse. Metcalf & Eddy. Tata McGraw Hill, USA. Thornthwaite, C. W., 1931. The climates of North America: according to a new classification. Geogr. Rev. 21, 633 655. Turnbull, L. A., Crawley, M. J., Rees, M., 2000. Are plant populations seed limited? A review of seed sowing experiments. Oikos. 88, 225 238. Tyler, C. M ., 1995. Factors contributing to postfire seedling establishment in chaparral: direct and indirect effects of fire. J. Ecol. 83, 1009 1020. USDOE, 2010. Energy efficiency and Renewal energy Theoretical yield calculator, http://www1.eere.energy.gov/biomass/ethanol_yield_calculator.html (accessed : Jan 4th, 2010) Unger, P. W., Cassel, D. K., 1991. Tillage implement disturbance effects on soil properties related to soil and water conservation: a literature review. Soil Tillage Res.19, 363 382. Venkataraman, C., Habib, G., Kadamba, D., Shrivastava, M., Leon, J. F., Crouzille, B., Boucher, O., Streets D. G., 2006. Emissions from open biomass burning in India: Integrating the inventory approach with high resolution Moderate Resolution Imaging Spectroradiometer (MODIS) active fire and land cover data. Global Biogeochem. Cy., 20,GB2013. Vieira, D. L. M., Scariot, A., 2006. Principles of natural regeneration of tropical dry forests for restoration. Resto r Ecol. 14, 11 20. Weihrich, H., 1982 The TWOS matrix a tool for situation analysis. Long Range Planning 15, 54 66. Working Plan Kadapa, 2004. Workin g plan for the management of forests of Kadapa division, Andhra Pradesh Forest Department, Government of Andhra Pradesh. India.
158 Working Plan Rajampet, 2008. Working plan for the management of forests of Rajampet division Andhra Pradesh Forest Department, Government of Andhra Pradesh. India. Zelder, J., 2005. Ecological restoration: Guidance from Theory. San Francisco Estuary and Watershed Science, 3(2) 1 31.
159 BIOGRAPHICAL SKETCH Sidhanand (Sid) Kukrety was bo rn in Pune, India to Mrs. Shantidevi and Mr. M. R. Kukrety. His initial s chooling was spread in different parts of India but higher studies were completed in Dehradun where he received Bachelor of Science and Master of Science in C hemistry. H e taught C hemistry for a brief period before joining the Indian Forest Service in 1991. Sid is a professional forester. He received his forestry training and M aster s in f orestry with hono rs in 1993 from the Indira Gandhi National Forest Academy, Dehradun, the premier instit ution for training of Indian Forest Service officers. As a member of the Indian Forest Service allotted to Andhra Pradesh cadre he held posts of District Forest Officer in Nizamabad and Chittoor Districts of Andhra Pradesh India before receiving promotion as Conservator of Forests He served as Conservator of Forests, Tirupati Circle before moving to Gainesville for his doctorate. While working with the Andhra Pradesh Forest Department he became aware of the myriad problems associated with forest managem ent in Andhra Pradesh, India. The motivation and ideas for th is docto rate research were born out of his field work and experience in India His desire to contribute towards conservation of this s pecies was met with equal ly enthusiastic support from his advisor Dr. Janaki R. R. Alavalapati. He is keen to follow his work on community participation in natural resource conservation poverty alleviation, sustainable energy initiatives and conservation of forests. After completing his doctorate study, Sid plans to continue work ing for sustainable energy initiatives, forest restoration and conservation.