• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Executive summary
 Introduction
 Description of the study area
 Farming systems and cropping...
 Wheat crop management: Practices...
 Wheat problems: A summary and an...
 Reference
 Appendix 1. Participants in the...
 Appendix 2. Annual rainfall data...
 Appendix 3. Recommended cultivation...
 Appendix 4. Weeds and fodder trees...






Group Title: Wheat and rice in the hills : farming systems, production techniques and research issues for rice-wheat cropping patterns in the mid-hills of Nepal : report on an exploratory survey conducted in Kabhre District, 20-28 April, 1992
Title: Wheat and rice in the hills
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00080081/00001
 Material Information
Title: Wheat and rice in the hills farming systems, production techniques and research issues for rice-wheat cropping patterns in the mid-hills of Nepal : report on an exploratory survey conducted in Kabhre District, 20-28 April, 1992
Physical Description: x, 84 p. : ill. ; 28 cm.
Language: English
Creator: Harrington, L. W ( Larry W )
Publisher: Nepal Agricultural Research Council
International Maize and Wheat Improvement Centre
Place of Publication: Khumultar
s.l
Publication Date: 1992
 Subjects
Genre: non-fiction   ( marcgt )
Spatial Coverage: Nepal
 Notes
Statement of Responsibility: L.W. Harrington ... et al., editors.
 Record Information
Bibliographic ID: UF00080081
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 63530002

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Title Page
    Acknowledgement
        Page i
    Table of Contents
        Page ii
        Page iii
    List of Tables
        Page iv
    List of Figures
        Page v
        Page vi
    Executive summary
        Page vii
        Page viii
        Page ix
        Page x
    Introduction
        Introduction 1
        Page 2
        Page 3
        Page 4
    Description of the study area
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
    Farming systems and cropping patterns
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
    Wheat crop management: Practices and problems
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
    Wheat problems: A summary and an agenda for action
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
    Reference
        Page 79
        Page 80
    Appendix 1. Participants in the diagnostic survey held in Kabhre District, Nepal during the wheat season, 20-28 April, 1992
        Page 81
    Appendix 2. Annual rainfall data for three metrological stations in the study area averaged over 10 years (1981-1990)
        Page 82
    Appendix 3. Recommended cultivation practices for rice and wheat in the hills
        Page 83
    Appendix 4. Weeds and fodder trees in the mid-hills
        Page 84
Full Text










it
1:t1,` ji
0 t-

W m





WI-









I,, 40








-71 't


*49,
:tk "I

Jt "Vi



;Ily
ji

it








34
V"t











1A i,,













tit, ""k,














jlt
Ai t













Wheat and rice in the hills:


Farming systems, production techniques and research issues
for rice-wheat cropping patterns in the mid-hills of Nepal


Report on an exploratory survey conducted in
Kabhre District

20-28 April, 1992





L. W. Harrington
P. R. Hobbs
D. B. Tamang
C. Adhikari
B. K. Gyawali
G. Pradhan
B. K. Batsa
J. D. Ranjit
M. Ruckstuhl
Y. G. Khadka
M. L. Baidya

(Editors)











Nepal Agricultural Research Council, Khumultar (NARC)

International Maize and Wheat Improvement Centre (CIMMYT)


September, 1992







Acknowledgements


The editors would like to acknowledge the important support given to this survey by
the following institutions and people:

The Nepal Agricultural Research Council narcC), Khumultar, and in particular the
Executive Director, P. Amatya, and the Director of Technical Services, V. P. Sharma.

The agronomy division of NARC and the head of the program, P. P. Regmi and the
lead centre coordinator, D. B. Tamang.

The farming systems division and the encouragement and advice of M. P. Panth, the
head of the division.

The Land Resource Mapping Project in Kathmandu for their help in providing maps
and other useful information on the command area. In particular, we are grateful to
P. B. Shah, Senior Resource Manager, Integrated Survey Section, and the help of his
cartographers.

The heads of all the other research divisions of NARC who provided scientists of
various disciplines to collaborate on this project.

The funds provided by USAID (Nepal) under the ATSP project for research on rice-
wheat systems in Nepal.

The funds provided by the Asian Development Bank (ADB) through IRRI to support
research on rice-wheat systems in South Asia.

The CIMMYT core funds provided by the Kathmandu regional office to supplement
the resources contributed by USAID and ADB, and the help of the CIMMYT staff in
the Kathmandu regional office to type, compile and prepare various drafts.

A number of reviewers, especially Kelly Cassady, Michael Morris, Marlin van der
Veen, and Hari Gurung who provided helpful comments on early drafts of this paper.

The secretarial assistance provided by Khun Valairat Kuslasayanon, CIMMYT Asian
Regional Economics Program in Bangkok, for composing and printing the final ver-
sion of this publication.

The many scientists listed in Appendix 1 who formed the backbone of the survey but
who were not mentioned as paper co-editors. This report was made possible entirely
through the team effort of all participants.

The numerous farmers who shared their time and experience in answering the many
questions that emerged during the survey. The hospitality extended by farmers to the
survey team during overnight stays in the villages was also warmly appreciated.

Any remaining errors are the responsibility of the editors. Opinions expressed are not
necessarily those of NARC, CIMMYT, USAID, or ADB.







Contents

Acknowledgements i
Contents ii
Tables iv
Figures v
Executive summary vii

Introduction 1

Background to the survey 1
Methods 2

Description of the study area 5

Location 5
Climate 6
Topography, land types and soil types 11
Land types 11
Trends in land types 12
Soil types 14
Soil fertility 14
Irrigation infrastructure 16
Ethnicity and caste 18
Input and product markets 19

Farming systems and cropping patterns 20

Crops and cropping patterns 20
Khet 21
Bari 27
Grain production and other income sources 29
Grain and straw utilization: rice, wheat, maize and millet 30
Animals in the farming system 31
Bullocks and buffaloes 31
Fodder 35
Farm yard manure and compost 37
Forest resources, fuel use and soil erosion 38
Forest resources and fuel sources 39
Soil erosion 40
Food security and the food calendar 43
Labor use and labor scarcity 44
Trends in yields, input use and land quality 48

Wheat crop management: Practices and problems 49

Tillage and sowing 51
Practices 51
Problems: Late planting 53
Problems: Poor plant populations 53
Wheat varieties and wheat seed management 54







Soil fertility management and soil health 56
Practices 56
Problems: Soil nutrients and soil health 57
Irrigation and moisture management 60
Wheat diseases 60
Weeds and weed control 63
Insects and rodents 65
Harvesting, threshing and storage 66

Wheat problems: A summary and an agenda for action 69

Problem listing and ranking 69
Links among problems: A synthesis 71
An agenda for action 73
References 79

Appendix 1. Participants in the diagnostic survey held in
Kabhre District during the wheat season,
20-28 April, 1992 81
Appendix 2. Annual rainfall data for three meteorological
stations in the study area averaged over 10
years (1981-1990) 82

Appendix 3. Recommended cultivation practices for rice
and wheat in the mid-hills 83

Appendix 4. Weeds and fodder trees in the mid-hills 84







List of tables


Table 1 Villages visited during the
diagnostic survey 4

Table 2 Soil fertility status of bari
and khet lands in the
Dhulikhel Khola Watershed 15

Table 3 Area and production of different
crops in Mahadevsthan Sub-centre
Kabhre District 21

Table 4 Utilization of wheat, rice and
maize grain, by ethnic group 31

Table 5 Preliminary scoring of wheat problems 72

Table 6 Suggestions for action 75








List of figures


Fig. 1 Location of the Kabhre District
study area in Nepal 7

Fig. 2 Average monthly rainfall
1981-85 average for three sites 8

Fig. 3 Monthly min and max temperature
1981-85 average for two sites 10

Fig. 4 Percentage area sown to different
double-crop patterns (khet) 23

Fig. 5 Percentage area sown to different
triple-crop patterns (khet) 24

Fig. 6 Cropping calendars for major cropping
patterns khet (triple-crop systems at
lower elevations with assured irrigation) 25

Fig. 7 Cropping calendars for major cropping
patterns khet (double-crop systems with
less assured irrigation or poor
soil fertility) 26

Fig. 8 Cropping calendars for major cropping
patterns (bari) 28

Fig. 9 Grain utilization, by crop and
ethnic group 32

Fig. 10 Straw utilization by crop 33

Fig. 11 Fodder calendar 36

Fig. 12 Fuel calendar 41

Fig. 13 Food calendar, by income class 45

Fig. 14 Labor calendar for typical mid-hills farm 46

Fig. 15 Labor calendar for typical river valley farm 47

Fig. 16 Annual wheat yields, Kabhre District
1970-86 50

Fig. 17 Hypotheses on problems and causes:
Poor plant population reduces wheat yields 55








Fig. 18 Hypotheses on problems and causes:
Declining soil fertility reduces wheat
yields and threatens future productivity 59

Fig. 19 Hypotheses on problems and causes:
Rust reduces wheat yields 62

Fig. 20 Hypotheses on problems and causes:
Weeds reduce wheat yields, especially
on khet 64

Fig. 21 Hypotheses on problems and causes:
Storage losses reduce effective wheat
yields 68

Fig. 22 Summary of hypotheses on problems and
causes 74








Executive summary


Introduction and methods.

The mid-hill area of Kabhre District was selected as one of nine sites in South
Asia to participate in a National Agricultural Research System (NARS)-CIMMYT-
IRRI collaborative study focusing on the productivity and sustainability of rice-wheat
cropping systems. This report presents the results of a diagnostic survey conducted in
Kabhre District in late April, 1992, during the wheat season.

The survey employed methods of rapid rural appraisal and adaptive research
planning, modified to match the characteristics of the study area. (The survey team
walked more, stayed overnight more frequently with farm families, and used more
group interview techniques, compared to other rice-wheat diagnostic surveys.) Alto-
gether, team members interviewed over 100 farmers in six villages within a one-week
period of time. Survey participants included scientists (including two women) from
agronomy, soils, pathology, entomology and farming systems divisions of NARC,
Khumultar, plus extension staff from the Kabhre ADO office and the Department of
Agriculture, and three scientists from CIMMYT.

Survey objectives included the following:

- understand farmers' terms for land and soil types;
- understand links between crop production and animal husbandry activities; exam-
ine farmers' soil fertility management practices, including use of organic and inorganic
fertilizer and interactions with fuel needs;
- elicit farmers' perceptions on soil erosion, and how they maintain land quality;
- ascertain production practices for rice, wheat and other major crops;
- determine the utilization of major crops, as well as major sources of crop and non-
crop income;
- understand farmers' use of fuel, labor, food and crop by-products by constructing
yearly calendars and identifying periods of scarcity; and
- discuss system problems as perceived by farmers.

Survey data was used to develop specific hypotheses on problems and their
possible causes, where problems were defined to encompass issues of yield loss, input
use efficiency, efficiency of use of farm resources, and sustainability. A preliminary
ranking of problems was made by means of a weighted scoring model.

Farmers' circumstances in the study area.

Study area farmers distinguish two major land type categories -- hilly areas that
are sometimes terraced, and river valley bottoms. In hilly areas, specific land types
include bari (unbunded terraces), khet (terraced and bunded land where puddled rice
is grown at some point in the cropping system), and pakho or bhiralo (steep waste-
land). Khet is concentrated on the lower slopes. In river valley bottoms, specific land
types include several sub-categories of khet, including tar (having only partial irriga-
tion, and therefore unable to support double rice cropping) and phant (having as-
sured irrigation, enabling more intensive cropping).







Information on soil types and soil fertility was also obtained from farmers, and
compared to secondary data. Overall soil fertility appears notably poor, and soil
acidity and phosphorous availability is of special concern. It appears that soils on bari
lands are being degraded through erosion and leaching, but that some of the nutrients
that are lost are redeposited on khet lands at lower elevations.

About 20-30% of farm land is irrigated, with river diversions, streams (kholas)
and natural hillside springs being the major sources of irrigation water. Most irriga-
tion water is used on khet, but in a few places irrigated wheat was observed on bari
land. Management of irrigation systems varies among villages, with local manage-
ment by village inhabitants being the rule.

Links with input and product markets are reasonably good, and farmers sell
milk, potatoes, grain and other farm products in nearby markets. They purchase ferti-
lizer, usually on cash terms, but buy other inputs infrequently.

Farming systems and cropping patterns.

Major cropping patterns on khet include rice-wheat, rice-rice, rice-potato, rice-
rice-wheat and rice-rice-potato, with rice-wheat being predominant. Triple crop
patterns, and patterns including potatoes, are concentrated in river valley lands with
assured irrigation. Other minor crops grown on khet include toria, spring maize,
vegetables, blackgram and soybeans. The major cropping pattern on bari is maize
followed by a relay crop of millet. Other minor bari crops include mustard, cowpeas,
soybeans, barley and wheat.

Farmers reported that the bulk of their income stems from crop production,
especially rice, wheat and maize. Sales of fresh milk account for a surprisingly high
(median about 20%) of income, however. Uses of grain (rice, wheat, maize and mil-
let) include direct household consumption, cash sales, brewing of alcoholic beverages,
payment in kind for agricultural labor, and use as feed for livestock (especially lactat-
ing buffaloes and working bullocks). Wheat is important among lower-income farm-
ers for food security during the months of May and June. Some differences in grain
and straw utilization were found among ethnic groups and castes.

Although farmers raise many kinds of animals, buffaloes and bullocks are the
most important. In river valley villages, each household has at least one pair of bul-
locks. This is where the triple-cropping systems are concentrated and where rapid
turnaround between crops is important. In villages in the mid-hills, however, bullocks
were found in only 20% of farm households. Farmers in these villages have to share
(or rent) bullocks in order to prepare their fields. Bullocks are less common at higher
altitudes, where they are less useful for tillage on the small terraces found there.
Buffalo populations appear to have increased, with most farms owning at least one or
two animals. This has increased manure availability, while increasing fodder re-
quirements. Additional information was collected on fodder sources (numerous),
farm yard manure use (mostly composted), compost management practices, and fuel
use (dung cakes not used; community forests not a major source of fuelwood; most
fuelwood obtained from trees on the farm).

Erosion and leaching of nutrients from the soil appear to be a problem, partic-
ularly on bari lands. Farmers, however, interpret erosion primarily in terms of gully
formation, landslides and terrace face collapse. Ironically, farmers also intentionally
cut into the terrace face. Numerous reasons were given for this practice.


viii








Farmers indicated that rice yields are increasing (because of the introduction
of higher yielding varieties), that maize yields are declining (because of soil insects,
stem borers and reduced soil fertility), and that wheat yields are also going down
(increase in rust damage, substitution of potatoes for wheat in some of the more
productive areas in river valley bottoms). Secondary data confirm that wheat yields
have declined since 1982-83.

Wheat crop management: Practices and problems.

Land preparation for wheat is usually performed between early and mid-
November and is normally restricted to one or two bullock plowings, with one or more
plankings to break clods and cover the seed before sowing. Farmers customarily pre-
irrigate before tillage. When compost is applied, it is conveyed to the field and spread
by hand before tillage. Turnaround time between rice harvest and wheat planting
ranges between two and four weeks, although early wheat tillage and planting can
overlap with late rice harvest (usually on different farms). Wheat seed is broadcast by
hand, with seed rates around 70-90 kg/ha. When seed quality is poor, higher seed
rates are used (as high as 200 kg/ha). Most of the wheat is planted by the end of
November: little is planted late.

Direct observations by the survey team indicated that (especially in mid-hill
villages) wheat plant populations were very low, often fewer than 100 spikes per
square meter. A flow diagram of possible chains of causes was elicited with farmers.
Hypotheses on the causes of low plant populations included low seed rates, poor
germination, poor tillering, in turn due to other factors such as poor soil fertility,
moisture stress, soil acidity, etc.

Almost all farmers use the variety RR21 which is now susceptible to rust.
Many farmers expressed an interest in trying other varieties, but indicated that seed is
difficult to obtain. Direct observation by the survey team confirmed that stripe rust
was widespread, with virtually all observed fields affected. Farmers noted that when
infections were severe, yield losses could be as high as 50%, and grain would be shriv-
eled and bitter-tasting.

Virtually all farmers apply chemical fertilizer to wheat fields, typically on the
order of 60 kg/ha nitrogen and 30 kg/ha phosphate. Most fertilizer is purchased with
cash, not credit. Compost was only occasionally applied to khet wheat, especially in
mid-hill villages. (Compost use is more common on bari.) Hypotheses on problems
and their causes associated with soil health were discussed and diagrammed with
farmers. Possible problems include soil nutrient depletion, erosion, and build-up of
pests and diseases.

Other wheat problems and practices discussed with farmers include irrigation,
wheat diseases, weeds and weed control, insects and rodents, and harvesting, thresh-
ing and storage.

Wheat problems: A summary and an agenda for action.

In this survey, a problem was defined as: a factor associated with near-term
yield loss; a factor associated with inefficient use of external inputs or farmers' re-
sources (regardless of effect on yields); or a factor associated with reduced system
sustainability, or land degradation over the longer term. Wheat-related problems
identified in the survey include rust infection, rat damage, weed competition, erosion,








nutrient deficiencies and/ or toxicities, low plant populations, late planting, late
season moisture stress, and wheat storage losses.

These problems are not, however, equally important. In an effort to deter-
mine the relative importance of each, a weighted scoring model was constructed. In
this model, the annual regional productivity loss (ARPL) was estimated for each
problem. The ARPL is measured as a proportion of a good yield for a particular
location under "good" management and as estimated as the product of the frequency
of occurrence of a problem, the proportion of crop area affected by the problem, and
the productivity loss associated with the problem. When available information was
inadequate to make even a rough judgement on the value of a parameter, a question
mark was inserted.

Among the problems identified, rust and storage losses were judged as the
most important. Late planting (because of the small area affected) and drought stress
(because of its low frequency) were judged as less important. Weed competition was
judged to be of intermediate rank. Several problems could not be scored due to a
lack of information. These included problems of erosion and land quality which most
survey participants felt were supremely important. Additional diagnostic research is
needed to clarify the relative importance of each.

Finally, links among problems and their causes were discussed and dia-
grammed with farmers, and agenda for action developed. Actions include expanded
diagnosis (research aimed at improving the definition of problems thought to be
important); assessment of possible solutions (research aimed at assessing alternative
solutions for problems that are thought to be well understood and well defined);
extension (activities undertaken to promote a seemingly feasible solution to a well-
understood problem); and policy implications.








1 Introduction


1.1 Background to the survey


Rice and wheat are prominent sources of food and income for many farm

families in Nepal. The two crops are typically grown sequentially in a rice-wheat

cropping pattern. Although much of the rice-wheat area in Nepal is found in the

terai, the pattern is also conspicuous in the middle mountain physiographic region,

where it is concentrated on khet land (bunded, leveled terraces), where partial or full

irrigation is available.


This middle mountain region has a number of advantages. It enjoys a climate

fit for year-round agriculture, the hillsides are suitable for terracing, and water re-

sources from springs or streams have been captured to irrigate these terraces.

Consequently, the resource base has proved adequate to maintain a sizable human

population for a considerable period. Recent population increases, however, are

exerting pressures that threaten the capacity of this region to sustain suitable levels of

production.


A middle hill area in Kabhre District, in the Central Development Region of

Nepal, was selected as one of nine sites in South Asia to participate in a National

Agricultural Research System (NARS)-CIMMYT-IRRI collaborative study focusing

on the productivity and sustainability of rice-wheat cropping systems.1 In all nine

rice-wheat sites, research has commenced with diagnostic surveys conducted during

both rice and wheat seasons. These surveys have aimed to describe and analyze the


1. Two sites per country have been selected in Pakistan, Nepal and Bangladesh, with an additional
three sites in India, to represent different rice-wheat systems in South Asia.







rice-wheat systems found in each site, with particular attention to interactions be-

tween rice and wheat, and between the rice-wheat pattern and the rest of the farming

system. Survey analyses have resulted in the development of a research agenda for

each site, to guide the actions of a multi-disciplinary NARS team working in partner-

ship with (and with support from) CIMMYT and IRRI.1


This report presents the results of a diagnostic survey conducted in the Kabhre

District rice-wheat site in late April, 1992, during the wheat season. The site was

selected by scientists of the Nepal Agricultural Research Council narcC), Khumul-

tar, Nepal, and features an on-going farming systems program in the village of Nal-

dung. The collaborative program of rice-wheat research aims to build on this existing

FSR activity, expanding the area of study to include both hill terrace and lower alluvi-

al river valley areas.


1.2 Methods


The diagnostic survey employed methods of rapid rural appraisal and adaptive
research planning (Tripp and Woolley, 1989; Fujisaka, 1991), also known in Nepal as

samuhik bhraman (Chand and Gibbon, 1989). Survey methods were modified to

match the characteristics of the study area.2


It should be noted that the Naldung site is quite different from the other eight

collaborative rice-wheat research sites in South Asia (all of which are located on the
----------------

1. Results of earlier diagnostic surveys conducted as part of the collaborative rice-wheat project are
available for Bhairahawa, Nepal (Fujisaka and Harrington 1989); Pantnagar, India (Hobbs et al. 1991);
and Faizabad, India ( Hobbs et al. 1992).

2. In this paper, the terms "site", "study area", and "project area" are used interchangeably.







Indo-Gangetic Plain). In Kabhre District, the terrain is mountainous and access to

villages is mainly on foot, as there are few roads. Survey methods were adjusted

accordingly. Instead of frequent but brief visits to the field, punctuated with daily

discussions among members of survey teams (to synthesize information across teams),

longer but less frequent visits were used. Similarly, guidelines to help focus the semi-

structured farm interviews were not changed daily, but rather every two or three days.



Survey participants were divided into two groups for each visit. One group

approached the study area from Nagarkot (2150 masl) and walked down to the se-

lected villages. The other group approached the area from the lower Panchkhal side

(865 masl) and drove and walked up to the villages. Most of the farmers who operat-

ed farmlands in the lower river valley had located their houses in the mid-hill bari

areas above the lowland.



In a typical visit, survey participants arrived at a selected village in the mid-

afternoon and invited farmers to group discussions that often lasted the rest of the

day and into the evening. Participants arranged to stay in the village overnight1

Arrangements were then made for farmers to accompany survey participants the

following morning in making transects of the village, to observe and discuss agricul-

tural themes together while walking. Survey participants then returned to Khumultar

in the afternoon to discuss findings and develop guidelines for visits to the next set of

villages. This procedure was repeated twice, so that six villages were visited during

the course of the survey (Table 1).



-----------*---------

1. Often the overnight stays were arranged with no advance preparation -- the spontaneous hospitality
extended by farmers was warmly appreciated.











Table 1. Villages visited during the diagnostic survey

Village name Ethnic groups in the village


1st Day:
Group 1 Gairi Bisauni 90% Brahmin, 10% Tamang
Group 2A Raniban 95% Brahmin
Group 2B Chhap 95% Tamang

2nd Day:
Group 1 Nayagaun 65% Brahmin+Chhetri, 35% Tamang
Group 2A Chitre 100% Tamang
Group 28 Halede 85% Brahmin, 5% Chhetri,10% Damai
.................................................................





Farm interviews had numerous objectives, among them:

to understand the layout of the village, the proportion and locations of dif-

ferent land types and soil types, the crops and crop sequences grown on these land

types, irrigation sources and areas irrigated, and the areas where rice and wheat were

grown;

to understand the links between crop production and animal husbandry

activities in the village, including sources of livestock fodder;

to examine farmers' soil fertility management, including use of organic and

inorganic fertilizer sources and the interaction with fuel needs;

to elicit farmers' perceptions of erosion, how it has influenced land quality

and how farmers tackle this problem;

to ascertain production practices for rice, wheat and other major crops;

to determine the utilization of major crops, as well as major sources of crop

and non-crop income;

to understand farmers' use of fuel, labor, food and crop by-products by

constructing yearly calendars and identifying periods of scarcity;







to obtain prices of various inputs and outputs to allow construction of simple
cost and return budgets; and

to discuss problems of crop production as perceived by farmers.


Discussions were held at Khumultar after each village visit. The information

brought out in these discussions was immediately entered into a "personal information

management" database,1 for easy access and sorting prior to report-writing. Addi-

tional information was obtained from various secondary sources referenced in this

report. Supplementary information from secondary sources was particularly impor-

tant for this site, because the area covered by the survey was relatively small, whereas

the study area it aims to represent is notably diverse. Maps and statistical information

were obtained from the Agricultural Development Office, Kabhre District and the

Land Resource Mapping Project of Nepal.


Participants included scientists from agronomy, soils, pathology, entomology

and farming systems divisions of NARC, Khumultar, plus extension staff from the

Kabhre ADO office and the Department of Agriculture. In addition, three scientists

from CIMMYT participated in the survey (Appendix 1).


2 Description of the study area


2.1 Location


Kabhre District is located near the southeastern border of the Bagmati Zone

of the Central Development Region of Nepal. It lies between 85024' and 85059'E


1. Lotus Agenda, release 2.0.








and 27020' and 27045'N, to the east of Kathmandu, Bhaktapur and Lalitpur Districts.

Parts of it are accessible by road from Kathmandu.


For the most part, the district is located in the Mahabharat (a middle moun-

tain range), with a very small portion in the Siwalik hill range. The site selected for

the survey is located on the eastern slope of the ridge at Nagarkot (2150 masl) arid

extends down to the river valley area adjacent to the Indrawati River at Mahadev-

sthan (850 masl) (Fig. 1).


2.2 Climate


The climate of the survey area is variable, ranging from a humid warm tem-

perate climate at Nagarkot to a sub-tropical climate at Mahadevsthan. Average

annual rainfall is close to 1300 mm, based on data collected from three sites over 10

years1 (Fig. 2, and Appendix 2). Annual precipitation increases with elevation by 50

mm per 100 meters.


The climate of the study area is typical of Nepal, which is dominated by the

"wet" southeastern monsoon (May through September), during which 87% of the rain

falls each year (Fig. 2). The "dry" southwestern monsoon in the other months carries

little moisture, resulting in relatively dry weather from October to April. The south-

eastern monsoon begins in eastern Nepal in mid-May and arrives in the central re-

gions about a week later. The rains normally end in early to mid-September.



--------------------

1. Rainfall data were averaged from weather stations at Nagarkot, Panchkhal and Khumultar from
records available from 1981-90. These three sites are situated at the highest, middle and lowest eleva-
tions of the study area.









Gairi basauni


khola


;ipaghat


SINDHU PALCHOWK
DISTRICT


Naldung


Chitre


SunkosI River


Main Highway


Kay: ----= Dirt road ---....
---- VDC boundary
Khola (river)
O Villages
Main road


Kathmandu


Kabhre District


Figure 1. Location of the Kabhre District Study Area.















Fig. 2. Average monthly rainfall
1981-90 average for three sites
(Nagarkot, Panchkhal and Khumultar)
350
TOTAL = 1296 mm
300-

250

E 200-

rC 150
1500 -------- --- ---- --
100-

50 J --------_ -.. y S OtoD

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec








Rainfall variability in the study area is substantial. In 1991-92, for example,

only 30 mm of rain fell between mid-September and late April (the time of the sur-

vey). A five-year average for the same period is about 170 mm. This year's drought

has forced wheat to mature almost two weeks earlier than usual and has prevented

farmers from planting maize on their rainfed bari land. Variation also exists among

different locations within the district: one village may receive ample rainfall while

another experiences drought. Variability exists with regard to the timing of the start

of the rains, periods of drought, the occurrence of high intensity storms and the timing

of the end of the rains.


Rainfall erosivity for the Kathmandu Valley was measured by Fetzer and Jung

(1978), using a rainfall intensity recorder. The rainfall erosivity or R-value is one

component of the Universal Soil Loss Equation, and permits comparison of rainfall

erosiveness between Nepal and other countries. Over six years the R-value was 72,

indicating relatively low rainfall erosiveness. This makes possible reasonably stable

hillside cultivation.


Temperatures are strongly influenced by altitude -- for each 200 m gain in

elevation, the mean annual temperature drops by 10 C. The maximum and minimum
temperatures for the upper and lower elevations of the study area are shown in Fig. 3.

The mean monthly maximum temperatures are almost 15 degrees higher in the winter

and 10 degrees higher in the summer at the lower compared to the higher elevation.

Frost occurs at higher but not at lower elevations. Land aspect also influences tem-

perature. North-facing slopes remaining cooler than south-facing slopes -- mean

annual temperatures are about 3.50 C higher on south-facing slopes. The differences

in temperature brought about by elevation and land aspect has a strong effect on

cropping patterns in the area.
















Fig. 3. Monthly minimum and maximum
temperature, 1981-90
(average for two sites)


Panchkhal ma)
-U-M
Nagarkot max

Panchkhal min

Nagarkot min


P I I I I I I I I I I I I
Jan Feb.Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Note: Nagarkot (2150 m) and Panchkhal (865 m) represent the extremes in elevation for the study area.


(,


0)


CL.

E
12
~I


35


30-


25


20


15-


10


5-

Q'







2.3 Toography. land tpes and soil typ


2.3.1 Land types


Farmers in the study area distinguish two major land type categories -- hilly

areas that are sometimes terraced, and river valley bottoms. Within these are a

number of sub-categories. Land type classes often have strong effects on cropping

patterns and crop production practices. Farmers' terms for land type classes (and

explanations for each one) are as follows:


HILLY AREAS: The most important land types in hilly areas are bari and

khet. Relatively minor land types include pakho and bhiralo.


Bari lands feature unbunded (usually forward-facing) terraces on moderately

to steeply sloping (5-300 slope) hillsides, usually with well-drained soils more than 50

cm deep. Terracing is essential for controlling erosion when this land is used for

agriculture, but some erosion can occur even when terraces are used. Maize and

millet, but never puddled rice, are commonly grown on bari. Wheat can be grown on

bari land after maize or millet if irrigation is available.


Khet is terraced and bunded land, where puddled rice is grown at some point

in the cropping system.


Pakho and bhiralo are steeply sloping wastelands or erodable sloping lands

that are not normally cropped. Pakho is sometimes deforested, and then cultivated by

farmers for one or two years before being returned to fallow. Bhiralo is too steep for

cultivation and is usually left in natural vegetation or grassland. Jungle or ban is

natural forest that can occur on pakho or bhiralo. These two land types are often

selected for reforestation.










RIVER VALLEY BOTTOMS: Land in the river valley bottoms is called besi

and is usually categorized as lowland khet. Sub-categories of khet in the lower river

valleys include tar, hant and jimaha.


Lowland khet (like upland khet) is bunded and used for puddled rice culture.

The soils are deep and well to moderately well drained. Terracing or contouring and

provisions for control of runoff and drainage are used to control erosion. Depending

on irrigation availability this land type can be used for double- or triple-cropping,

usually rice-wheat or rice-rice-wheat.


Phant lands are lowland khet lands in river valley bottoms that enjoy year-

round irrigation. Water for irrigation comes from river diversions or from hill streams
or springs, making triple-cropping common. About 50-75% of the lowland khet lands

are phant. Tar lands feature partial irrigation, usually from hill springs, and do not

commonly support double rice cropping. About 25-50% of the lowland khet land is
tar. Tar is also a term for old relic river terraces. Jimaha is a minor land type (less

than 1%) that is marshy in nature. It is also found on upland khet terraces where

springwater emerges throughout the year.


2.3.2 Trends in land types


Wymann (1991) reports:

Over the last 17 years agricultural land, including bari and khet, has increased
significantly (total increase of 7%) at the expense of grazing land. Grasslands in
more marginal areas have been converted into agricultural land with the highest
increases occurring in areas with slopes above 29" and on south-facing aspects.
In addition to expansion of the agricultural landbase, cropping patterns were
intensified. Improved varieties of wheat and potatoes made it possible to grow
two and three crops per year.








This commentary was verified by the farmers during the survey. Farmers

noted that early rice (the first rice crop in a double rice cropping pattern) was not

grown 20 years ago; that there is less forest and grazing land than before; and that

they have to use more chemical fertilizer to maintain customary levels of yields.


It is not common for farmers to convert bari land into khet land, even when

new irrigation water becomes available. However, there appears to have been a small

increase in the proportion of khet to bari over time, mainly due housing construction

(which is understandably concentrated on bari land), driven by population increase.

The proportion of khet to bari land in the study area varies by village, and depends on

the availability of irrigation water and the closeness of the village to the lowland river

valley bottoms. Most of the lowland river valley khet land belongs to the farmers

living higher up on bari land on the ridges above the river valleys. This results in a

relatively high proportion of khet to bari in these areas.


Any review of land types is incomplete without some reference to land classes

used by the government for taxation. These are:


Khet Awall year round irrigation; the most productive.
Khet Doyam supplemental (five to six) monsoon irrigation.
Khet Sim less than two months supplemental irrigation.
Khet Chahar paddy production in only some years.
Pakho Awal no soil limitation.
Pakho Doyam minor soil limitations.
Pakho Sim major soil limitations.
Pakho Chahar severe soil limitations.



Farmers, however, were not accustomed to using these terms; many did not

even know what the terms meant.


1. This classification of land is based on productivity. Awal is the most productive, followed by
Doyam, Sim and Chahar. Chahar the least productive.







2.3.3 Soil types


The soils in the survey area are mostly derived from metamorphic rocks of
gneiss and quartzite. The soil textures are generally sandy loam to loam on bari land

and silty loam to loam on lower river terraces.


Red soils (lateritic), locally termed ratomato, are sticky when wet, hard when

dry, and have a high clay content. Red soils account for about 25% of the cultivated

area. Farmers further divide the red soils into lower yielding (rukhomato) and higher

yielding (malilomato) classes. Cropping patterns on the same land type did not differ

by soil type.


Sandy soils are locally termed balaute mato and account for about 75% of

cultivated area. These include kalo (blackish, high yielding), seto (whitish, low yield-

ing) and kailomato (light brown). Farmers also refer to a khumremato sandy soil that

has high infestation of white grub (Phyllophaga sp.).


2.3.4 Soil fertility


Secondary data sources (e.g., Wymann 1991) have described soil fertility
properties in a nearby, similar area, the Dhulikhel Khola Watershed. These sources

are used to approximate soil fertility conditions in the study area.


The bari soils are normally quite acidic, with low nitrogen and carbon content
(Table 2). Use of acidifying fertilizers, like ammonium sulfate and urea, should be

monitored to prevent further aggravation of the problem. Concentration of ex-

changeable bases appears to be negatively correlated with elevation. Two possible








reasons are proposed for this relationship: 1) increasing precipitation with elevation

leads to more surface soil erosion and higher leaching rates in the upper parts of the

watershed, and 2) accumulation of eroded and leached nutrients in the lower part of

the watershed.



The fertility status of khet is considerably better than bari (Table 2), although

exchangeable bases appear negatively correlated to elevation in a way similar to bari.

Presumably, the bases are transported with the irrigation water and accumulate in the

lower terraces.



In conclusion, erosion and leaching appear to be important factors in the

decline of soil fertility in higher altitude bari land, whereas khet land benefits from

these processes. The khet soils are less prone to declining fertility than bari soils

because of the input of fresh sediments from irrigation water. The possibility of

liming of bari land might be considered in future research.





Table 2. Soil fertility status of bari and khet
lands in the Dhulikhel Khola Watershed


pH % % ppm Ca Mg Na K EB CEC BS
Land type C N P -------------meq/0lg--------- %

Bari
4.3 0.73 0.07 22.3 2.2 0.7 0.02 0.14 3.2 8.6 38.2
Khet
2-4cm 4.8 0.68 0.06 21.1 3.5 0.4 0.20 0.18 4.2 8.5 51.6
5-8cm 4.6 0.54 0.04 18.1 2.9 0.4 0.10 0.13 3.4 7.1 49.1


Source: Wymann (1991)

EB=exchangeable bases, CEC=cation exchange capacity, BS=base saturation.
Average for 60 bari and 43 khet samples. Khet sampled at two depths to
distinguish between newly transported sediments and underlying soil.







Shah and Schreier (1991) have also discussed soil fertility issues in the water-

shed. They conclude that overall soil fertility is notably poor, and that soil acidity and

phosphorus availability is of particular concern. Less desirable soils had lower pH

and lower soil-P values. Shah and Schreier also showed that irrigated khet lands had

significantly better levels of certain nutrients than bari lands, possible due to enrich-

ment of nutrients from irrigation water.


Information on the red soils found in the watershed is also provided by Shah
and Schreier (1991). They find these soils very acidic, low in carbon content, very

prone to erosion and difficult to rehabilitate. The dominant clay mineral in these red

soils is kaolinite, which makes it difficult to maintain a good nutrient pool. Moreover,

at low pH and in the presence of extractable iron and aluminum, phosphorus is

converted into unavailable forms. Lacking basic cations and organic matter, the soil

structure deteriorates, and the soil becomes more difficult to work and hardens when

dry.


This might be an explanation for the hardening of soils observed by farmers

(chamro), especially in river bottom khet land where the use of compost is restricted

because of distance from the homestead. Farmers also reported that, in the mid-hills,

about 25% of land was becoming amilo (acidic) because of the use of acid-producing
fertilizers.


2.4 Irrigation infrastructure


Statistics from the Kabhre District ADO office show that 25% of the cultivated
land in the district is fully irrigated, 18% partially irrigated and 57% rainfed. Data

obtained from the Mahadevsthan Sub-Centre show that 31% of the cultivated land in







the 10 Village Development Committees (VDC) in their command area is irrigated.

However, these data do not distinguish between full and partial irrigation.


The main sources of irrigation are river diversions, streams (kholas) and

natural hillside springs. Many of the small streams in the hills are tapped by farmers

through farmer designed and built canal systems. Nayagaon, Chitre, Chhap and

Raniban villages have good examples of this type of system. In the Gairi Bisauni,

Devpur and Sallebas villages, the government has provided a grant to excavate a

canal from the Chitre Khola. Farmers are providing the labor, and the water will be

used to extend the irrigation in a tar area in the lowland areas of the villages.


Most irrigation water is used on khet land, but in a few villages (e.g., Nayag-

aon) farmers also use canal irrigation for bari wheat. In Nayagaon, farmers were

reluctant to convert bari to khet because of the work and cost involved, and the

exposure of poor sub-soil at the base of the terraces. When bari wheat is irrigated,

there is sufficient water only for a pre-irrigation and one or two additional irrigations.

There is not enough water to irrigate rice. Farmers can harvest three crops from this

bari land when it is irrigated (maize-millet-wheat pattern).


Springs are a less important source of water; the water they provide is ade-

quate to cover only around 5% or so of irrigated cropped area. The volume of water

from these sources is said to be declining, possibly because of deforestation and poor

recharge. Some springs are now used only for drinking water and similar household

needs.


Management of irrigation systems varies among villages. In one village, for

example, a committee decides on water distribution policy, which is then implement-

ed by a locally hired watchman. This watchman is paid 7-9 kg grain by each farm







household per 4 ropanis1 of land irrigated. He is responsible for proper distribution
of water and for minor system repairs. Village farmers provide their labor when

major annual repairs to the system are needed. All paying farmers get one irrigation

before any farmer gets a second one. Apart from this, however, farmers near the
irrigation source have priority in receiving water supplies. In other villages the supply
of water is handled more informally, with farmers called together as needed for canal

maintenance.


Management of irrigation systems by farmer groups appears to work reasona-
bly well. In one village, for example, farmers agreed to reduce conveyance losses by

irrigating at night those fields farther from the water source. When water supplies are

seasonally scarce, e.g., at the time of early rice establishment, farmers often agree to

limits on planted area. There is usually no water shortage at the time of establishing
the dry season wheat crop. However, in some drought years, water scarcity can lead

to some fields getting no irrigation water at all. Under these conditions, the number

of irrigations for winter crops is also usually reduced.


No ideas were brought forth in the survey regarding ways for improving the
efficiency of local irrigation systems.


2.5 Ethnicity and caste


Three main ethnic and/ or caste groupings were found in the villages visited --
Brahmins, Tamangs and Chetris. Minor groups were also found, including Kamis
(blacksmiths) and Damai (tailors). Some of the villages visited were mostly Brahmin

----1. 1 ha = 20 -i.
1. 1 ha = 20 ropani.







or Tamang while others had a mix of groups (Table 1).1 Tamangs appear to control

fewer resources, resulting in lower incomes. In river valley villages that are ethnically

mixed, Brahmins control virtually all of the irrigated khet land.


A few differences were found among ethnic groups:


1) Some farmers reported that Tamangs own more bullocks than Brahmins or

Chhetris, though other farmers disagreed.


2) Tamangs utilize a larger proportion of their grain in the preparation of

alcoholic beverages (raxi and chhyang). (Brahmins, for religious reasons, do not drink

alcohol.) Tamangs prepare these beverages throughout the year for festivals, wed-

dings and funeral ceremonies.


3) Tamangs seem more likely than Brahmins to leave the village for off-farm

employment. The village ties of the latter group are more effective in keeping young-

er people in the village.


4) Labor sharing is popular among the Tamang communities, whereas labor

hiring is more common among Brahmins.


2.6 Input and product markets


Fertilizer and seed are available at Agricultural Inputs Corporation (AIC)

Saha (cooperative) stores located in the area. However, farmers claim that they are
--------------------

1. Village profiles of other villages in the district confirm the proportions of ethnic groups. See
Socioeconomic Research Division (SERED) Reports 8,12, 14, 18 and 21.







reluctant to buy wheat seed because of the high price. Data from the ADO office

indicate that in 1988 improved wheat seed distribution in the district reached only 6

tons. Figures for rice and maize were 2 tons and 1 ton, respectively.


Fertilizer availability in the study area was reasonably good this year, although

farmers claimed that in previous years desired fertilizers were not available at the

correct time (Diammonium phosphate [DAP] was available when urea was needed).

Some villages have constructed access roads to allow tractors to haul fertilizer.

Otherwise fertilizer is carried in from either Nagarkot or Mahadevsthan manually.

Fertilizer is usually purchased on cash, not credit, terms.


Cereal grain is sold for cash in local and nearby major markets (e.g., Nagar-

kot). Potatoes are conveyed by porter to the main towns of Nagarkot and Mahadev-

sthan for sale.


Money, when borrowed, is obtained from local moneylenders. Usually,

however, farmers use their own cash resources when purchasing inputs. Some report

obtaining loans from moneylenders or traders, but only a very few from the Agricul-

tural Development Bank (ADB). The ADB does make credit available through its

Small Farmer Development Programs to farmer groups for milk production, irriga-

tion, input, bullock, buffalo, and goat purchases as well as for vegetable and potato

production.


3 Farming systems and cropping patterns


3.1 Crops and croppin patterns


As might be expected in an area so diverse, farmers in the study area cultivate








numerous crops (Table 3), organized into a complex array of cropping patterns that

feature up to three crops per year. Major cropping patterns found on khet land are

somewhat different than those on bari land.







Table 3: Area and production of different crops
in Mahadevsthan Sub-Centre, Kabhre District a/

% Area Prod. YLd
Crop area (ha) (t) (t/ha)
-.--.----..--..-.-------------------.-------------------..--
Rice 27 3363 13452 4.0
Wheat 11 1425 2850 2.0
Maize 27 3422 8555 2.5
Milet 25 3125 4375 1.4
Potato 10 1280 23040 18.0

a/ Total cultivated area for Mahadevsthan Sub-Centre = 8416 ha,
with a cropping intensity of 150%.







3.1.1 Khet



Crops grown by farmers on khet land include rice, wheat, potatoes, toria,

spring maize, vegetables, blackgram and sometimes soybeans. Blackgram and soy-

beans are often grown on alley ways, bunds and terrace faces. Among these crops,

rice and wheat are the most important. Potatoes are of secondary importance in

lower elevations.1 Principal cropping patterns include rice-wheat; rice-rice; rice-

fallow; rice-potato; rice-rice-wheat; and rice-rice-potato.



1. Potatoes were especially important in one river valley village linked to markets by a road. In vil-
lages without road access, potatoes are likely to be less predominant.







Patterns with two crops are more typical of khet lands in the mid-hills and tar

areas (khet lands with less assured irrigation) at lower elevations. Patterns with three

crops are concentrated at lower elevations in phant lands (khet lands featuring as-
sured irrigation and reasonably good soil fertility). The predominant double-cropping

pattern was found to be rice-wheat (about 70% of double-cropped area); similarly,

the principal triple-cropping pattern was found to be rice-rice-wheat (about 60% of

triple-cropped area) (Figs. 4 and 5). Triple-cropping patterns are extraordinarily

intensive, with little time between crops. Calendars for major khet cropping patterns

are shown in Figs. 6 and 7.


Previous studies conducted by SERED in similar areas have reached similar

conclusions with regard to khet cropping patterns. For example, rice-rice-wheat was

reported to be the main cropping pattern in "fully irrigated lowland areas" in several

VDCs of Sindupalchok District, covering 62% and 78% of the area found in this land

type. Similarly, rice-wheat, rice-mustard, or rice-fallow were identified as the major
patterns in "partially irrigated lowland areas" in the same VDCs (Shrestha, et al. 1989

-- also see SERED reports No. 12, 14 18, and 21).


In many rice-wheat areas of South Asia, differences in cropping intensity (for

systems featuring rice and wheat) are often associated with differences in rice variety

-- shorter duration varieties tend to be used in more intensive systems. This may not
be entirely true in the Naldung study area, however. Some farmers indicated that,

about 20 years ago, no early rice crop was grown. According to these farmers, triple-

cropping became possible only with the introduction of short-duration sathiya rice.

Sathiya rice matures about a week earlier than the commonly grown variety Taichung.

However, sathiva does not share some of the favorable characteristics that have made

Taichung a favorite, especially among Tamang farmers. These farmers note that














Fig. 4. Percentage area sown to
different double-crop patterns (khet)
(Concentrated in mid-hills, and lower elevations with partial irrigation .





Rice spring maize (5.0%)

Rice potato (25.0%)


Rice -wheat (70.0%)
.. ... :-. -" *., ..,. ^. ...^f^ *













Fig. 5. Percentage area sown to
different triple-crop patterns (khet)

(Concentrated in lower elevations with assured irrigation)


Rice-rice-potato (20.0%)



Rice-rice-other (15.0%)lll Rice-rice-heat
Other (5.0%)


















Fig. 6. Cropping calendars for major cropping patterns khet
Triple crop systems (lower elevations with assured irrigation)


Month a/


A/M M/J J/J J/A A/S S/O O/N N/D D/J

Bai Jes Asa Sra Bad Asw Kar Man Pou


Pattern


J/F F/M M/A

Mag Fal Cha


/Z -7


I I -Q


Rice nursery


Wheat


Potato


a/ English months shown in top line, Nepali months in second line. Note that the Nepali
months bisect English months (e.g., May extends from the middle of Baisak to the middle of Jesth).


RICE-RICE-WHEAT
Early rice
Normal rice
Wheat

RICE-RICE-POTATO
Early rice
Normal rice
Potato


Key:


Rice


r....~r "'r""""''77///~7//~'


















Fig. 7. Cropping calendars for major cropping patterns khet
Double crop systems (mid-hills or lower elevations with less
assured irrigation or poor soil fertility)

Month a/


Pattern


RICE-WHEAT
Rice
Wheat

RICE-POTATO
Rice
Potato


A/M M/J J/J J/A A/S S/O O/N N/D


Bai Jes Asa Sra Bad Asw Kar Man


D/J J/F F/M M/A


Pou Mag Fal Cha


Key: M M m


Rice


Rice nursery


Wheat


Potato


a/ English months shown in top line, then Nepali months in second line. Note that the Nepali
months bisect English months, (e.g., May extends from the middle of Baisak to the middle of Jesth).








Taichung makes excellent flattened rice1 and superior rice beer (chhyang), and that it

also "fills the stomach" better. Equally important, it commands a higher price in the

market. Given these advantages, farmers have been reluctant to reduce Taichung

area and, instead, have reduced turnaround time between rice crops to as little as one

or two days. As a consequence, a rice-rice-wheat cropping pattern featuring Taichung

rice for both rice crops is not unknown.


3.1.2 Bari



Roughly half of the land in the study area was found to be bari land. Crops

grown by study area farmers on bari land include maize, finger millet, mustard,

cowpeas, soybeans, barley and wheat. Of these, maize and millet are by far the most

important and are usually cultivated in a maize/ millet pattern, whereby millet is

transplanted and relayed into the standing maize crop.2 A third crop is occasionally

grown after the harvest of finger millet. This third crop is usually wheat or wheat

mixed with mustard (lower elevations), but may also be monocropped mustard

(middle elevations). The presence of this third crop typically depends on the avail-

ability of supplementary irrigation.3 In higher elevations, maize-fallow tends to

become predominant. Calendars for major bari cropping patterns are shown in Fig.

8.



1. Flattened rice is made by soaking the paddy in warm water overnight and then heating the rice in a
pan to dry it. It is then beaten and flattened. The husks are removed by winnowing.

2. The dominance of the maize/ millet cropping pattern (occasionally followed by a third crop) on bari
land types in mid-hill areas was also found by various SERED researchers as reported in SERED
Research Reports No. 8, 12, 14, 18 and 21.

3. Not all bari land is irrigated -- in fact, relatively little of it is. It should be noted, moreover, that bari
irrigation, while adequate for preirrigation or one or two irrigations for wheat, is not normally ade-
quate for rice cultivation.



















Fig. 8. Cropping calendars for major cropping patterns bari


Month a/


A/M M/J J/J J/A A/S S/O O/N N/D D/J

Bai Jes Asa Sra Bad Asw Kar Man Pou


Pattern


J/F F/M M/A


Mag Fal


Cha


zZZZZZZZZZZZ~


Key:
Millet


Millet nursery


a/ English months shown in top line, then Nepali months in second line. Note that the Nepali months bisect English
months, e.g., May extends from the middle of Baisak to the middle of Jesth. Note that the maize/ millet wheat
pattern is concentrated in lower elevations, while the maize-fallow pattern is concentrated in higher elevations.


MAIZE/ FINGER MILLET
Maize
Millet

MAIZE/ FINGER MILLET WHEAT
Maize
Millet
Wheat

MAIZE FALLOW
Maize


Wheat


Maize


i i


I _
~c~B~'~


~:,: :;j:~i~ii::::~:;I:;I:L:L:i.i:i:~;;r








The predominant maize variety mentioned by farmers was "Khumal Yellow";

the predominant millet varieties were local cultivars called "Dalle" and "Latte"1.


3.2 Grain production and other income sources



Sources of income2 for study area farmers are even more diversified than their

cropping patterns. Income sources mentioned by farmers include: sales and home

utilization3 of rice, wheat, maize, millet, barley and other crops; sale of potatoes; sale

of fresh milk; home manufacture of baskets, mats and brooms; and off-farm employ-

ment (especially during the dry season). Many farms reported raising goats and

chickens as well as cattle and buffalo.



Farmers reported that the bulk of their income came from crop production,

especially from rice, wheat and maize. One lowland village with relatively good road

infrastructure was found to specialize in potato production. Sales of fresh milk con-

sistently accounted for a surprisingly high 15-40% of income (median around 20%),

which reflects a comprehensive network of village-level milk collection points. This

milk is delivered to a cheese factory in Nagarkot (not far from the study area) and to

dairies in Kathmandu. Income earned off-farm seems limited for most families,

possibly due to the year-round labor requirements associated with milk production

and marketing. It should be noted, however, that some differences are apparent

between major ethnic groups: Brahmins tend to stay in the village, whereas Tamangs



1. Latte is the most common local variety with more closed heads. Dalle is of more recent origin and
has more open heads.

2. Income is broadly defined here to include production for home consumption as well as for cash
sale.

3. Including the use of grain for brewing and distilling. Details on crop utilization are presented in
the next section.








are more likely to seek seasonal employment in Kathmandu.


With their high dependence on crop production for income, farmers are vul-

nerable to yield variability associated with extreme weather. The drought prevailing

at the time of the survey was said to have reduced many farmers' incomes and to have

forced some farmers to sell livestock or borrow money to buy food.


3.3 Grain and straw utilization: Rice, wheat, maize and millet


Grain and straw of rice, wheat, maize and millet are put to a number of uses by

farm households in the study area. The uses of grain include direct household

consumption, cash sales, brewing of alcoholic beverages, payment in kind for agricul-

tural labor, and use as feed for livestock (especially lactating buffaloes and working

bullocks) (Table 4). More than half of the grain produced is consumed directly as

food for the household.


Secondary uses of grain depend somewhat on the crop: rice tends to be sold

for cash1, wheat to be given as payment in kind for agricultural labor, and maize to be

fed to livestock (Fig. 9). The use of grain for brewing alcoholic beverages is limited to

the Tamang ethnic group.


The uses of straw also depend on the crop from which it is derived. Rice straw

and millet straw are fed to livestock, as is most of the maize stover (a small proportion

of maize stover is used as fuel). Uses of wheat straw are more varied. Most wheat

straw is used for thatching roofs, with less than a third used as livestock feed (Fig 10).


1. Rice is often sold in the form of paddy because of low recovery rates when milled on the farm.












Table 4: Utilization of wheat, rice
and maize grain, by ethnic group
(%)

Use/ Ethnic Group Wheat Rice Maize

Direct consumption/ food
Brahmin 50-75 47-70 55
Tamang 25-50 25-90 25-75

Sale for cash
Brahmin 0-40 0-50 0
Tamang 10-25 35-50 10-20

Brewing
Brahmin 0 0 0
Tamang 35-40 10-25 10-40

Payment in kind/ Labor
Brahmin 10-25 0 20
Tamang 5 0 0

Animal feed
Brahmin 0 0 25
Tamang 0 0 25







Wheat straw is also extensively used for livestock bedding, after which it is composted

for three to four months with a mixture of cow dung and urine.



3.4 Animals in the farming system



3.4.1 Bullocks and buffaloes



Although farmers in the study area raise many kinds of animals, buffaloes and

bullocks appear to be the most important. Moreover, the number of buffaloes and

bullocks per household appears to vary considerably over villages, and over time















Fig. 9. Grain utilization
by crop and ethnic group

BRAHMIN |TAMANG

















Wheat Rice Maize Wheat Rice Maize
= Food f Cash sale E Brewing
Pay labor Mi Feed


100-
90-
80-
70-
60
50
40
30-
20
10-
0-














Fig. 10. Straw Utilization








-----


100
90-

80
70'
60
50


by Crop




- *


I I
Rice Maize Millet
Crop

hatch M Bedding Fuel


Wheat


I


I


I


I = Feed M.


I ~







within villages. Two to three buffaloes per household is common.


Bullocks are important in the farming system because they are used in land

preparation. In some parts of the study area, e.g., in the river valley, women reported

that each household had at least one pair of bullocks. This area is, of course, where

the triple-cropping systems are concentrated and where rapid turnaround between

crops is important.


In villages in the mid-hills, however, bullocks were found in only about 20%

of farm households. Farmers in these villages have to share (or rent) bullocks in

order to prepare their fields. In one Brahmin village, it was noted that the village

bullock population has declined from about 60 animals (10 years ago) to less than 20

today. In this case, labor scarcity was said to have played a major role.1 Bullocks are

apparently less common at higher altitudes, where they are less useful for tillage on

the small terraces found there.


In contrast, buffalo populations appear to have increased, with most farms

owning at least one or two animals. Richer farms were said to have up to four

buffaloes, with one or two lactating (duhuno). Interest in buffaloes is increasing

because of the establishment of the Nagarkot cheese factory and the consequent

development of a network of milk collection points.2 This growth in buffalo popula-

tions appears to have increased the amount of compost available to farmers, but at

the same time it has increased the amount of fodder farmers must obtain for their


1. Many younger members of farm households are absent from the farm, attending village schools,
and so were not available to tend the animals.

2. Note that buffalo milk has a higher butterfat content (and therefore commands a higher price) and
that buffaloes produce more milk per animal than cattle.







animals. In one Brahmin village, the number of buffaloes per family had decreased

from five (10 years ago) to less than three at present, because of labor scarcity.


3.4.2 Fodder


According to women respondents, livestock fodder includes rice straw, maize

and millet stover, wheat straw, cut grasses from own fields (including cuttings from

terrace facings) and cuttings from fodder trees. Women were said to spend one to

two hours per day cutting and carrying fodder.


No common grazing areas were found in any village; rather, most animals

are stall-fed or grazed on the farmers' own lands. Similarly, few women reported

collecting forest grasses from a community forest (community forest areas were

effectively available in only one village). The purchase of straw for fodder was said

to be uncommon, restricted to farmers with little khet land (and therefore little rice

straw) or farmers with an usually high ratio of large animals to land area.1


In addition to the dry and green fodder sources noted above, lactating buffa-

loes and working bullocks are sometimes fed concentrates, e.g., oilseed cake, rice

bran, maize flour or wheat flour mixed with water.


A calendar indicating the availability of different fodder sources during differ-

ent months of the year is shown in Fig. 11. A listing of grass species and trees cut for

fodder is given in Appendix 4.



1. Rice straw prices were quoted as Rs 100 per bhari of 50-60 kg, or about Rs 2 per kg. This is about
20-25% of the price of paddy, depending on grain quality and variety. Wheat straw prices were said to
be determined by the market for thatching material, not for livestock fodder.
















Fig. 11. Fodder calendar
Month a/
A/M M/J J/J J/A A/S S/0 O/N N/D D/J J/F F/M M/A


Fodder Source


Bai Jes Asa Sra Bad Asw Kar Man Pou Mag Fal Cha


Wheat straw
Grass from bund
Millet straw
Maize stalks
Fodder trees
Legume residues
Rice straw
Maize husks
Forest grasses
Maize thinnings


Greatest scarcity


-1S30 RauIe


OEMr


a/ English months are shown in the top line, Nepali months in the second line. Notet that the Nepali
months bisect English months (e.g., May extends from the middle of Baisak to the middle of Jesth).
Notes: Lactating buffaloes are also fed oilseed cakes, rice bran, maize or wheat flour, and other supplements.
Forest grasses are obtained from the community forest (one group only). Some farmers do not feed wheat
straw, claiming that it reduces milk yields. Rice straw is most important in the months of from Kartik through Asadh.










3.4.3 Farm yard manure and compost


Farm yard manure and compost availability was said by many farmers to be

increasing, in accord with the increase in buffalo populations.1 Compost applied to

farmers' fields as fertilizer is a combination of animal manure and urine, and plant

materials used as animal bedding. Bedding may include straw, stover or leaf litter

from the forest (rare). These are collected and piled up until needed.2 A composting

pit is usually not used, nor do farmers add water or inorganic fertilizer to the compost,

nor do they usually turn it to facilitate even decomposition. Not all animal manure is

used in composting: in the dry season, animals are sometimes grazed on bari fields.

Dung cakes are not made for use as fuel (as in other rice-wheat areas of South Asia,

located in the Indo-Gangetic Plains), so compost is not lost in this way.


Compost is usually applied to the field before or after the first plowing.

Compost application strategies are complex, however, and are affected by land type,

crop selection, distance from household to field, labor availability, the availability of

compost at a particular time, and other factors. Given labor scarcities, for example,

compost must be gradually conveyed to the field well ahead of plowing -- up to two

months ahead of plowing in the case of rainfed maize on bari lands. Farmers are

aware that nutrients may be lost if compost is left in the field for an extended time

without being incorporated, but they appear to accept these losses as inevitable, given

labor scarcity. Compost is typically carried to the field and spread by women.



1. This was not true of all villages. In a few cases, buffalo and bullock populations were said to be on
the decline.

2. Sometimes, this "collecting" and "heaping" involves nothing more than sweeping out the area where
animals are tethered in a low spot in front of the barn or house.








In deciding where to apply compost, farmers give preference to rice nurseries,

millet nurseries (much of the millet is transplanted), valuable cash crops (e.g., pota-

toes)1, and fields close to the household. In addition, there is an apparent preference

for applying compost to bari land as opposed to khet land, other things being equal

(probably because, unlike rice and wheat on khet land, maize and millet on bari land

receive little if any chemical fertilizer).2 Farmers appear to have no preference for

applying compost to particular soil types.


For those fields receiving compost, an application is usually made at least once

per year, sometimes once per crop cycle.3 Compost application rates were found to

range from about 10 t/ha for potatoes, to about 6 t/ha for khet wheat. Rice on khet

was said to receive even lower doses than wheat. (Compost applications to bari crops

were not determined.)



3.5 Forest resources, fuel use, and soil erosion



The collaborative research program on rice-wheat systems described in the

introduction of this paper aims to assess, in the various collaborative research loca-

tions, issues of sustainability as well as issues of near-term productivity. For the mid-

hills of Nepal, concerns about sustainability and land quality are inevitably linked witl

forest resource use and soil erosion. As a consequence, these themes were included


1. In applying compost to cash crops, farmers appear to take account of carry-over effects. Compost
applied to potatoes in a river valley village was said by farmers to be beneficial to the following spring
maize or early rice crops.

2. In several villages compost was used almost exclusively on bari lands. In another village, however,
compost use was concentrated on khet lands. It appears that in the former villages, households were
located closer to bari lands, whereas in the latter village, households were located closer to khet lands.

3. This may be sensitive to farm size, however. It was not ascertained whether farmers with larger
holdings were compelled to rotate their limited compost supplies around their several fields.








in the guide issues for the survey.1


3.5.1 Forest resources and fuel sources



Different villages expressed different experiences with respect to forest

management. In mid-hill villages, forest resources were often claimed to be "con-

stant", i.e., not subject to reduction in forest area or degradation of forest quality.2 In

some cases, farmers claimed that the village cared properly for the forest, to the

extent of employing local forest guards and restricting access to forest resources. In a

river valley village, however, farmers admitted that forest area had been substantially

reduced over the last 25 years as population increased and as land was cleared for

building, for agriculture, and for fuelwood. Even here, however, farmers claimed that

deforestation had been halted as better forest protection, community forest manage-

ment, and some replanting of forest areas had been introduced.



Deforestation is often linked with demand for fuelwood. In this survey,

however, wood from community forests was not found to be a major source of fuel. In

some villages, community forests were not a source of fuelwood -- because there were

no community forests. In the few villages with access to community forests, long

distances (five to six hours of walking) precluded any significant continued use of this

resource for fuel.





1. There is an abundance of literature on deforestation, forest management, land quality and soil
erosion in Nepal. This paper does not summarize that literature; rather, it offers farmers' perspectives
on forest and land quality issues elicited during the diagnostic survey. Serious students of land quality
issues for Nepal are urged to examine the broader literature.

2. In one village, however, virtually no common forest land was left to degrade.







According to women farmers, most firewood comes from trees on the farm

near the homestead, whether on khet or bari land. Some of this wood is cut and used

as needed. In addition, wood is cut green and stored for use during the rainy season.

Minor sources of fuel included twigs collected as available; wood cut from the

community forest; and maize stalks. Dung cakes were not used for fuel, unlike other

rice-wheat areas located in the Indo-Gangetic Plains of Bangladesh, India or Pakistan,

where dung cakes are an important fuel source.


Women reported spending one to two hours daily collecting firewood from

trees near the homestead for fuel (during those months of the year when firewood

was collected). About 30-35 kg of wood1 was said to be adequate for a family of six to

eight people for three days (equivalent to about 1.7 kg of wood per person per day).

Unfortunately, survey participants were not successful in ascertaining trends in

women's time spent in collecting fuelwood. If women are spending an increasing

amount of time at this task, their labor input into other tasks may suffer, with harmful

consequences for agricultural productivity.


A calendar showing major sources of fuel for the different months of the year

is shown in Fig. 12.


3.5.2 Soil erosion


A review of secondary data on soil characteristics and soil erosion was given in

section 2.3.4, above. In this section, farmers' opinions on soil erosion, and farmers'

actions to control erosion are described.



1. This is the amount of wood that fits conveniently in a doko or bamboo basket.


















Fig. 12. Fuel calendar


Month a/


Fuel Source


A/M M/J J/J

Bai Jes Asa


J/A A/S S/O O/N

Sra Bad Asw Kar


N/D D/J J/F F/M M/A

Man Pou Mag Fal Cha


a/ English months are shown in the top line, then Nepali months in the second line. Note that the Nepali months bisect
English months, e.g., May extends from the middle of Baisak to the middle of Jesth.
b/ Only one group of survey participants found farmers with access to a community forest. Distance to this forest
was 3-4 hours or 5-6 hours, according to the place visited. Other groups found that farmers didn't cut firewood
in the forest, but rather only on their own land.
c/ Firewood collected during the dry season is used during this period.


Wood from own land
Cutting in forest b/
Collecting twigs in forest
Maize stalks
Stored firewood c/








Farmers' and researchers' concepts of erosion appear to differ somewhat. For

farmers, erosion includes gully formation, landslides, and (most important) terrace

face collapse. Only a few farmers included the continuous washing of soil from ter-

race to terrace as erosion. In some villages, farmers informed survey participants that

there was no erosion problem in their farms, particularly on khet lands.



Terrace face collapse (TFC) was the erosion problem of most concern to

farmers. Some farmers noted that TFC was associated with land aspect: north- or

west-facing lands were said to be at greater risk, because there was less sunlight to dry

out terrace faces. Farmers also noted that TFC is concentrated in khet lands, because

of more frequent irrigation.1 When terraces are observed to be in danger of collapse

(usually during the rainy season), farmers temporarily strengthen them with stones,

leaves or branches, then carry out more permanent repairs with stones during the dry

season.



It is ironic, given concerns about TFC, that many farmers intentionally cut into

the terrace face. Several reasons were given for this practice.

terrace face soil is relatively fertile, and farmers can increase land productivi-

ty by spreading it on the surface of the terraces below;

terrace faces should be cut to even out the effects of any TFC that may have

occurred;

rodent problems can be reduced by cutting off the terrace face where their

burrows are located;



1. Some farmers blamed increased problems of terrace face collapse on the use of chemical fertilizer.
Survey participants were not successful in understanding the links between them. There was some
speculation, however, that chemical fertilizer use may induce soil pests, e.g., mole crickets, to concen-
trate in terrace bunds, thus weakening their structure.








terrace faces are weakened when roots from grasses grown in earlier seasons

begin to rot; these weakened areas should be cut to avoid later problems of more

general collapse;
soil cut from terrace faces can be used to deepen the soil on the terrace

below, e.g., to cover rocks that may be emerging;
cutting terrace faces contributes to disease, insect and weed control in crop

production.


The washing of soil from one terrace to another was not generally seen as a

problem. Loss of soil to the next terrace below was thought to be compensated by the

addition of soil from the next terrace above. (It should be noted that all villages visit-

ed were located half-way or more down the slope -- farmers at the top of the slope

might see this problem differently.) Those few farmers concerned about this washing

of soil indicated that it was concentrated on bari lands, especially where irrigation had

been introduced, where cropping patterns had recently been intensified and, as a

consequence, where the disturbance of the soil had increased. These farmers noted

that soil depth on bari terraces had declined and that rocks and stones were being

encountered close to the soil surface when plowing.


3.6 Food security and the food calendar


Given the heavy reliance of study area farms on crop production for both cash

income and food, some attention should be given to the food calendar and problems

of food security. During the survey, it was found that farms could be conveniently

grouped into two classes. Higher-income farms were defined as those where rice and

maize were readily available as food during the entire year. Lower-income farms

were defined as those compelled to switch from one source of food to another as the








year progressed. Food calendars for both groups are shown in Fig. 13. These calen-

dars clearly show the importance of wheat as a source of food for the "deficit period",

when food of any kind is relatively scarce. When crops fail, or crop yields are low,

farmers are sometimes compelled to borrow money or sell livestock to purchase food.


3.7 Labor use and labor scarcity


Given the several land types typically found on a farm, the intensity of land

use, the presence of livestock activities, and the need to gather fuelwood and fodder

on a daily basis, farm labor requirements are substantial. Peak labor demands typical-

ly occur during the months of Ashad and Shrawan (June-July and July-August), when

rice transplanting, rice weeding, maize weeding, and millet transplanting coincide. In

contrast, relatively little labor is required in the month of Falgun (February March).

Farmers indicated that there was no labor in-migration to help cope with peak season

labor requirements. (Some farmers, especially Tamangs, use labor sharing arrange-

ments.) In addition, labor out-migration during the slack season was less than expect-

ed, possibly due to the importance of livestock activities. Labor calendars were

developed with farmers for two typical situations:
rice-wheat on khet lands, plus maize/ millet on bari lands, representative of

farms in the mid-hills (Fig. 14);
rice-rice-potato on khet lands, plus maize/ millet on bari lands, representa-

tive of farms in the river valley (Fig. 15).


Labor scarcities affect a number of farm activities. As noted in section 3.4.3,

compost must be conveyed well ahead of plowing time to the fields where it is to be

applied; otherwise, there will be insufficient time for transporting and incorporating

it. This appears to lead to a decline in the effectiveness of compost applications. In

addition, some farmers blame late planting of wheat on scarcity of labor for harvest-

ing and threshing of the preceding rice crop. Farmers insist on finishing the harvest
















Fig. 13. Food calendar, by income class
Month a/
A/M M/J J/J J/A A/S S/O O/N N/D D/J J/F F/M M/A


Food staple


Bai Jes Asa Sra Bad Asw Kar Man Pou Mag Fal Cha


Wheat b/
high income
low income
Rice
high income
low income
Maize
high income
low income
Millet
high income
low income


Deficit period c/

Key: High Income


Low Income


Deficit Period


High vs. low income in terms of whether a farm family can afford to consume rice and maize all year.
a/ English months shown in top line, then Nepali months in second line. Nepali months bisect English months.
b/ Wheat eaten occasionally, to vary the diet.
c/ When crops fail, low-income farmers sometimes have to borrow money or sell livestock to purchase food.


~1\\\\\\\\\\I ROSS I q I 11's, I\\

















Fig. 14. Labor calendar for typical mid-hills farm (rice-wheat on khet and maize/ millet on


Crop/ task


Month a/


A/M M/J J/J J/A
Bai Jes Asa Sra


A/S S/O O/N N/D D/J
Bad Asw Kar Man Pou


M/A
Cha


Rice nursery
Rice transplant
Rice weeding
Rice harvest b/
Wheat tillage c/
Wheat planting c/
Wheat harvest
Maize tillage d/
Maize planting d/
Maize weeding e/
Maize harvest
Millet transplanting
Cut firewood
Carry compost


Peak labor demand f/ w////

a/ English months shown in top line, then Nepali months in second line. Nepali months bisect English months.
b/ Begins earliest in lower altitudes.
c/Exchange labor used.
d/ Extends later in drought years.
e/ Weeding (once or twice) and earthing up.
f/ Farmers' perceptions, supported by evidence of seasonal increases in wages.


J/F F/M
Mag Fal


1811100
111811181


SENOR
















Fig. 15. Labor calendar for typical river valley farm
(rice-rice-potato on khet, and maize/ millet on bari)


Crop/ task


Month a/


A/M M/J J/J J/A A/S S/O O/N N/D D/J
Bai Jes Asa Sra Bad Asw Kar Man Pou


Early rice nursery
Early rice transplanting
Early rice weeding
Early rice harvest
Rice nursery
Rice transplanting
Rice weeding
Rice harvest
Potato tillage, planting
Potato harvest
Maize tillage, planting
Maize weeding b/
Maize harvest
Millet transplanting
Cut firewood
Carry compost


Peak labor demand c/ v//////

a/ English months shown in top line, then Nepali months in second line. Nepali months bisect English months.
b/ Weeding (once or twice) and earthing up.
c/ Farmers' perceptions, supported by evidence of seasonal increases in wages.


J/F
Mag


F/M M/A
Fal Cha


ma
anal
MOE=BB


m1sim
Emm








ing and threshing of rice before commencing wheat activities.1 The possibility that

labor shortages may delay rice establishment, and thereby delay rice harvesting and

wheat establishment, was not assessed.


3.8 Trends in yields, input use and land quality


Unravelling trends in crop yields, input productivity, and land quality over time

is challenging, because numerous factors have unfolded simultaneously. For example,

over the last 20 years or so, the following processes have been at work:

some bari land has been converted to khet when major new sources of irriga-

tion have been developed;

bari land has been taken out of production to make way for housing;

improved varieties and chemical fertilizer have been introduced;

availability of chemical fertilizer has improved because of farmer-built roads

that allow the passage of small tractors with trailers;

triple cropping has become more common, spurred by the introduction of

sathiya rice varieties, and by the development of new irrigation facilities, particularly

canal water diverted from streams (some farmers blame this crop intensification for a

more rapid buildup of pests and diseases);

potato cultivation has replaced wheat in some river valley villages;

fresh milk sales have become more important as a source of income;

soil erosion (terrace face collapse, washing of soil from one terrace to anoth-

er) has gathered pace, especially on bari lands; and

forests (especially in river valley villages) have been degraded.



1. This is probably sound judgement, considering losses to rats and other pests that might occur if rice
were stacked and threshing delayed until after wheat sowing was finished.







Some of these processes reinforce each other (e.g., use of bari land for housing

and the rapid erosion of bari fields both suggest that total production from bari land

will decline over time). Other processes counteract each other (e.g., erosion on bari

has accelerated, but much of what is lost in the way of soil or plant nutrients from bari

fields is deposited in khet fields.) A full understanding of the interactions among

these trends might require a sophisticated dynamical model, together with a very

considerable quantitative data base. In the absence of these, however, farmers'

opinions on trends may provide useful insights. Some farmers' perceptions on trends

are:


Rice yields are increasing because of the introduction of higher yielding

varieties (but may be falling where pest buildup is severe);

Maize yields are declining, because of ever-increasing problems of soil in-

sects and stem borers, combined with soil nutrient depletion (more compost and ferti-

lizer are required to produce the same yields as before);

Wheat yields are decreasing, because of an increase in rust damage on the

predominant variety (RR21) and because some of the best wheat lands have been

converted to potato cultivation. Time-series data on wheat yields for Kabhre District

from 1970-71 to 1986-87 indicate that wheat yields rose until 1983-84, but since that

date they have been falling (Fig.16).


4 Wheat Crop Management: Practices and Problems


A major objective of the diagnostic survey was to examine wheat production

practices, identify associated problems,1 indicate some of the causes of these


1. In this report, as in other rice-wheat diagostic survey reports, a "problem" is taken to mean a factor
associated with near-term yield loss; a factor associated with inefficient use of external inputs or
farmers' resources (regardless of effect on yields); or a factor associated with reduced system sustain-
ability or land degradation over the longer term.


















Fig. 16. Annual Wheat Yields
Kabhre District, 1970-86


1.8
1-.8---------------------------------------_---


1.6 -

1.4 --.-- ------ ---- --- -



1 ...... ........


0.6

0.4-

0.2-

0-


1970 1972 1974 1976 1978 1980 1982 1984 1986


Source: Agricultural Plan for Kabhre District, 1970-1986.


.. .......-


m


]







problems (particularly causes linked to other activities in the farming system), and

highlight possible avenues for action, including research, extension and implications

for policy. In this section, wheat production in the study area is examined practice by

practice, from land preparation and sowing through harvest and post-harvest activi-

ties. Discussion will focus on wheat in a rice-wheat or rice-rice-wheat pattern on khet

land. Where wheat is grown after maize on irrigated bari, differences in practices or

problems will be highlighted.


4.1 Tillage and sowing


4.1.1 Practices


Land preparation for wheat is usually performed between early and mid-

November and is normally restricted to one or two bullock plowings, with one or more

plankings to break clods and to cover the seed after sowing. There is some variability

in the number of plowings and plankings given, but tillage does not appear to be

excessive.1 Farmers with reliable sources of irrigation water customarily pre-irrigate

before tillage, as the soil is hard and dry after rice culture. Raised beds are not made;

rather, fields remain flat after tillage. When compost is applied, it is conveyed to the

field before tillage and spread by hand just before tillage so that it may be incorporat-

ed into the soil during plowing.2


In river valley villages, practically every household owns a pair of bullocks. In

some villages in the mid-hills, however, only around one fifth of households maintain


1. In contrast, in rice-wheat areas located on the Indo-Gangetic Plains in India or Pakistan, more than
half a dozen tractor plowings are not uncommon.

2. Note, however, that compost use is more common on bari land than khet land. Some khet wheat
fields receive little or no compost.







even a single bullock, and renting or sharing bullocks is common. In these villages, a

shortage of bullocks was said by farmers to result in delayed land preparation for

wheat, occasionally leading to late planting.1


Some farmers reported experimenting with zero-tillage techniques, whereby

soaked wheat seed is broadcast on moist untilled soil. These farmers affirmed that

this method produces a good wheat crop, but that land preparation for the following

rice crop becomes more arduous.


Turnaround time between rice harvest and wheat planting ranges between two
and four weeks, although early wheat tillage and planting can overlap somewhat with

late rice harvest (usually on different farms) (Figs. 14 and 15). Turnaround time may

be shorter in the intensive triple-cropping systems found in the river valleys.


Wheat seed is broadcast by hand. No line sowing or (understandably) seed

drills were found. Estimates of seed rates ranged from 60-120 kg/ha (calculated from

local units of athi per ropani), with one or two estimates as high as 200 kg/ha.

Higher seed rates were said to be associated with poor quality seed, especially seed

infested with storage pests. Median seed rates, however, were said to be surprisingly

low -- around 70-90 kg/ha. Farmers obtain seed either from their own stored stock, or

from neighbors. Apparently, seed exchange among neighbors is an established and

common practice. Access to seed of new varieties, however, appears poor.


Two problems were found that are associated with tillage and sowing of wheat.

First, some of the wheat is planted late. Second, poor plant populations appear to


1. In these villages, the bullock population is not only low but also appears to be declining, as the
number of dairy buffaloes increase. See section 3.4.







reduce wheat yields.


4.1.2 Problems: Late planting


Farmers' definitions of "late planting" varied somewhat among villages. In

some villages, planting was considered to be late if it fell after the second week of

Mangsir (end of November). In other villages, "late" was interpreted as planting

during Pous (middle of December to middle of January). In an effort to standardize

definitions, the survey team asked farmers about the proportion of wheat fields plant-

ed after the second week of Mangsir. Responses ranged from "none" (in a Brahmin

hillside village), to about 20% (in a Tamang hillside village). In a river valley village,

only 20% of farmers plant by this particular date, but most of the wheat is usually

planted by the end of Mangsir (mid-December). Late planting is not widespread --

nor, however, is it entirely unknown.


Reasons given for late planting include scarcity of bullocks; scarcity of labor to

harvest and thresh rice; and the need to wait for irrigation water to become available.

In the scoring of individual problems (Table 5), however, it became clear that late

planting is less important than some other problems affecting the wheat crop.

Consequently, it is not discussed further in this section.


4.1.3 Problems: poor plant populations


Although much of the wheat in the study area had been harvested by the time

the survey was conducted, it was still possible to observe wheat plant populations in

some fields. In mid-hill villages, wheat plant populations often looked very low (less

than 100 spikes/m2). Examination of wheat stubble in some river valley wheat fields

confirmed a generalized impression of low plant populations.







Low populations can be agronomically linked to one or both of two factors:

low initial germination and poor tillering. Poor tillering might be caused by late plant-

ing (but see section 4.1.2, above), acid soils or nutrient deficiencies (see section 2) or

drought stress. Low germination might conceivably be traced to low seed rates or

poor seed germination, in turn due to a number of factors. Hypotheses on causes of

poor tillering, poor initial stands, and poor germination are spelled out in greater

detail in Fig. 17.


It is important to note that the causes of poor plant populations described

above are only hypotheses. No evidence was gathered on germination rates or initial

plant stands (although information gathered on seed storage methods suggests that

seed quality might well play a role -- see section 4.2). Actions to address the problem

of poor plant populations might encompass a wide range of issues: from amelioration

of acid soils to provision of quality seed. Further diagnosis on this theme is needed.

Why, for example, do farmers use low seed rates? Why is tillering apparently so

poor? These questions need to be answered before setting detailed research direc-

tions.


4.2 Wheat varieties and wheat seed management


Farmers use a number of wheat varieties, including RR21, Lerma Rojo 64,

Nepal 297, Triveni, Vaskar Annapurna 1 and Annapurna 2. Most farmers, however

-- about 90-95% of them -- were said to be using RR21, which is now susceptible to

rust. Many farmers expressed interest in trying other varieties, but indicated that seed

of new varieties was difficult to obtain.1 It would appear that turnover of new varie-


1. Farmers living in villages where the FSR program has been most active appeared to have the high-
est rates of adoption of new wheat varieties. In these villages, of course, seed availability is less of a
problem.












Fig. 17. Hypotheses on problems and
Poor plant population reduces wheat


causes:
yields


Key: rectangles problems; ovals causes; hexagons "primary causes"







ties, while slow in the study area, does occur. Farmers in one village noted that 30

years ago, they were using local Nepali varieties; about 20 years ago, they adopted

Lerma Rojo; then about 15 years ago, RR21 came along. Now Nepal 297, Annapur-

na 1 and other rust-resistant varieties are being introduced. Stripe rust (Puccinia strii-

formis) infections were widespread, with virtually all observed fields seen to be affect-

ed. Farmers noted that when infections were severe, yield losses could be as high as

50%, and grain would be shriveled and bitter-tasting.1


It was noted in the previous section (4.1.3) that wheat seed storage may be

deficient, leading to poor seed viability and low germination rates. During the survey,

farmers provided information on farm-level seed storage. Seed is stored essentially in

the same way as wheat grain, i.e., in gunny sacks or bamboo baskets, not in earthen

jars. Common storage containers include the bhakari or bamboo mat, and the sukul

or rice straw mat. Seed (or grain) is placed in these mats and rolled into a cylinder.

Some farmers reported 20-25% loss of seed from insect and rodent damage when it is

stored in this fashion. Seed and grain are often treated with insecticide, e.g., BHC,

but without any great reduction in pest populations. For further information on

problems associated with grain and seed storage techniques, the causes of these prob-

lems, and some possible actions, see section 4.8.


4.3 Soil fertility management and soil health


4.3.1 Practices


Almost all farmers apply chemical fertilizer to wheat fields. Doses vary

somewhat over villages, but survey participants estimated that farmers use about 60

1. The survey was conducted during a wheat season in which a drought occurred. As a consequence,
rust caused fewer problems than usual. Farmers noted, however, that problems of rust are becoming
continually worse over time. (See section 4.5.)








kg/ha nitrogen (split fairly evenly between a basal application and a topdressing) and

about 30 kg/ha phosphate. Nitrogen doses range from less than 40 kg/ha to around

100 kg/ha. Doses are higher in river valley villages with relatively good access to

markets.


Basal application features the use of 20-20-0, whereas topdressing relies on

urea, or occasionally ammonium sulfate. Basal fertilizer is broadcast before seeding

at the time of the second plowing, whereas the topdressing is broadcast before the

first irrigation about one month after sowing.


Availability of fertilizer is apparently uncertain. Farmers reported that ferti-

lizer was readily available this year, but that fertilizer supply has been unreliable in

previous years. Surprisingly, farmers claimed that most fertilizer was purchased with

cash, not credit. Fertilizer is transported to the village on trailers pulled by small

tractors, on roads constructed by farmers (river valley villages), or by porters over

footpaths from Nagarkot (midhill villages). Farmers reported paying Rs 5.2/kg of

urea and Rs 6.4/kg of 20-20-0, including Rs 0.4-0.7/kg transport cost for porters.

depending upon the distance carried.


In addition to chemical fertilizer, compost was sometimes (but not always)

applied to khet wheat, especially in mid-hill villages. Typical rates for compost appli-

cation were around 6 t/ha. It is not known if wheat grown on irrigated bari receives

lower or higher doses of compost (see section 3.4.3 for additional information on

compost).


4.3.2 Problems: Soil nutrients and soil health


Farmers broached specific concerns about the effects of chemical fertilizer on








soil health. Farmers in a river valley village, for example, claimed that soils become

harder (chamro) if chemical fertilizer is used without the addition of compost.1 Other

farmers in the mid-hills claimed that continuous use of chemical fertilizer caused soils

to become more acidic (amilo). As noted above (section 3.5.2), some farmers blame

terrace face collapse on the use of chemical fertilizer, though the links between the

two are not clear. Finally, there was a general feeling that increased levels of fertilizer

were needed just to maintain earlier yield levels.2


The survey team came to share numerous misgivings about soil health, includ-

ing soil nutrient depletion. Farmers' fertilizer and compost management practices

may be inadequate to deal with the pressures imposed on the system by increased

cropping intensity and soil erosion. Hypotheses on trends in soil health are summa-

rized in Fig. 18.


A sense of perspective is needed, however. Cropping intensification has taken

place -- but mostly in river valley villages where erosion most likely works in favor of

farmers (i.e., they benefit more from nutrients brought down from above than they

lose by the erosion of soil to lower terraces). Compost use may not have increased as

much as desirable -- but it does not appear to be declining (as it is in other rice-wheat

areas of South Asia). Soil erosion on bari lands is of special concern -- but only a

small proportion of wheat area (and virtually no rice-wheat area) is found on bari.


Considerable diagnostic work remains to be done to sort out the real threats to

soil health in areas devoted to the rice-wheat cropping pattern in the mid-hills. A


1. This concern was raised in an area where khet wheat fields are located far from the farm household,
making compost conveyance costly and difficult.

2. This decline in fertilizer productivity does not necessarily imply growing problems in soil health.
For example, the marginal productivity of fertilizer application is likely to wane over time as rust
damage increases.













Fig. 18. Hypotheses on problems and causes:
Declining soil health reduces wheat yields
and threatens future productivity


INCREASED
IRRIGATION OF CROPPING INTENSITY
BARI LRND FOSTERS FOSTERS EROSION AS
EROSION (L?) LAND IS DISTURBED
MORE FREQUENTLY


FARMERS DON'T
USE SOME EROSION
- CONTROL MEASURES
(GRASSY STRIPS,
TREES)


FERTILIZER USE INADEQUATE
INADEQUATE TO CONTROL MEASURES
S COMPENSATE FOR FOR PESTS (WEEDS,
NUTRIENTS EXTRACTED INSECTS, DISEASE)
BY CROPS ('?) (? ?)

Key: Rectangles problems; ovals causes; hexagons "primary" causes







large number of research themes might be pursued: agroforestry; management of

acid soils; improving the effectiveness of farmers' compost resources; testing of alter-

native cropping patterns; integrated management of organic and inorganic fertilizers

to maintain soil health; etc. Additional diagnosis is necessary to ascertain which of

these avenues are most likely to be productive.


4.4 Irrigation and moisture management


Pre-irrigation is essential for wheat establishment because the hard, dry soils

left after rice harvest need to be softened to allow tillage, and moisture is needed for

germination. After establishment, wheat receives one or two additional irrigations

depending on water availability. The first irrigation is performed 20-30 days after

emergence, and the second (when given) is timed at about two to three weeks before

maturity, during flowering. In the current drought year many farmers had insufficient

water to give the second irrigation, resulting in moisture stress during grain filling,

forced maturity (up to two weeks earlier maturity) and shriveled seed.1


No ideas were brought forth regarding how irrigation system efficiency might

be improved, except through development of new sources of irrigation (only possible

by tapping river water) or through reduced production of early spring rice.


4.5 Wheat diseases


The three wheat diseases mentioned most by farmers were rust (Puccinia strii-

formis), loose smut (Ustilago tritici) and foliar blight. The farmers' term for blight is


1. One might speculate, of course, that even two irrigations are inadequate for wheat grown on sandy
or sandy loam soils during the dry season.







dadhuwa, a generic term for yellowing, necrosis, chlorosis and leaf firing that can be

caused by many factors (nutrient deficiency or toxicity, drought, excess water, soil

insects and foliar diseases, etc.). Because of the early maturity of the crop during the

survey year, participants were not able to observe this symptom in the field. Visits are

needed at the proper time to get a better perspective on this problem.


Loose smut was observed to affect as many as 5% of plants in some fields.

This is a seed-borne disease perpetuated when farmers continually use their own seed

and/or not treat seed with fungicides before planting. The yield loss for this disease is

in direct proportion to the percentage field incidence, since the entire wheat spike is

lost when the plant is infected. The easiest way to control the disease is to discard and

replace infected seed with new healthy seed. However, given the problems farmers

have in replacing seed, this method of control may not be feasible without a strong

extension effort and village-level seed production. Another way to reduce the prob-

lem is to rogue out infected plants before they have a chance to shed their spores.


Rust, especially stripe rust (panhele, sindhure) is the disease most commonly

mentioned by farmers. This is not surprising, since 90-95% of the farmers grow the

rust-susceptible variety RR21.

Fig. 19 outlines hypothesized causes of the wheat rust problem. These are

mainly associated with the continued use of RR21 by farmers, the presence of races of

rust that infect RR21, and agroclimatic conditions that favor disease transmission.

One possible complication is that farmers like the taste characteristics of RR21 and

its wide adaptability under varying soil fertility conditions. Many farmers may be

unwilling to give up this variety without being given a replacement of similar yield and

quality.















Fig. 19. Hypotheses on problems and causes:
Rust reduces wheat yields


RUST REDUCES WHEAT
YIELDS


FARMERS USE FARMERS LIKE
AGROCLIMATIC RUST-SUSCEPTIBLE THE TASTE
CONDITIONS FAVOR VARIETIES CHARACTERISTICS OF
RUST DEVELOPMENT (ESPECIALLY RR21) RR21




SEED OF RUST MANY FARMERS RR21 HAS THE
RESISTANT HAVE LITTLE ABILITY TO ADAPT
VARIETIES NOT KNOWLEDGE OF TO DIVERSE
EASILY AVAILABLE ALTERNATIVE CONDITIONS IN THE
VARIETIES HILLS


Key: Rectangles problems; ovals causes







It would seem important to popularize and make available new, rust-resistant

varieties of wheat at the village level. Farmers should be encouraged to save and

store the seed of preferred new varieties for sale to other farmers in subsequent

seasons. Work should continue to screen resistant germplasm that has some of the

quality and broad adaptability properties of RR21. This can be done on-station -- but

with a good testing program under farmers' field conditions, and with farmer partici-

pation in screening for characteristics other than yield.


4.6 Weeds and weed control


Wheat in the study area is rarely weeded, although weeds in fields near the

homestead may be occasionally removed for fodder. Farmers do not use herbicides.

Although wheat fields are not weeded, farmers did complain about yield losses to

weeds, especially the weed called pire (Polygonum hydropiper water pepper) the

most common weed in khet lands. Yield losses from pire of up to 50% were reported

by farmers, with an average loss of about 10% on half of the fields.


Various other grassy and broadleaf weeds were observed in the khet and bari

wheat fields and are listed in Appendix 4. Alopecurus pratensis, and Polypogon fugox

were other important weeds on khet land and Chenopodium album was important on
bari land.


Fig. 20 summarizes hypotheses on causes of the pire weed problem.

Difficulties with weeding wheat fields because of broadcast seeding was frequently

mentioned; however, this does not explain why farmers do not remove weeds from

the field while they harvest the wheat. Pire is not a palatable weed (it has a hot

peppery taste) and is not fed to animals, which may explain why it is not removed












Fig. 20. Hypotheses on problems and causes:
Weeds reduce wheat yields, especially on khet


Key: Rectangles problems; ovals causes







from the fields. However, it can be composted. The weed is associated with continu-

ous rice-wheat cultivation. Farmers noted that after a crop of potatoes, fields were

relatively free from this weed. The problem is exacerbated by poor plant stands.


Alternative control measures need to be assessed by farmers. These include

the use of alternative rotational crops such as potatoes, spring maize and legumes;

mixed cropping with toria; or the removal of the weeds from the field before they set

seed. Improvement in plant stands would result in better wheat competition with this

weed. Finally, chemical control is always an option, although poor availability of

sprayers and farmer inexperience with herbicides may reduce its feasibility.


4.7 Insects and rodents


The major insect pests of wheat do their damage during storage. These pests

will be briefly discussed in section 4.8. Rodents were identified by farmers as a prob-

lem in the field as well as during storage. Rat damage in the field includes cutting of

stems and destruction of grain. Farmers use no rat control measures. The incidence

of this problem, and the damage caused by it, remain unclear.


Insect problems in other crops were reported by farmers. In rice, leaf folder

(Leptocorisa oratorius), rice bugs (Cnaphalocrocis medianalis), hispa (Dicladispa

armigera), stemborer1 and armyworm (Mythemma separate) were said to reduce

yields. A buildup of rice insect pests because of cropping intensification and introduc-

tion of early rice into triple- and double-cropping systems was noted as a possible


1. According to entomologists on the survey team, four species of stemborer are found in the area and
more work is needed to assess their importance. They are: stripe stem borer (Chilo suppressalis); pink
stem borer (Sesamia inferens); yellow stem borer (Scircophaga incertulas); and white stem borer
(Trvporvza innotata).







cause of the gradually declining rice yields mentioned by farmers. Additional infor-

mation on rice pests should be obtained during the rice survey scheduled for later this

year.


Insects were also identified by farmers as a problem in maize. Farmers re-

ported that the white grub (Phyllophaga spp.) was a problem in bari maize, especially

on sandy balautemato soils. Continuous application of undecomposed compost to

bari fields would encourage the proliferation of this pest. Most farmers do not control

white grub, either in the field or in the compost pit. However, some farmers reported

collecting large numbers of grubs by hand during plowing, while others reported using

BHC dust at 20-40 kg/ha. Gradual yield declines in maize might conceivably be linked

to a white grub buildup. Maize borer (Sesamia inferens) was another pest observed

in emerged spring maize in lower khet lands. The incidence and importance of this

pest in spring khet and summer bari maize remain unclear.


Many farmers have experienced terrace face collapse in bari and khet. Some

farmers blame this on the increased use of chemical fertilizers coupled with water and

rain erosion. One explanation given by farmers has to do with the movement of mole

crickets from the main field to the bunds as a result of applying chemicals. These

crickets are said to burrow in the terrace face and bunds and weaken them, causing

them to collapse when water flows through the holes. This notion needs further study.


4.8 Harvesting, threshing and storage


Wheat is harvested from the end of Chaitra to the third week of Baishak (early

April to early May). Cutting is done manually with sickles, with women doing most of

this work. (Men are busy at this time preparing the maize fields for planting.) The







wheat may be carried to the homestead for threshing (if it is near), or it may be

threshed directly in the field.


Wheat is usually threshed by beating it with wooden sticks. Sometimes thresh-

ing is conducted by beating bundles of wheat against a stone. There are no mechani-

cal threshers in the area, although it might be possible to introduce them in the more

accessible parts of the lowlands. The threshed grain is carried back to the farm house

for cleaning and winnowing by women. (Women are also usually responsible for

transporting the wheat straw back to the household, often uphill and over long dis-

tances.) After cleaning, the grain is usually sun dried and then stored.


Storage losses are a problem for wheat seed and grain. Many farmers report

storing wheat seed and grain in the same way -- in gunny bags, fertilizer bags, bamboo

baskets, or sometimes heaped on the floor in the house. Weevils and rodents are

largely responsible for these storage losses. Some farmers use BHC dust to protect

their stored wheat seed, but weevils may have developed resistance to this chemical --

its effectiveness is said to be reduced. Other farmers report using methyl parathion

on both seed and food grain to control insects. Many farmers frequently sun dry their

seed to rid it of insects.


Fig. 21 summarizes hypotheses on the causes of wheat storage losses. Some

work on farmer assessment of alternative storage methods is needed. The "Save the

Grain" project, and work conducted at the Agricultural Research Station (ARS)

Lumle, and the ARS Pakhribas, are relevant here. Farmers may not be fully aware of

the hazards of using pesticides, so evaluation of indigenous plants with insecticidal

properties may also be useful.















Fig. 21. Hypotheses on problems and causes:
Storage losses reduce effective wheat yields


INSECTS FUMIGATION
BECOMING CHEMICAL USE NOT SOMETIMES
RESISTANT TO A( ALWAYS EFFECTIVE INEFFECTIVE BECAUSE
CHEMICALS, E.G., CONTAINERS NOT
BHC () TIGHT


STORAGE LOSSES FROM
WEEVILS, RODENTS
AND MOISTURE REDUCE
EFFECTIVE WHEAT
YIELDS


Key: Rectangles problems; ovals causes










5 Wheat problems: A summary and an agenda for action


5.1 Problem listing and ranking



In the discussion of study area farming systems (section 3) and wheat produc-

tion practices (section 4), a number of problems affecting wheat were uncovered. In

this report, as in other rice-wheat diagnostic survey reports, a "problem" is taken to

mean:



a factor associated with near-term yield loss;

a factor associated with inefficient use of external inputs or farmers' re-

sources (regardless of effect on yields);

or a factor associated with reduced system sustainability or land degradation

over the longer term.



A listing of wheat-related problems uncovered during the survey includes:1



Rust infection reduces wheat yields. This problem was traced to the use of

susceptible varieties, combined with agroclimatic circumstances favorable to rust

development.

Rats reduce wheat yields before harvest, by cutting tillers and destroying

grain.



1. Given the unfortunate timing of this survey (much of the wheat had been harvested by the time the
survey was conducted), this list of problems is necessarily somewhat tentative. Note that numerous
problems associated with enterprises other than wheat were also uncovered. Given the objective of
this report (and the mandate of the broader collaborative rice-wheat research program), these prob-
lems are not discussed here.







Weed competition reduces wheat yields, especially pire (Polygonum hydro-

piper) on khet lands. This was traced to farmer weed management practices.

Erosion (interpreted as washing of topsoil or removal of nutrients by water

as well as terrace face collapse) reduces near-term wheat yields and threatens the

longer-term productivity of wheat fields, especially in bari.

Nutrient deficiencies and/ or toxicities, possibly associated with soil acidifica-

tion, reduce near-term wheat yields and threaten the longer term productivity of

wheat fields.

Low plant populations reduce wheat yields. This was traced to issues of low

initial plant populations (due to low seed rates and poor quality seed), as well as poor

tillering (most likely associated with the land quality issues noted above).

Late planting reduces wheat yields. Bullock and labor scarcity appear to play

a role here.

Late season moisture stress reduces wheat yields (especially in a drought

year such as this one).

Wheat storage losses from rodents, weevils, moisture, etc. reduce effective

yields and damage stored seed.


The problems listed above are not, however, equally important. In an effort to

determine the relative importance of each problem, a weighted scoring model was

constructed following Tripp and Woolley (1989) and Harrington (1991). In this

model, the annual regional productivity loss (ARPL) was estimated for each problem.


The ARPL is measured as a proportion of the maximum obtainable yield for a

particular location under "good" management (i.e., when factors under farmers'

control do not limit yields), and is estimated as the product of:


the frequency of occurrence of a problem;








the proportion of crop area (in this case, wheat harvested area on both khet

and bari) affected by the problem; and

the productivity loss associated with the problem for areas affected and for

years when it occurs.1



When the information available is inadequate to make even rough judgments

on the value of a parameter, a question mark is entered in the model (thus signaling

areas for future diagnostic research). The weighted scoring model developed during

this diagnostic survey is shown in Table 5.



Among the problems identified, rust and storage losses were judged as the

most important. Late planting (because of the small area affected) and drought stress

(because of its relatively low frequency) were judged as less important. Weed compe-

tition was judged to be of intermediate rank, i.e., below rust and storage losses, but

above late planting and drought stress. Several problems could not be scored,

however, due to lack of information. These included the problems of soil erosion and

land quality (including soil acidification and nutrient depletion) which most survey

participants felt that these problems were supremely important. Many participants

also felt that the problem of plant populations is important. Additional diagnostic

research can be used to clarify the relative importance of each of these problems.


5.2 Links among problems: A synthesis



In section 4, individual problems were discussed, and hypotheses on their

causes were presented. An understanding of causes is often important, because they


1. This does not imply that researchers or farmers should strive for maximum yields. Note that esti-
mates of ARPL from diagnostic surveys are necessarily very imprecise and are only used as a first
approximation at separating the more important from the less important problems.












Table 5
Preliminary scoring of wheat problems

Prod- Wheat
activity area
Problem a/ loss b/ affected
(%) (%)


Rust
Rats
Weeds d/
Erosion e/ **
Acid soils/ nutrients f/ **
Plant population


10-;

8-:


20 90-100 100
? ? 100
37 27-75 100
? ? 100
? ? 100
? ? 100


Late planting 5-30 0-10 50 0-2
Moisture stress 10-20 20-50 15 0.3-1.5
Storage losses/ wheat 5-15 100 100 5-15
. . . . . . . . . . . . . .- - -


TOTAL:


21-46


a/ For more complete descriptions of problems, see text.

b/ Yield loss or efficiency loss for the affected area only
(productivity Loss and area affected estimated simultaneously).
For sustainability problems, this is interpreted in terms of the
annual contribution to "unsustainability", usually quite diffi-
cult to quantify.

c/ Annual regional productivity Loss -- the product of percentage
productivity loss, percentage wheat area affected and percentage
years. Expressed as a percentage of unconstrained wheat yield.

d/ Primarily khet lands.

e/ Primarily bari lands.

f/ Both khet and bari lands.

** Both of these problems are related to sustainability issues,
and are somewhat interlinked.


can suggest avenues of action (research, extension, policy) useful in attempting to

solve the problems. A problem-by-problem approach is not entirely satisfactory,

however, because of the possibility of linkages among problems. A review of informa-


Years
(fre-
quency)
(%)


ARPL c/
(%)


10-18
7
6-10
?
?
?








tion acquired during the survey led to a recognition of several ways in which problems

interact.1 These interactions were usually found to be somewhat indirect. For

example, storage damage to wheat seed was thought to contribute to poor plant

populations and weed competition by means of its unfavorable effect on initial plant

stands. Similarly, continuous rice-wheat cultivation was found to be linked to both

land degradation (in the form of nutrient depletion) and the buildup of problem

weeds. These linkages are summarized in Fig. 22.


5.3 An agenda for action


A brief diagnostic survey of the kind described above is obviously an insuffi-

cient base for taking definitive decisions on a research agenda. Given the understand-

ing of the farming system achieved to date, however, along with an appreciation for

the relative importance of problems identified and the several causes of these prob-

lems, some suggestions can be made with regard to research themes that may be

appropriate.


Actions aiming to address the problems examined in the sections above are not

necessarily restricted to agricultural research. In this paper, four kinds of action are

considered, as follows:


diagnosis (research aimed at improving the definition of problems thought to

be important -- e.g., with regard to productivity loss, area affected, frequency, land

type or soil type affected, major causes, etc.);




1. It should be noted, however, that one of the major problems --rust infection -- does not appear to
be linked in any way to other problems.












Fig. 22. Summary of hypotheses on problems and causes


AGROCLIMATIC RUST REDUCES WHEAT FARMERS USE A
CIRCUMSTANCES FAVOR YIELDS RUST-SUSCEPTIBLE
RUST DEVELOPMENT VARIETY (RR21)



Key: Rectangles problems; ovals causes









assessment of possible solutions (research aimed at assessing alternative

solutions for problems that are thought to be well understood and well defined;1

extension (activities undertaken to promote a seemingly feasible solution to a

well-understood problem);

policy (activities undertaken to inform policy makers of implications of policy

for farmer adoption of new technology; or of agricultural problems that are best

solved through policy interventions).



A list of actions developed by review of diagnostic survey results is listed in

Table 6. This list is necessarily incomplete. Future diagnostic research will undoubt-

edly serve to refine and improve this listing.




Table 6. Suggestions for action


RUST

Continue to develop and screen "flexible" varieties Solutions
that have the quality and environmental adaptation of
RR21, while incorporating appropriate disease and
insect tolerance.

Farmer assessment of new wheat varieties, evaluations Extension
of varieties in farmers' fields. Solutions

Develop systems of village-level seed multiplication Extension
and storage.

PLANT POPULATION

Ascertain reasons for farmers' use of Low seed rates. Diagnosis

Ascertain reasons for poor tillering of wheat in Diagnosis
farmers' fields.

Develop systems of village-level seed multiplication Extension
and storage.



1. This can be extended to cases where complementary diagnostic research is proceeding concurrently.









Table 6, continued...


Research on establishment methods for wheat, e.g., Solutions
surface seeding, to improve populations while
forestalling weed growth.

Research on improved wheat seed storage methods (farmer Diagnosis
assessment of alternative improved systems; evaluation Solutions
of farmer-developed materials and methods).

Research on the use of agricultural implements and Solutions
tools to increase the efficiency of hill farming.

WEEDS

Research on establishment methods for wheat, e.g., Solutions
surface seeding, to improve populations while
forestalling weed growth.

Diagnosis of yield loss and research on control methods Diagnosis
for major weeds. Solutions

Study of wheat+toria cropping systems and effects on Solutions
weed control.

Research on weed control by means of crop rotations, Solutions
e.g., with potato, legumes, spring maize, to replace
wheat.

Research on chemical weed control. Solutions

Research on possible benefits of removing weeds after Solutions
harvested, before they set seed, in infested fields. Extension

Weed survey to collect, identify and preserve weeds. Diagnosis

STORAGE

Develop systems of village-level seed multiplication Extension
and storage. Demonstrate storage methods.


Research on improved wheat seed storage methods (farmer
assessment of alternative improved systems; evaluation
of farmer-developed materials and methods). Research on
use of indigenous plants with insecticidal properties.

FERTILITY

Research on methods of erosion control, including
terrace face collapse in bari land. This is likely to
require farmer participatory methods.

Demonstrate the benefits of pH correction.


Diagnosis
Solutions


Diagnosis
Solutions



Extension











Table 6, continued...


Analysis of soil and plant material to determine Diagnosis
nutrient deficiencies or toxicities and development of
lime requirement indices.

Long-term studies on causes of land productivity Diagnosis
decline; farmer participatory research on alternative
land management systems.

Research on methods to improve nutrient availability Diagnosis
and use efficiency (affecting compost, chemical Solutions
fertilizer, micronutrients, etc.) -- may include
studies on placement, timing, application methods, etc.

VERTEBRATE PEST MANAGEMENT

Research on yield loss and control of rats in fields and Diagnosis
in storage. Solutions

Research on insect pest buildup (monitoring farmers' Diagnosis
fields)
...................................................................







Many of the actions suggested in Table 6 can be undertaken in the near term.

These are the actions labeled for extension attention, and focus on issues of seed

availability. More of the actions call for additional diagnostic research, however. The

next step in such additional diagnosis should be a diagnostic survey conducted during

the forthcoming rice season. Diagnostic activities are not restricted to surveys,

however. Improved understanding of problems and their causes may spring from

farmer participatory research, on-farm experiments, and even from laboratory tests

(e.g., seed germination tests) if they are focused on specific hypotheses. A glance at

the suggested diagnostic activities listed in Table 6 will confirm this.



Suggested actions to assess alternative solutions to well-defined problems go

well beyond traditional researcher-managed trials. For many of the issues and prob-







lems, farmer participatory trials are likely to be the most efficient approach. Whether

a program of farmer and field monitoring should be commenced in this site remains

to be seen. While it would be undoubtedly valuable, the costs of commencing and

faithfully continuing such a program must be compared to its benefits (weighted by

the probability that the program will continue for a minimum number of years).


Finally, it should be noted that this survey was conducted in a fairly small,

compact area in a single district in the mid-hills of Nepal. Caution should be used

when attempting to extrapolate these results to other rice-wheat areas in the mid-

hills. In the future, efforts should be made to compare the results reported in this

paper with results obtained by other researchers working on similar problems in

similar areas.


In conclusion, visits to study area villages have allowed survey participants to

gain new insights into farming systems, land quality issues, and wheat production

practices and associated problems. The tentative agenda for action outlined in Table

6 should provide a sound basis for collaboration between Nepali scientists and re-

searchers and extension workers with similar interests from other rice-wheat collabo-

rative research sites.







References


Carson, B., P. B. Shah and P. L. Maharjan (eds.), 1986. Land Systems Report: The
Soil Landscapes of Nepal. Land Resource Mapping Project. HMG-Government of
Canada. Kathmandu, Nepal.

Chand, S. and D. Gibbon, 1989. "Samuhik Bhraman: A Rapid and Appropriate
Method of Prioritising and Replanning Agricultural Research in Nepal". Presented
at the International FSRE Symposium, Fayetteville, Arkansas, October, 1989.

Fetzer, K.D. and L. Jung, 1978. "Factors affecting soil erosion by water within a re-
search area of Northwestern mountain ridge of Kathmandu Valley". Journal of
Nepalese Research Centre. Vol. 2/3.

Fujisaka, S., 1991. "A Set of Farmer-Based Diagnostic Methods for Setting Post-
Green Revolution Rice Research Strategies". Agricultural Systems. 36 (1991) 191-
206.

Fujisaka, S. and L. Harrington (eds.), 1989. The rice-wheat cropping pattern in Nepal
Tarai: Farmers' practices and problems and needs for future research.
NARC,CIMMYT,IRRI publication, Nepal.

Harrington, L., 1991. "Setting research priorities: Concepts and Applications to On-
Farm adaptive research". Presented at the Training course in agricultural research
operations, Department of Agriculture, Bangkok, Thailand, 3 July 1991.

HMG, 1990. Agricultural Plan for Kabhre District. Support to District Agricultural
Planning. HMG Ministry of Agriculture FAO/UNDP report NEP/87/010, Kath-
mandu, Nepal

Hobbs, P. R., G. P. Hettel, R. P. Singh, Y. Singh, L. Harrington and S. Fujisaka, eds.,
1991. Rice-wheat cropping systems in the Tarai areas of Nainital, Rampur and
Pilibhit Districts in Uttar Pradesh, India: Diagnostic surveys of farmers practices
and problems and needs for further research. Mexico,D.F.:CIMMYT.

Hobbs, P. R., G. P. Hettel, R. K. Singh, R. P. Singh, L. Harrington, V. P. Singh and K.
G. Pillai (eds.), 1992. Rice-wheat cropping systems in Faizabad District of Uttar
Pradesh, India: Exploratory surveys of farmers' practices and problems and needs
for further research. Mexico D.F.: CIMMYT.

Krishna, K. C., A. R. Sharma, and D. L. Gait, 1987. Baseline survey report for produc-
tion program. National Agricultural Research Centre, Khumultar, Nepal. Sered
Report No.8.

Paudyal, B. K. S., R. B. Shrestha, S. B. Mathema, D. L. Galt and K. B. Kadayat, 1988.
An initial site description survey report for outreach research program.
Lamatar,Dhadakhel and Nallu village Panchayat, Lalitpur District. National
Agricultural Research Centre, Khumultar, Nepal. Sered Report No.14.

Shah, P. B. and H. Schreier, 1991. "Nutrient deficiency and soil fertility issues". In Soil
Fertility and Erosion Issues in the Middle Mountains of Nepal. A Workshop
Proceedings. Jhikhu Khola Watershed, April 22-25,1991. IDRC, Canada. pages
260-266.







Shrestha, R. B., B. K. S. Paudyal, S. B. Mathema, D. L. Gait and K. B. Kadayat, 1987.
An initial site description survey report for outreach research program. Chittapol
and Nangkhel village Panchayats, Bhaktapur District. National Agricultural Re-
search Centre, Khumultar, Nepal. Sered Report No.12.

Shrestha, R. B., B. K. S. Paudyal, and S. B. Mathema, 1988. An initial site description
survey report for outreach research program. Mulpi and Subba village Panchayats,
Kabhre District. National Agricultural Research Centre, Khumultar, Nepal. Sered
Report No.18.
Shrestha, R. B., B. K. S. Paudyal, and S. B. Mathema, 1989. An initial site description
survey report for outreach research program. Pokhare and Phataksila village
Panchayat, Sindhupalchowk District. National Agricultural Research Centre,
Khumultar, Nepal. Sered Report No.21.

Tripp, R. and J. Woolley, 1989. The planning stage of On-Farm research: Identifying
factors for experimentation. Mexico D.F. Mexico, and Cali, Colombia: CIMMYT
and CIAT.

Wymann, S., 1991. "Land use intensification and soil fertility in agricultural land: A
case study in the Dhulikhel Khola Watershed". In Soil Fertility and Erosion Issues
in the Middle Mountains of Nepal. A Workshop Proceedings. Jhikhu Khola
Watershed, April 22-25,1991. IDRC, Canada. pages 253-259.










APPENDIX 1.
Participants in the diagnostic survey
held in Kabhre District Nepal during the
wheat season, 20-28 April 1992


Position


Chiranjibi Adhikary
M.L.Baidya
B.K. Batsa
B.K.Gyawali
S.K.Joshi
Y.G.Khadka
Gautam Pradhan
Laxmi Rai
J.D. Ranjit
R.K.Subedi
D.B.Tamang
V.N.Upraity
L.Harrington
M.Ruckstuhl
P.R.Hobbs


Asst. Agronomist
Site agronomist
Pathologist
Deputy Entomologist
Asst. Economist
Asst Soil Scientist
Asst. Agronomist
Community Development
Asst. Agronomist
Assistant ADO
Agronomist
Asst.Agronomist
Ag.Economist
Plant Pathologist
Wheat Agronomist


NARC, Khumultar
NARC, Khumultar
NARC, Khumultar
NARC, Khumultar
Dept. of Agric.
NARC, Khumultar
NARC, Khumultar
Kathmandu
NARC, Khumultar
Kabhre District
NARC, Khumultar
NARC, Khumultar
CIMMYT, Bangkok
CIMMYT, Nepal
CIMMYT, Nepal


Institution









APPENDIX 2:
Annual rainfall data for three
metrological stations in the study area
averaged over 10 years (1981-1990)


Year Nagarkot Panchkhat Khumultar


1981 1066 1107 1126
1982 1045 1215 1176
1983 1266 1044 1419
1984 1435 1153 1399
1985 1788 1664 1566
1986 2089 1276 1404
1987 1645 1243 1275
1988 1581 1293 1490
1989 1402 1151 1006
1990 2132 1212 1149
1991 na na 858

Average 1545 1236 1301










APPENDIX 3:
Recommended cultivation practices
for rice and wheat in the hills



RICE:


Variety Maturity Production Planting Recommended
(days) potential date altitude
(t/ha) (feet)
---------------------------------------------------------
CS45 120-130 3.5-4.0 15 Apr-30 Apr Lowland
Taichung 176 160-170 4.0-5.0 3 Jun-2 Jul 1000-5000
Khumul-2 130-153 3.5-7.7 3 Jun-2 Jul 3000-4500
Khumul-4 139-148 4.2-8.4 3 Jun-2 Jul 3000-4500
Palang-2 158-186 4.9-7.2 3 Jun-15 Jul 5000-6000
............----------------------------------..---.
Seed rate: 40-50kg/ha; fertilizer: 60-30-20.



WHEAT:
.............---------------------------------.. ----
Variety Maturity Production Planting Recommended
(days) potential date altitude
(t/ha) (feet)
----------------------------------------
Lerma-52 176 4.5 15 Oct Mid-hills
Lerma Rojo-64 168 5.0 1 Nov Mid-hills
RR21 146 4.0 1 Nov Mid-hills
Annapurna-1 168 5.5 1 Nov Mid-hills
Annapurna-2 168 5.0 1 Nov Mid-hills
---Seed rate: ------------------------------10.
Seed rate: 120-140 kg/ha; fertilizer: 80-40-20.










APPENDIX 4:
Weeds and fodder trees in the mid-hills.



Grasses and weeds
...............................................................

Polypogon fugax
Avena fatua
Atopecurus pratensis
Vicia sativa
Vicia angustifolia
Echinochtoa colona
Echinochloa crusgatti
Cyperus sp.
Polygonum hydropiper
Polygonum viscosum
Kalmo *
Kachhi *
Pankhuyu *
Phalkinchhe *


Trees* Scientific name
...............................................................
Dudhilo Ficus neriifolia
Gogan Saurauia nepautensis
Kutmiro Litsea monotepta
Nivaro Ficus auricuLata
Katus Castonopsis indica
Chhampu Magnolia campbelle
Amala Phyttanthus emblica
Kyaksim
Berkum
Sutchinchhe
Tangirum




* Local name.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs