• TABLE OF CONTENTS
HIDE
 Title Page
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Description of Lake Izabal and...
 The Local Climate
 Transmitted radiation in veget...
 Precipitation throughfall
 Biomass and element inventory
 Summary and Conclusions
 Appendix
 Literature cited
 Biographical Sketch
 Copyright














Title: Ecological studies on tropical moist forest succession in eastern lowland Guatemala
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00086527/00001
 Material Information
Title: Ecological studies on tropical moist forest succession in eastern lowland Guatemala
Physical Description: 133 leaves : ill. ; 28 cm.
Language: English
Creator: Snedaker, Samuel C
Publication Date: 1970
 Subjects
Subject: Forest ecology   ( lcsh )
Botany thesis Ph. D
Dissertations, Academic -- Botany -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis (Ph. D.)--University of Florida, 1970.
Bibliography: Includes bibliographical references (leaves 125-131).
Statement of Responsibility: by Samuel Curry Snedaker.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00086527
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 000415265
oclc - 37751440
notis - ACG2521

Table of Contents
    Title Page
        Page i
    Acknowledgement
        Page ii
    Table of Contents
        Page iii
        Page iv
    List of Tables
        Page v
        Page vi
    List of Figures
        Page vii
        Page viii
    Abstract
        Page ix
        Page x
        Page xi
    Introduction
        Page 1
        Page 2
        Page 3
    Description of Lake Izabal and the study area
        Page 4
        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
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
    The Local Climate
        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
    Transmitted radiation in vegetation
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
    Precipitation throughfall
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
    Biomass and element inventory
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
    Summary and Conclusions
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
    Appendix
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
    Literature cited
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
    Biographical Sketch
        Page 132
        Page 133
    Copyright
        Copyright
Full Text








ECOLOGICAL STUDIES ON

SUCCESSION IN EASTERN


TROPICAL MOIST FOREST

LOWLAND GUATEMALA


By
SAMUEL C. SNEDAKER














A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY










UNIVERSITY OF FLORIDA
1970












AC1O0WLEDGIETITS


The author wishes to express his sincere appreciation to Dr. Hu:gh

L. Popenoe for his efforts in assuring the success of this investigation.

Eis guidance, both at the University and in Guatemala, C. A., made com-

pletion of this work possible.

Special appreciation is due thu author's committee members,

Dr. A. F. Carr, Jr., Dr. J. F. Gerber and Dr. A. E. Lugo, for their

influence in providing the necessary foundation in ecology.

The support and encouragement given by Mr. Christopher HempEtead

in Guatemala are gratefully appreciated.

Special thanks are due Dr. D. B. Ward, Dr. J. F. Gamble, Dr. H. L.

Breland, Dr. R. G. Selfridge, Dr. F. G. Martin and Mr. R. R. Parks for

technical support in the preparation of field data for interpretation.

Dr. H. T. Odum and Mr. J. J. Ew'el'of the University of North C .roclJ.n'

provided insight into the theory of ecology. Mrs. M., M. Go:.cz tyrpei

the manuscript.

The author Nvishes to extend his appreciation to the Center for

Tropical Agriculture for providingg the necessary financial assistance.

The staff of the Center graciously administered project details.














TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS ................. .........................

LIST OF TABLES ....................... ....... ........ v

LIST OF FIGU`ES ............................................ vii

ABSTRACT ............................................ ix

INTRODUCTION ......................... ..................... 1

DESCRIPTION OF'LAKE IZABAL AND THE STUDY AREA ............... 4

The Lake ......................... .......... ...... ....... 6

Geology and Soils ..................................... 9

Vegetation and Shifting Agriculture ................... ..... 10

Selection and Description of Fallows ...................... 15

1-Year-Old Vegetation ............... ................ ... 17

2-Year-Old Vegetation .................................. 19

3- to 6-Year-Old Vegetation ............o............... 21

7- to 10-Year-Old Vegetation .......................... 22

THE LOCAL CLIMATE .......................................... 25

Radiation Balance ......................................... 25

Temperature ..................... 30

Precipitation ...................................... ......... o.. 38

Life Zone Classification .................................. 45

TRANSMITTED RALIATIO IN VEGETATION ......................... 9

Statement and Methods ..................................... 49











TABLE OF CONTENTS (CONT'D.)

Page

Results and Discussion .................................. 53

Spectral Composition Under a Forest Canopy ........... 53

Variation in Intensity Under Fallow Vegetation ....... 56

PRECIPITATION THROUGFFALL ..................................... 59

Statement and Methods .................................... 59

Results and Discussion ................................ 62

Factors Affecting Throughfall ........................ 62

Predictive Models Based on Total Precipitation ....... 69

1IOMASS AND ELEMENT INVENTORY ............................ 74

Statement and Methods .................................... 74

Results and Discussion .................................. 83

Floristic Composition of Fallow Vegetation ........... 83

Biomass Standing Crop ................................ 88

Element Inventory .................................... 97

Calculation of Study-Forest Age .................... 99

SUMMARY AND CONCLUSIONS ................................... 102

Theoretical Consideraticrs ...................................102

Shifting Agriculture and Fallow Succession in
Ecological Perspective ................................. 104

Events Leading to Fallow Succession .................. 105

Fallow Succession: A Descriptive Model ............... 108

APPENDIX ................................................ ... 113

LITERATURE CITED ............................................ 125













LIST OF TABLES


Table Page

1 Estimates of the mean hourly net radiation during
May, June and July at Finca Los Murcielagos, Guate-
mala .... .......... ............ .. .................... 29

2 Monthly temperatures at Finca Los Murcielagos,
Guatemala ......................................... ... .. 31

3 Monthly temperature characteristics for Finca Los
Murcielagos, Guatemala ................................ 33

4 Hourly temperature frequency during year at Finca
Los Murcielagos, Guatemala (expressed as a 5-year,
31-day rate) ...... ...... ........................ 35

5 Hourly march of temperature and biotemperature for an
average day during each month of the year at Finca
Los Murcielagos, Guatemala ..............c................ 36

6 Monthly precipitation at Finca Los Murcielagos,
Guatemala ................................. ......... ........ 41

7 Monthly precipitation at Las Dantas, Guatemala ........... 43

8 Average rain-day characteristics, summarized by
month, for Finca Los Murcielagos, Guatemala 4............. 44

9 Frequency of hours with precipitation of specified
amounts, summarized by month to provide a 5-year
rate for Finca Los Murcielagos, Guatemala ............... 46

10 Radiation at soil surface under fallow vegetation
and forest expressed as a per cent of total incident
solar radiation ............................0 ............. 57

11 Regression equations of precipitation throughfall on 3
precipitation-related variables (duration, intensity and
amount) for 4 forest fallows of known ages atFinca Los
Murcielagos, Guatemala ................ ............... 65












LIST OF TABLES (CCOT'D.)


Table


12 Biomass
element
fallows


standing crop and leaf compartmcent nmacro-
inventories in selected tropical lowland
and mature forests .............................


13 Relative frequency of occurrence of selected plant
species in vegetation of 4 general ege classes at
Finca Los Murcielagos, Guatemala ......................

14 Element inventory of the leaf corpartiment starndi.'g
crop in fallows of 6 selected ages at Finca Los
Murcielagos, Guatemala .............. ................


Page












LIST O FIGURES


figure Page

1 Map of the Lake Izabal Region in eastern Guatemala.
(Adapted from map in the Clasificacion de los Suelos
de la Republica de Guatemala.) 1959. Ministerio de
Agriculture .............................. .... 5

2 Weekly water level fluctuation of Lake Izabal, Guate-
mala, during> 1967. Water levels were recorded near
Las Dantas in meters above-the-lowest-level of the
year, which occurred during the week En:in,: Janxs..y 14.
(Data interpreted from records maintained by the Ihter-
national Nickel Corporation) .............................. 7

3 Relative weekly water temperature of Lake Izabal,
Guatemala, during 1967. Water temperatures were
recorded weekly near Las Dantas. (Data interpreted
from records maintained by the International Nickel
Corporation) .................................... ........ 8

4 Mean annual daily march of temperature and biotempera-
ture averaged from hourly terLp-r.'.tures rec:re.i from
July, 1962 through July, 1967 at Finca Los Murr:iclgs,
Guatemala ................................................ 39

5 Spectral distribution of the incident radiation between
0.5 and 1.1/t, and the per cent transmission throi[h a
32-year-old forest at Finca Los Murcielagos, Guatemala ... 54

6 Spectral distribution of the transmit: .e radiation be-
tween 0.5 and 1.1,u, and in bands 5 to 3 (see text), in
a 32-year-old forest at Finca Los Murcielagos, Guate-
mala ....... ....... .. ................ ... 55

7 Relationship between precipitation throughfall and (A)
hours with precipitation, and (D) total pre -i:itation
in a 14-year-old fallow at Finca Los Murcielagos,
Guatemala ........................................ ........ 67

8 Relationship between precipitation throughfall and (D)
maximum one-hour intensity, and (D) total precipitation
in a 32-year-old forest at Finca Los Murciel sc:,
Guatemala ............................................. 68


vii













LIST OF FIGURES (CONT'D,)


Figure Page

9 Regression of precipitation throughfall on (D) total
precipitation (expressed in per cent) for a 14-year-
old fallow and a 32-year-old forest at Finca Los
Murcielagos, Guatemala ................................... 70

10 Comparison of per cent throughfall versus total precipi-
tation during growing season for mature mixed hardwood
forests in the eastern United States and eastern lowland
Guatemala ............... ......................... 73

11 Comparison of species similarity among 4 age-classes.
The percentage value indicates the observed number of
species common to 2 age-classes (linked by arrow) out
of the total number of observed species in both age
classes ................................................... 89

12 Relationship between total above-ground standing crop
and vegetation age at Finca Los Murcielagos, Guatemala ... 91

13 Relationship between wood compartment standing crop
and vegetation age at Finca Los Murcielagos, Guatemala ... 92

14 Relationship between leaf compartment standing crop
and vegetation age at Finca Los Murcielagos, Guatemala.
The vertical lines for each age represent the mean
standing crop + 1 standard error. The horizontal line
is the weighted mean for .ll samples ...................... 94

15 Frequency distribution of the sample plots with leaf
standing crops falling within 200 gms m-2 class inter-
vals ..................................................... 95


viii








Abstract of Dissertation Presented to the Graduate Council of the
University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy


ECOLOGICAL STUDIES ON TROPICAL MOIST FOREST
SUCCESSION IN EASTERN LOWLAND GUATEMALA

By

Samuel C. Snedaker

August, 1970

Chairman: Dr. Hugh L. Popenoe
Major Department: Botany

Successional-fallow vegetation resulting from the abandonment of

shifting agriculture was studied at Finca Los Murcielagos in the Lake

Izabal region of eastern lowland Guatemala. Transmitted radiation and

precipitation throughfall vere compared among fallow of known ages and

a mature secondary forest. Total above-ground standing-crop determina-

tions were made in fallows of known ages, 1 through 10 years. These

studies were supplemented by an analysis of the regional climate, and

detailed descriptions of shifting agriculture and fallow vegetation

characteristics.

The levels of transmitted light in developing fallow ecosystems

decrease over the first 3 years to a minimini of 5.4% of the total in-

cident radiation. Observations in fallows of older ages suggest an

oscillating time-dependent pattern. This variation over age is thought

to result from changes in the vertical organization of canopy components.

The solar radiation intensity spectrum within a mature secondary forest

shows the canopy to be semi-transparent in the infrared range between

0.7 and 1.1 The lowest transmitted light levels in the visible

spectrum were recorded in this forest,









Precipitation throughfall is a function of the amount, duration

and intensity of precipitation and varies with the age of the vegeta-

tion. The total amount of precipitation falling with a 24-hour period

was determined by regression analysis to be the most important factor.

Variations in the relative importance of each precipitation factor in

throughfall were discussed for fallows of ages 8 and 14 years, and the

mature secondary forest. The regression of throughfall on total preci-

pitation in the secondary forest was compared with similar results

reported for eastern U. S. hardwood forests. The observed variations

in each of the fallow throughfall studies emphasize the structural and

spatial differences in fallow canopies.

Total above-ground standing-crop increases over time and after the

first year is the result of stem-wood growth. Large variations exist in

the standing-crop of leaves per unit area in the successional fallow.

The statistical analysis permits the conclusi.on that, based on the study

results, there is no discernible trend in leaf biomass changes during

the first 10 years of succession. The net annual production during

this time period was calculated to be 634 gms m-2 year-1 and is attributed

to increases in the wood compartment. The age of the mature secondary

forest was calculated by extrapolating the regression equation, total

biomass on time, to a steady-state value of 25450 gms.m-2 derived from

the literature. A cursory evaluation of the standing-crop of nutrients

in the leaf biomass was inconclusive because of uncontrolled variation

among samples.

Species composition and the gross structural characteristics of

developing fallow ecosystems are related to the length of the preceding







cropping sequence and fallow period. Lengthening the period of con-

secutive annual cropping and/or decreasing the length of the intervening

fallow period leads to site degradation and retards the development of

the subsequent fallow succession. The successional flora was divided

into 4 age-class categories based on their observed affinities in fallows

of known ages. Site degradation favors the flora of the 1- to 2-year

age-class and retards replacement by species of the later age-classes.

In the early stages of a developing successional ecosystem, 3

major structural changes were observed. Initially, the abandoned agri-

cultural site quickly becomes covered with green, leafy material. The

only woody plants observed in a 1-year-old fallow result from coppicing.

By the second year, the herbaceous species are succeeded by soft-woody

species capable of rapid highth growth. As they mature they are slowly

replaced by species with denser, harder stem-wood. These species

usually become dominant after the seventh year and add additional height

to the fallow canopy. The effects of site degradation on the time of

inception of these structural changes were discussed.












INTRODUCTION


Fallow second-growth successions cover large areas of the low-

land humid tropics where shifting agriculture is the dominant crop-

ping system of the local inhabitants. In 1963, Conklin published a

bibliography listing over 1300 titles covering all aspects of this

system and closely related topics. Very few of these papers cover

details of natural fallow succession, in spite of the almost universal

recognition of its importance in the maintenance of site quality. In

view of its role in the shifting agriculture cycle and tropical forest

regeneration, Popenoe (1960) and Richards (1964) consider the intensive

study of succession to have great practical value and theoretical im-

portance.

Shifting agriculture is a generic term referring to the pantropical

system of subsistence agriculture which involves the clearing of an

area of natural vegetation and the subsequent planting of one or more

crops, without mechanically disturbing the soil. Later, the site is

abandoned to the regenerating successional fallow vegetation, and another

area is selected for clearing and planting. In a closed cycle, the new

clearing is made in an area of previous cropping which, at the time of

selection, is in some stage of fallow succession. The reasons which

motivate the rotation of crops with fallows involve both the local

ecology and culture. In general, the subsistence agriculturist weighs

the effort necessary to achieve a certain minimum harvest on the present


- 1 -




- 2 -


site versus a fallov or forest site. For Fsme specific examples and

research orientations, see Newton (1960), Nye and Greenland (1960),

Allan (1965), Reina (1967), and Carter (]969).

The biotic and abiotic characteristics of cropping and succes-

sion, such as increasing weed and pest problems and changes in site

quality, ultimately foster the decision to abandon a site. But the

fact that a successful cropping sequence can be reinitiated following

a period in fallow emphasizes the importance of the recuperative aspect

of successional second-grocth vegetation. The duality of succession,

first in site degradation, then in site improvement, is recognized,

but only the fo-r.er aspect is treated extensively in the literature.

The relatively few ecological observations on succession.

vegetation in the American tropics have generally been restricted

to floristic composition and physiognomy (e.g., Stevenson, 1928;

Kenoyer, 1929; Lundell, 1937; Allen, 1956; and Richards, 1960), or to

specific aspects such as litter accumulation (Ewel, 1968) and nutrient

element uptake by 1-year-old vegetation (Tergas, 1965). Budowski (1960

and 1963), however, has outlined the principal differences between

early and late secondary successions and climax forest, and has formu-

lated a series of useful generalizations for tropical lowland forest

succession. The importance of Budowski's syntheses has been neglected

in most field studies on succession. The reports by Gamble, Snedaker,

et al. (1968) and Golley, McGinnis and Clements (1968), on bioenviron-

mental feasibility studies for a proposed Atlantic-Pacific sea-level

canal in Panama and Colombia, are a significant step forward in an

attempt to understand tropical forest succession as a portion of





- 3 -


the overall agricultural and terrestrial ecology. The Rain Forest

Project of the Puerto Rico Nuclear Center, initiated in 1962 (Odum,

1964), is currently providing much detailed information on succession

and radioecology in a tropical environment.

Before a complete understanding of tropical succession is

achieved, associated biotic and abiotic parameters must be isolated

and examined over time. The objectives of this study are to examine

the changes in transmitted radiation, precipitation throughfall and

biomass and nutrient element inventories that occur with increasing

age in fallow second-growth succession. The current understanding of

ecosystem development (i.e., succession) and function has been enhanced

by the use of certain descriptive terms and concepts such as maturity,

stability, structure and complexity (Margalef, 1968 and Odum, 1969).

The present inability, however, to fully evaluate quantitatively some

of these concepts requires that inferences be drawn from related, but

more vigorously defined, parameters. The data from this investigation,

interpreted within the context of ecosystem theory, provide a basis

for increasing our understanding of both succession in general and

tropical moist forest succession in particular.

A study area near Lake Izabal, Guatemala, C. A., was selected

where the shifting agriculture pattern has resulted in a heterogeneous

pattern of various even-aged, or nearly so, fallows and mature forest.

The relative environmental homogeneity afforded by the close proximity

of sites presents a unique opportunity to estimate the above changes

by comparison of successional fallows of differing ages. The study

was made during the wet seasons of 1964 through 1967.












DESCRIPTION OF LAKE IZABAL AND THE STUDY AREA


Lake Izabal is located about 40 km from the Atlantic coast of

Guatemala (Figure 1). It covers approximately 700 km2 and is roughly

)I7 km long in an ENW-WSW direction and 20 km wide. The lake is

formed by the drainage from the Polochic Valley and surrounding areas,

and flows through the Rio Dulce into the Eahia de Amatique of the

Caribbean Sea.

The relatively shallow lake, 16 m maximum depth, is bounded on

the north by the Sierra de Santa Cruz and on the south by the Sierra

de las Minas. To the extreme east lies the Montana del Mico which is

separated from the lake by a broad delta. The mountain elevations

rise to over 1.500 m and drop sharply along the southern shore. The

drop is less pronounced on the northern shore due to the presence of

two large deltas formed by alluvium from the Rio Sauce and Rio Tunico.

The entire western end of the lake is bounded by an extensive swamp

delta formed primarily by the Rio Polochic.

The study area of Finca Los Murcielagos is located midway along

the northern shore at 150 35' north latitude and 890 09' west longitude.

The finca covers an area of about 2000 ha and derives its name from

the village of about 30 families located on the shoreline of the property

(Figure 1). The elevation ranges from 10 m above sea level along the

lake shore to over 100 m further inland to the north.


- 4 -








































MINA
CE LAS-
t E. .... R. : ,
* ... '*' .. '*. .' :r".


Figure 1.-


Map of the Lake Izabal Region in eastern Guatemala. (Adapted from map in
the Clasificacion de los Suelos de la Republica de Guatemala. 1959.
Ministeric de Agricultura




- 6 -


The Lake


The water of Lake Izabal is fresh. Analyses by Tsukada and

Deevey (1967) show 300-400 reciprocal megohms of conductivity. The

preliminary results of a study of the bottom sediments by Brooks

(1970) suggest that within recent geologic times brackish conditions

never existed in the lake. However, the lake has a dominant marine

fish fauna (Saunders, Holloway and Handley, 1-950) and there is a

noticeably low species diversity and population of aquatic fauna and

flora.

Water levels and temperatures are recorded weekly by the Inter-

national Nickel Corporation (Exmibal) near their base camp at Las

Dantas. The water level averages about 10 m above sea level. Shown

in Figure 2 are the weekly water levels of 1967, expressed in 0.1 m

intervals above the lowest level which was recorded during the second

week in January. A general seasonal relationship with precipitation

exists; the highest water levels follow the months of highest precipi-

tation. The relative fluctuations in the weekly water temperature

are shown in Figure 3. The original data showed the water temperature

to average about 4 C higher than the ambient air temperature inland

from the lake, with the greatest differences occurring during the

periods preceding the highest water levels. The accuracy of these

temperature data are, however, open to question, particularly with

respect to the location (distance from shoreline and depth) of tempera-

ture readings. Reasons for the higher wet-season temperature are

unknown. The possible occurrence of hot springs (which are known in




. -7-


WEEK ENDING


Figure. 2.-


Weekly water level fluctuation of Lake Izabal,
Guatemala, during 1967. Water levels were recorded
near Las Dantas in meters-above-the-lowest-level of
the year, which occurred during the week ending
January 14. (Data interpreted from records maintain-
ed by the International Nickel Corporation)


<.9
>-

S.8
u_

uj.7
w
-J
c .6


5
1-.-



I *
xJ
S.4
UJ




6 _
F 3



- .1
UJ
2




-8-


I ,I l i | i




II Ili I |
SI
I t
I ji *i fI iI
I! I j i I
!~- h U 1 l


4 Mar


5 May


8 Jul
V/c.iEE EiD; G


2 Sep


II


II

* i L' I

2 Nov 29 Dec


Figure 3..-


Relative weekly water temperature of Lake Izabal,
Guatemala, during 1967. Water temperatures were
recorded weekly near Las Dantas. (Data interpreted
from records maintained by the International Nickel
Corporation)


0
C


7 Jan





- 9 -


the Lake region) in close proximity to Las Dantas could cause differen-

tially higher temperatures during periods of increased discharge and/or

changes in lake currents. In general, little is known of the lake it-

self or its influence on the local climate.


Geology and Soils


Roberts and Irving (1-957) and Brooks (1970) have given cursory

descriptions of the geology of the Lake Izabal area. Of particular

importance with respect to the Finca Los Murcielagos study area is a

late Paleozoic and early Mesozoic serpentine block which parallels the

northern perimeter of the lake. This serpentine block and the

occasional limestone deposits have contributed to the alluvial delta

which forms a large portion of Finca Los Murcielagos. In addition to

the terrestrial alluviums, Simmons, Tarano and Pinto (1959) have clas-

sified two soil series which are present at lower elevations along

the north lake shoreline. The Sebach series, dark-brown clays, and

the Guaipol series, red-brown clays, are developed wholly or in

part on serpentine parent material. Most of the upland soils at Finca

Los Murcielagos are Oxisols, with some hydromorphic forms, and Inceptisols

occur in the alluvial lowlands (Popenoe, personal communication).

Tergas (1965) partially described most of the soils of the

study area at Finca Los Murcielagos. They are, for the most part,

black-to-yellowish-brown clays derived from limestone, serpenting-and

other transported materials. Usually a good crumb structure is found

in the "A" horizon, and internal drainage ranges from good to poor.




- 10 -


Other soils within the study area are developed on beds of sand and

gravel, alluvial remnants of meandering rivers. Layers of sorted sand

or gravel appear to underlie all of the soils and have been observed

to depths of 2 m (unpublished data). Sediment cores taken in the lake

at the mouth of El Jagua Creek (Finca Los Murcielagos) and about 2 km

off shore ended in gravel at 50 cm and sand at 320 cm, respectively

(Tsukada and Deevey, 1967). Tergas (1965) reported uniformly high pH

values (up to 7.2) and highly varying Ca:Mg ratios for the soils of Finca

Los Murcielagos.


Vegetation and Shifting Agriculture


The vegetation along the northern perimeter of Lake Izabal is

composed primarily of broad-leaved hardwood forest and interspersed

areas of successional fallow vegetation. These latter areas, the

results of shifting agriculture, dominate the better-drained soils of

Finca Los Murcielagos, and extend onto the limestone foothills. The

majority of these fallow areas are less than 15-years of age, measured

from the time of abandonment of agriculture. The hardwood forests

reach heights of 35 m in the low alluvial areas and attain a somewhat

lesser height further inland. The fallow and forest components are

generally evergreen, but show various degrees of deciduousness on the

limestone foothills and on soils with a dominant percentage of sand

or gravel in the "A" horizon.

Soils on serpentine deposits are dominated by Pinus caribea in

open stands of less than 50p crown density (Ewel, 1968). These latter

areas have recently been exploited for lumber and are generally




- 11 -


unsuitable for crop production because of the soil characteristics.

None of these areas occur on Finca Los Murcielagos.

Selective logging appears to have been a regular activity around

the margins of the lake since the early days of the Spanish. The most

exploited tree species, probably then as now, are Swietenia sp. (mahogany),

Cedrela sp. (Spanish cedar), Tabebuia sp. (oak, no relation to Quercus

sp.), Calophyllum sp. (Santa Maria), Achras sp. (Zapotilla) and a few

others. Usually the logging is done within easy access of the lake to

facilitate transportation.

From the investigations so far completed by Yale University, it

appears that the area was inhabited by Indians of the pre-Maya stock,

probably as early as around 2000 B.C. (Voorhies, 1969 and personal

communication). Housesites, sherds and domestic artifacts found on

Finca Los Murcielagos indicate that this site has also been inhabited

since the Conquest, although inclusive dates are not yet known. Pollen

cores taken from Lake Izabal and examined by the palynologist Tsukada

(1966) with Deevey (1967) suggest that corn was the main crop during

the periods) of known habitation. Hernan Cortes (Gayangos, 1868)

observed large quantities of both corn and cacao being grown in the

Izabal area during his travels in 1526. These bits of information

almost conclusively indicate that the area has been under, at least,

a minimal intensity of shifting agriculture for the past several thousand

years, and that corn, then as now, was an important crop.

Population of the Lake Izabal area is presently dominated by Kekchi

Indians who have recently migrated from the highlands of Guatemala





- 12 -


(Carter, 1969). Watters (1966) notes that in-migration leads to un-

stable forms of shifting agriculture resulting in problems detrimental

to both the people and the natural resources upon which they must rely.

Presently the population pressure, specifically on Finca Los Murcielagos,

is light enough to allow agricultural sites to remain in fallow for a

period sufficient to regain their productive potential after each

cropping sequence.

.Shifting cultivation in and around Finca Los Murcielagos is

primarily of the voluntary type (Allan, 1965) in that cropping sites

are abandoned in anticipation of increasing weed and pest problems and

not as a result of them. Sufficient fallow and forested land exists

in the area, and the cultivator prefers to prepare a new site rather

than risk a smaller yield on the old site. In some instances the area

is not completely abandoned and certain crops which require more inten-

sive care, such as chile, sugar cane, cassava, pineapple, beans, bananas,

pumpkins, squash and melons, are.planted on the better drained and

more fertile portions.

On the finca lowlands, of which this study is representative,

successional fallow areas are usually cut over for cropping before

they reach roughly 10 years of age. Variations in the length of the

fallow period are dependent upon many ecological and cultural factors.

Most of the latter have been extensively treated by Carter (1969), who

has studied the Kekchi Indians and their agriculture in a nearby area.

Effects of the methods of site preparation, number of cropping

seasons and site care are reflected in the subsequent regenerating





- 13 -


vegetation. Site preparation, beginning with clearing, is finished

several weeks prior to the estimated arrival of the rainy season in

May. This allows for adequate drying of the slash before burning

yhich is delayed as long as possible before the rains start. In some

instances, the cultivator waits too long and- achieves an incomplete

burn of the dam? slash. This necessitates a second and sometimes a

third attempt at burning. A prolonged delay in attaining a thorough

burn sometimes requires an additional c-earing of the quickly regener-

ting vegetation.

The clearing procedure begins with the cutting of the smaller

vegetation and then the felling of the remaining larger trees. Gener-

ally all cutting is done with a machete, severing stems in such a

manner as to leave stumps ranging in height to 0.5 m. Large diameter

trees in older vegetation are cut with an axe at variable heights on

the bole.1 On loping terrain, vegetation at the lower end of the

site is cut first. In all instances the smaller trees are cut first.

This permits ease of movement in felling the large trees, ensures that

all trees are cut for subsequent drying, and probably enables the over-

lying wood to burn more thoroughly.

Planting is done with the aid of a dibble stick. A hole is made

in the ash and soil in which 3 to 5 corn seeds are dropped. Soil is



1 Johnston (1958) notes that in Chad, West Africa, the cutting of trees
at ground level, instead of a higher level, promotes stump regeneration
and coppice growth. This contrasts with the Iban of Sarawak whom
Freeman (1955) observed deliberately left high stumps to promote
coppicing when the fields were abandoned.




- 1 -


moved by the foot in such a manner as to partially cover the seeds.

Planting distances average about 1 m. A buffer zone of 5 to 10 m around

each site is also planted to corn, ostensibly to alleviate pressure

from herbivores and granivores. The corn is more widely spaced in the

buffer zone. At the time of planting, seed germination and root sprout-

ing of fallow vegetation have often already begun. (Root sprouts from

members of the Musaceae, particularly the Heliconias and Calatheas, grow

to I m in height during the first week.)

Various methods are used to protect the corn crop from animal

pests. Sometimes, crudely rigged scarecrows are placed in the field,

and subsurface snare traps are built of sticks and vines to catch

gophers vbich are eaten. When nocturnal damage by deer, racoons and

peccary is evident, the cultivator may stand night watch, with a rifle

if available. The deer and peccary are eaten. In general though, the

cultivator is not prepared to control insect and crop diseases. Recent

introduction of aldrin into the area has permitted a few cultivators

to spray or dust their crops, with relative success. It is also used

to control leaf-cutting ants by direct application on the nests.

During the maturation of the crop, the regenerating native vegeta-

tion is cut back with a machete one or more times. Coppice growth is

cut by striking the shoots against the stump with the machete. Sur-

face plants are gathered in clumps with a small hooked stick and cut

off at ground level with a machete.

Crops are sometimes planted twice on the same site within the

first year. The first, or summer, crop is grown during the wet season

and is almost exclusively corn. The second, or winter, crop may be




- 15 -


planted in October or November and is usually accomplished with only a

cutsory clearing of weeds which are not burned. A second corn crop is

considered an emergency crop planted to augment a poor vet season har-

Vest or dwindling corn supplies. A few entrepreneurs consistently

plant a winter crop for subsequent sale, even though harvest yields

are consistently lower than the wet season crop. Occasionally, the

second crop may be planted to some cultigen other than corn, such as

chile,,cassava, melon, pumpkin or squash.

Secondary vegetation regeneration initially begins with the clear-

ing operation, but does not dominate the site until after the final

weeding, which occurs around July or August. The final weeding, prior

to harvest, is not as thorough as the earlier weedings; for instance,

herbaceous vines entwining corn stalks are not removed. These stalks

continue to support vines during the early regrowth stages.


Selection and Description of Fallows


On the alluvial lowland portion of Finca Los Murcielagos and ex-

tending northward to the limestone foothills, an area of some 200 ha

was delimited around a permanent weather station for the selection of

various-aged second-growth fallow vegetation sites. It was assumed

the relatively small area would permit microclirnatic and other

comparisons to be made among the study sites and the permanent weather

station. The vegetation of the study area was all in various stages

of second-growth hardwood vegetation with the exception of swamp forests

bordering the lake shore and several areas of mature secondary forest

located, for the most part, on more poorly drained soils. The majority

of the fallow areas were under 15 years of age.









During the 4-year study period, 14 areas of various-aged second-

growth vegetation were selected to provide representative sample ages

of from 1 to 10 years since the abandonment of shifting agriculture.

Despite being able to select from an area of some 200 ha, a limited

number of potential sites prevented the study of every age-class fallow

within any one year. The progressive inclusion of finca activities

within the 200 ha further compounded the difficulty of selecting un-

disturbed second-growth vegetation areas.- This resulted in some

instances in the same site being studied for 1, 2 or 3 years in suc-

cession. The repetitive studying of the same site in successive years

proves useful by allowing the variation attributable to sites to be

taken into consideration.

The actual selection of second-growth fallow sites during any of

the years was made using the following criteria: (1) first preference

was given to age-classes not previously studied, (2) the site closest

to the permanent weather station was chosen if several of the same age

were available, (3) when there was no indication of grazing2 or

continued domestic use of the site by the local Indians. In addition,

several older-aged secondary forests were selected for special study.

Each location was given an identification number and further reference

to a -oecific site is by site number, and age, if applicable.

During 1966 and 1967, Ewel (1968) studied the dynamics of litter

accumulation under forest succession and selected many of these sites



2Cattle brought to the finca in June, 1965, were maintained within
fenced pastures, though on occasion would manage to get into the
second-growth vegetation. There were no instances of disturbance on
any of the permanent study plot areas.


- 16 -





- 17 -


for his detailed investigations. His findings, as they relate to this

study, are discussed elsewhere. The results of other investigators,

Tergas (1965) and Popenoe (personal communication), who have studied

these and nearby sites, are also mentioned.

The following descriptions of fallow vegetation are based on

numerous observations and the descriptions of the individual 9 m2

sample plots selected for biomass studies. The descriptions are given

for 4 age categories (1, 2, 3 to 6 and 7 to 10 years) since similari-

ties were observed aiong fallows of these age intervals. The ages

were determined from local informants, many of whom had previously

cropped the site in question and remembered details of the harvest.

The sample plots used in the cursory description are the same as those

listed by age in the Appendix, Table C.


1-Year-Old Vegetation

Second-growth vegetation at 1-year of age has a fairly uniform

appearance and averages about 2 m in height with scattered emergents.

The latter arise as a result of coppicing, usually several stems from

each stump. Woody coppice emergents with stem diameters of at least 1 cm

at a height of 20 cm were found to be present on about 56% of the 9 n2

sampled plots. Woody shrubs of small stature tend to be the dominant

component of the vegetation, and there is a noticeable paucity of

herbaceous ground cover, particularly of grasses and sedges. Individual

ligneous and large herbaceous stems were counted at a level of 20 cm on

each plot and the average density was 20 stems m-2. The vegetation

supports numerous herbaceous vines, predominantly of the families




- 18 -


Cucurbitaceae and Solanaceae, and are, for the most part, vertically

oriented.

Generally a mixture of species is found, though occasionally

areas varying in size have a monospecific dominant. Some of the more

important species observed to form homogeneous stands were Heliconia spp.,

Calathea spp., Gynerium sp., Allophyllus cominia, Waltheria americana,

Lantana camera, Piper auritumn, Trema micrantha, Plumeria prisahouai,

Smilax spp., and an unidentified fern. Tergas (1965) compared the stand-

ing crop (dry weight) of pure stands of Heliconia and Gynerium near San

Felipe, and mixed vegetation of the same age and found no unusual dif-

ferences.

Travel through vegetation of this age fallow is not difficult,

but is facilitated with a machete. This is due to a minimum of horizon-

tal integration of branches and vines, despite an apparently complete

canopy closure. Surface sunfleck area, based on 20 randomly placed

1 m2 quadrats, was estimated at 35e between noon and 1300 hrs. The

low vegetation density at 30 cm above ground level permits horizontal

visibility for several meters.

Sites which were cropped during both the summer and winter seasons

showed effects of the shorter growing period after abandonment prior to

the dry season.3 (These sites were not included in the more intensive

environmental and biomass studies.) Generally, the most character-



3 Local informants were relied upon for determining the number of crops
planted on a given site. This information was verifiable with relative
consistency for fallows up to about 3-years of age.





- 19 -


istic result of planting 2 crops within one year was the increased

number of herbaceous ground species, particularly the grasses. A

comparison of one of these sites 1 year later, with a 2-year-old site

showed a continuing relatively high frequency of these species.


2-Year-Old Vegetation

The greatest change in flora and structure appears to occur

during the second year. The second growth is doubled in height (to ca

4 m) and displays a greater number of emergents which may rise an

additional 2 m. The emergent species observed in 1-year-old fallows

dominate in the canopies of this and older fallows. The new emergents

are generally single stems of rapidly growing soft woods such as

Cecropia sp. and Cochlospermum sp. The relatively open canopy is

densest at about 3 m, though horizontal strata of vines begin to

close in at 1.5 to 2 m, thinning out and assuming a more vertical

orientation above 3 m. The vines are predominately woody, and many,

such as Byternia aculeata, are covered with thorns.

The monospecific dominant stands observed at 1 year of age are

less frequent at 2 years. The pure stands of Heliconia and Calathea

usually become overtopped by the surrounding vegetation, but appear

able to persist for some time in this condition. Piper auritum may

reach heights over 3 m and form a continuous canopy over a small area,

but is rarely seen, even as individuals, in fallows of older ages.

Gynerium, Waltheria .and Lantana were rarely observed in pure stands

in this or later ages, possibly because of their characteristics of

.height growth limitation and shade intolerance. The other species




- 20 -


mentioned as occurring in pure stands at 1 year were observed in the

canopy of this and older-aged fallows.

Area stem counts gave an average of 16 stems m-2. The reduction

is largely due to a lower frequency of multistem shrub species, such

as the once-dominant Waltheria or Lantana, as species diversity in-

creases. An apparent paradox exists in the fact that woody stems were

present on only 56% (same as observed in the 1 year fallow) of the

sample plots. This is due to the small sample size (n = 16) and the

establish ment of a disproportionate number of plots in areas which

contained many erect stems but no woody stems greater than 1 cm in

diameter at 20 cm.4 For example, one plot was dominated by a single

Orbignya cohune seedling and eight tall Piper auritum. The stems of

the latter, though large, contain over 80O water and were not consider-

ed, by the working definition, to be wood.

Penetration through a 2-year-old fallow requires the vigorous

use of a machete and visibility is restricted at all levels by an

abundance of leafy material. The first stages of a distinctive ground

flora are evident and include the seedlings of a number of species

which are observed in the canopy at later ages.

Accumulation of litter on the soil surface begins during the

initial cropping sequence and completely covers the mineral soil by

2-years. The litter depth does not appear to change with increasing

age (see Ewel, 1968), although there are small increases in the depth

of the humus layer.



On coppice stems, height was measured from point of origin on stump.




- 21 -


3-to 6-Year-Old Vegetation

These successions, covering a 4-year period, are discussed

together because of certain similarities and trends which are distin-

guishable from both earlier and later ages. This is not to imply that

any characteristic mentioned is or is not wholly restricted to these

or other age fallows. With more intensive field study, exceptions may

be observed which deviate from the general-descriptions.

Three year-old fallows generally give the appearance of being the

most densely vegetated at all levels, an aspect which progressively.

decreases with time. This age fallow on all sites proved the most

difficult to work in because of severely restricted visibility and

movement. Penetration is slow and tedious except on hands and knees.

There is a more open appearance at the 4th and later years and pene-

tration is more easily accomplished.

The height of the 3--year-old vegetation is not too different

(4 to 5 m) from a 2-year-old fallow and increases to only 6 to 7 m

at 6-years of age. There is a general proliferation of woody vines,

including Smilax sp. and Lygodium sp., assuming a dense horizontal, or

nearly so, layer just below the canopy and extending through it. In

some instances, vines may traverse a horizontal path of up to 6 m from

ground origin to the canopy. Several species of woody vines have been

observed to completely cover the canopy over relatively large areas and

extend onto the emergents. Cochlospermum sp. and Cecropia sp.

continue as the more conspicuous emergents, but a diverse group of other

species gradually assume this role. Invariably, Cecropia sp. was

observed to be the tallest species in all fallows of ages 2 through




- 22 -


10 years. Cochlospermum sp. usually becomes completely dominated

by.vines and, following a slow necrosis of the crown, is eliminated

from the second-growth flora by about 5-years of age.

During the fallow-age progression from 3 to 6 years, the average

number of woody stems m-2 decreases from 13 to 10. This is thought to

be primarily the result of coppice stem mortality which occurs more

rapidly than increases in the number of new woody stems arising from

seed.


7- to 10-Year-Old Vegetation

The dominant aspect of these aged fallows is the relatively open

appearance below the canopy. Canopy closure begins at about 5 m and

extends up to about 10 m. Emergents are not obvious due to vines

obscuring the crown canopy surface. The only emergent observed that

is present in younger ages is the ubiquitous Cecropia sp. Between

the ground cover and the canopy there is very little foliar material,

but travel is hampered by randomly oriented -woody vines. There is a

sparse but distinctive ground. cover composed of members of the family

Liliaceae and Costus sp. The most conspicuous species are Piper

aeruginosibaccum and P. subcitrifolium, which occur in clumps over 1 m

in height.

The woody stems average about 4 m-2. These woody species are

almost exclusively canopy components arising as individual stems. In

most instances, it is difficult to distinguish the stems which have

arisen as the result of coppice regrowth, The exceptions are those

which have grown from large stumps. At these relatively older ages,

this aspect is useful in determining whether the previous vegetation




- 23 -


was a fallow regrowth or some stage of an older secondary forest or

mature forest.

Generally, the largest diameter stems belong to Bursera simaruba

and Nectandra sp.5 The former has been observed as a 31 cm dbh coppice

in a 10-year-old fallow. Several species have been observed in homoge-

neous stands of a dozen or more stems. Three of the more common are

Plumeria prisshouai, Allophyllus cor-ina and Tecona stans. Excluding

these, the distribution of other tree species is considered random.

Thus far, little has been said about the common corozo palm

(probably Orbignya cohune) which is a floristic component of all second-

growth fallows. Ostensibly, this species was protected by law because

of its fruit, which is comparable to the African oil palm. The corozo

palm is generally left uncut in the preparation of cropping sites around

Lake Izabal, which results in a greater proportion of these palms in

the fallow areas than might normally be expected. It appears the pri-

mary reason now for leaving the palm is that it is both difficult to

cut down and is the major source of the thatch used in Kekchi house

construction. The lower leaves, however, are cut off and distributed

over the site to ensure an even burn. It is doubtful in any case that

their preservation is a residual effect of the effort-by-statute to

protect them. In situations where the palm is present, in what the

cultivator feels to be an overabundance, they are thinned out by axe

when clearing the fallow vegetation prior to cropping. Those that are



Two locally important members of the Lauraceae, Nectandra sp. and
Ocotea sp., have not been identified to species, based on herbaria
material.




24



left are trimmed to a height of about 3 m or more to reduce crop shading.

In'this study, the large corozo palms, left uncut during preparation

for cropping, were avoided due to their disproportionate size and age

in relative to the regenerating vegetation.













THE LOCAL CLIMATE


Most general references agree that climatological data are

severely inadequate for Central America. Regional data that are

available are usually drawn from observations recorded at a minimum of

locations and suffer from the vagaries of extrapolation. This is par-

ticularly true for the lowland areas where climatological stations are

both limited in number and in the types of observations.

In July, 1962, a permanent weather station was established at

Finca Los Murcielagos. Initially, it was equipped with a Bendix-Freiz

Model 504 recording hygrothermograph, a Belfort recording intensity

rain gage and suitable reference instruments. During the course of

this study, other instruments were incorporated for varying lengths of

time. The results of these observations permit a description of the

local climate in terms of which the specific studies can be interpreted.


Radiation Balance


It is well established that the solar radiation incident upon

the biosphere is the ultimate source of energy for all abiotic and biotic

processes, and provides about 99.97% of the required heat energy.

Assuming the absence of an atmosphere, the radiant energy incident on

the earth's surface varies in a predictable manner with such factors

as solar disturbances, sun-earth distances, latitude, season and time

of day. The magnitude and variation of these factors can be determined


- 25 -




- 26 -


from meteorological tables such as those prepared by Berry, Bollay and

Beers (1945) and List (1966). Penetration through the atmosphere though

creates variations in solar radiation quantity and quality which are

dependent on local atmospheric conditions (Sellers, 1965). Detailed

knowledge of the quantity and quality of radiation incident anywhere

on the earth depends upon empirical data which are lacking for many

areas, especially in the geographical tropics.

The solar radiation incident upon an opaque body is differen-

tially absorbed and transformed into radiation of other wavelengths or

into energy of another form. The portion not absorbed is reflected

off the surface (Sellers, 1965). The difference between the radiation

incident upor an object and that directed back by both reflection and

reraiation into space defines the net radiation or radiation balance

for that object. When all forms of energy transfer are considered,

the result is an energy balance or beat budget. Budyko (1956) has

described the heat balance of the earth's surface and notes a zonal

irregularity in tropical isolines.

Based on Budyko's (1956) estimates, the region encompassing

Guatemala in Central America receives about 430 and 320 ly day-1 total

solar radiation in June and Decemiber, respectively. For comparison,

these values are respectively lo;er and higher than values for the

same months at higher latitudes in the northern hemisphere. The net

radiation balance for June and Decenber is about 260 and 190 ly day--,

respectively. The radiation balance at around 400 north latitude is

about the same as for the Guatemala region during June, whereas it is

tero or below during December. The distribution of the surplus energy





- 27 -


from tropical latitudes, giving a radiation balance equal to zero for

the entire earth, forms the basis of climatology discussed in most

textbooks on the subject.

In addition to the heat value of solar radiation, the spectral

quality and composition is also of great importance to living systems.

Of the three primary regions of the solar spectrum, ultraviolet,

visible and infrared, approximately 50% of the radiant energy is of

frequencies greater than 0.7u. in the infrared (Drummond, 1958, and

Gates, 1965a). Absorption within the infrared region by atmospheric

moisture produces the greatest spectral differences between sunlight

and cloudlight. With respect to a point on the earth's surface, sun-

light originates from a point source, whereas cloudlight is omnidirec-

tional and disproportionately lover in infrared energy. Diffuse sky

radiation, as opposed to direct solar radiation, is the integrated

result of all internal atmospheric scattering and reflection (Drummond,

1958). It is basically shortwave radiation in the ultraviolet,

visible and near infrared, and varies with the solar angle and sun's

declination.

Empirical determinations of the total radiant fluxes, the

radiation balance and the spectral distribution are notably lacking as

a part of the more intensive investigations of lowland tropical ecosystems.

The exceptions (Evans 1939; Schulz, 1960; Odum, Copeland and Brown,

1963; and others) consistently emphasize the necessity of this sort of

information.

Estimates of the hourly radiation balance were calculated from

data obtained with a Thornthwaite recording net radiometer maintained




- 28 -


at the permanent climatic station during June and July, 1963, and May

and June, 1964. The Thornthwaite instrument design (Fritschen, 1963)

permits measurement of the radiation balance (incoming minus outgoing

radiation) by integrating over time the temperature differential with-

in polyethylene-enclosed thermopiles exposed to both sky and earth.

The raw data indicated consistent positive net hourly losses up to 1.4

ly min-1 during most nights. The consistency and magnitude of these

losses are unexpected and have been suggested by Gerber (personal

communication) to be the result of moisture on the polyethylene hemi-

spheres distorting the transmission characteristics. At Finca Los

Murcielagos, nighttime precipitation is a common occurrence, and the

formation of dew on the upper hemisphere has been observed to occur

frequently. Moen (1968) has discussed an effect of dew formation on

the polyethylene window of a Suomi and Kuhn economical net radiometer

under nighttime sky exposure.

Presented in Table 1 are the estimates of the mean hourly net

radiation for three months. To correct for the apparently erroneous

nighttime values, the positive net losses were eliminated in the cal-

culations. Thus, a tacit assumption is made: the nighttime sky and

terrestrial radiation are equal, and heat losses occur by conduction

and convection. The values for June, 1964, probably represent the

best approximations. The net increase at 2200 hours agrees with

observations reported in the literature on net increases following a

minimum after dusk (Brooks, Pruitt, et al., 1965). Values for the

radiation balance are in magnitude agreement with literature values

(List, 1966), but high in comparison to the estimate made from Budyko

(1956).




- 29 -


Table 1. Estimates of the mean hourly net radiation(a) during May,
June and July(b) at Finca Los Murcielagos, Guatemala


Mean hourly net incident radiation(c)
Iv hr-1
May, 1964 June, 1964 June, 1963 July, 1963

0.2 2.6
0.2 2.6
0.2 2.8
0;3 2.9
0.5 2.8
1.8 3.0
7.6 6.6 2.7 0.4
17.9 19.1 12.4 8.9
33.8 30.1 24.3 18.6
46.5 44. 46.2 38.1
52,3 60.2 51.9 46.9
52.2 64.9 59.0 51.1
60.8 71.7 58.3 49.0
46.8 57.2 46.4 48.7
44.3 50.2 35.5 34.0
26.0 35.2 20.3 22.0
10.7 19.7 6.7 5.4
4.7 5.2 0.2
0.1 1.8
0.1 1.8
0.1 1.8
0.4 2.4
0.4 2.3
0.2 2.4


Total day"- 408.1 494.1 363.9 323.1


(a)Thornthwaite net radioneter.
(b)From continuously recorded data representing 18 days (May, 1964),
20 days (June, 1964), 19 days (June, 1963) and 16 days (July, 1963).
()Positive net hourly losses were averaged as zero due to errors
resulting from moisture condensation on radiometer hemispheres.


Hour

1
2
3
4
5
6
7
6
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24





- 30 -


Temperature


Descriptions in the literature of the annual temperature and pre-

cipitation for the Lake Izabal area are based primarily on records

maintained at Livingston (on the Atlantic coast), El Estor and stations

further to the west in the Polochic valley. The narrow range in tropical

temperature cycles though often permit relatively good estimates to be

drawn from few data.

The range in annual.temperature, calculated as the difference

between the warmest and coolest months, decreases to insignificance as

the equator is'approached. By interpolating the isohyet maps prepared

by the Observatorio Nacional (1964) and Riehl (1954), a mean annual

temperature of 25 C was estimated for the Finca Los Murcielagos area.

Both sources estimate the annual range to be within 2.5 to 5.0 C.

Bygrothermograph records were maintained during the period July,

1962, through July, 1967, at Finca Los Murcielagos. The number of

days in each month of the period for which 24-hour records are available

are listed in the Appendix, Table A. The temperature readings at each

hour were entered on punch cards and a cursory analysis of the data

was performed on an IBM 360 computer at the University of Florida.

Simple means were computed for all values except the mean monthly tem-

peratures for the 5-year study period. These values were weighted

according to Appendix, Table A.

The mean temperatures are presented by month and year in Table 2.

The mean annual temperature of 25.2 C, with a range of 4.4 C, are in

almost exact agreement with the estimates mentioned. The lowest





- 31 -


Table 2. Monthly temperatures at Finca Los Murcielagos, Guatemala


Nonth

January

February

March

April

May

June

July

August

September

October

November

December


Deorees centigrade
1962 1963 1965 1965 1966 1967 Mean(a)


26.1

26.9

27.6

26.7

24.4

23.7


23,8

23.7

25.9

27.1

28,6

NR(b)

NR

NR.

iHR

25.8

25.1

23.4


23,9

25.1

27.6

27,8

27.2

26.4

NR

25.5

25.1

22.8

22.7

21.7


20,3

22.7

23.8

25.9

26.7

26.8

26.6

25,9

25.8

26,1

NR

NR


Mean annual


22.8

24.4

22.4.

26.6

27.3

26.4

27.1

27.2

26.6

25.8

23,2

22.2


22.4

24.3

23.3

25.9

27.6

26.8

26.4


temperature


22.7

23.7

25.1

26.4

27.1

26.6

26.5

26.6

26.5

25.2

23.9

22.4


25.2


(a)Weighted means
(b)Not recorded.


(see text).




- 32 -


temperatures are recorded in December and January and may be associated

with cold air masses moving in from the north during the winter. Early

morning temperatures as low as 11 C and daily means below 20 C are not

uncommon during these periods. During the month of May, maximum hourly

temperatures above 37 C frequently occur, though daily means above 29

C are rare.

In addition to the monthly means, the values for eight associated

parameters were calculated. These are presented by month in Table 3.

The annual mean values were also weighted in accordance with Appendix,

Table A. All of the temperature characteristics evidence trends asso-

ciated with the seasons of the year. In general, there is a wet season

beginning sometime around May and tapering off toward the end of the

year. December through April is considered to constitute the dry

season. SeasoLal characteristics are discussed later in more detail.

With reference to Table 3, the monthly mean-maximum temperatures

show a general correlation with the hour of occurrence. The higher

mean values occur later in the afternoon, and generally occur at even

later hours during the drier months. The month of May, in particular,

and the following two months appear to be anomalous in this aspect

and also with respect to the other parameters. The mean-minimum tem-

peratures also show a general correlation with the hour of occurrence

similar to the maximum hour relationship. The higher mean-minimum

values occur at later hours in the morning, but there is no clear

association between the two parameter values and season. The daily

temperature range is greatest during the drier months and reflects the

ameliorating effect of precipitation on local climate during the wet

season.






Table 3.- Monthly temperature characteristics: for Finca Los Murcielagos, Guatemala


Mean


Hour of


maximum occurrence
temp. of max.
Month C temp.

January 28.0 14:02

February 29.6 14:19

March 31.3 14:30

April 32.9 14:43

May 33.3 14:25

June 31.9 14:33

July 31.3 14:20

August 31.9 14:43

September 32.0 14:37

October 30.5 13:59

November 29.1 13:37

December 27.8 13:25

Mean(b) 30.7 14:14

(bTBiotemperature (Holdridge, 1967)..
(b)Weighted means.


Mean
minimum
temp.
C

19.1

19.6

20.4

21.2

22.4

22.9

23.4

23.1

23.0

21.7

20.4

18.8.

21.3


Hour of
occurrence
of min.
temp.

4:50

5:14

4:51

4:44

4:35

4:49

4:37

4:49

4:57

5:03

4:55

4:44

4:50


Daily
temp.
range
C

8.9

10.0

10.9

11.7

10.9

9.0

7.9

8.8

9.0

8.8

8.7

9.0

9.4


Temp. Temp.


median Biotemp.(a)'


-cl


mode
C

21.1

22.2

22.2

23.9

23.9

23.9

24.4

24.4

25.0

24.4

23.3

21.1

23.3


C

22.2

23.3

24.4

26.1

26.7

26.1

26.1

26.1

25.6

25.0

23.3

22.2

25.0


C

20.6

20.3

17.1

15.6

15.3

18.8

20.1

18.1

17.5

18.9

20.9

20.9

18.7


-U


111-




- 34 -


Minimum temperatures usually occur at dawn as a result of

radiative and convectional cooling during the night. The minimum tem-

peratures at Finca Los Murcielagos, however, consistently occur up to

an hour prior to first sunlight. This may be in some way a result of

the nearness of the lake and its effect on the local ambient temperature.

The last colua in Table 3 gives the monthly mean biotemperatures

(Holdridge, 1967) calculated from the raw hourly data. Holdridge defines

the biotemperature as the "... mean of (all) unit-period temperatures

with substitution of zero for all tem:perature values below 0 C and

above 30 C, respectively." Zero and 30 C define the lower and upper

limits, respectively, of the temperature range over which vegetative

growth takes place. The exact upper limit is .tenta.tive pending further

investigation (Holdridge, 1967)

Both the 7conth]y teLperature modes and medians correlate with

the seasonal effects of precipitation; they are at their highest

during the rainy season. These values were calculated from the monthly

temperature-frequency listings in Table 4. The frequencies are grouped

in 1.0 degree Fahrenheit intervals, with 0.5 degree readings falling

in the next higher interval. As the temperatures were recorded in

degrees Fahrenheit, frequency tables were not converted to Centigrade.

Presented in Table 5 is a comparison by month of the hourly

march of ambient and biotemperatures. The biotemperature is given

only when it differs from the ambient temperature. Indicated differ-

ences are generally present during the daylight hours and reach the

maximum during the early afternoon. The daily march of temperatures

indicate a farming period during the period dusk to midnight in the

months of March, April and May. This may be due to warmer air moving





Table 4.- Hourly temperature frequency during year at Finca Los Murciela-
gos, Guatemala (expressed as a 5-year, 31-day rate)


Temperature frequency

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


55 3 6
56 7 6
57 5 14 3
58 17 14 4
59 23 12 3
60 45 45 15 1
61 27 47 7 4
62 23 41 41 3
63 48 24 21 11
64 107 45 36 11
65 117 51 64 17
66 199 88 80 27
67 188 118 88 58
68 191 116 106 80
69 191 141 156 95
70 272 241 207 94
71 215 233 155 103
72 260 257 230 126
73 246 230 176 203
74 213 198 201 217
75 140 188 155 223
76 129 186 152 1.70
77 129 155 162 159
78 118 128 142 14o
79 81 133 106 130
80 130 141 152 184
81 76 98 109 191
82 72 116 138 170
.83 94 102 107 113
84 81 104 103 14o
85 57 98 85 147
86 60 84 120 99
87 54 63 102 134
88 53 45 74 109
89 21 39 78 129
90 21 31 74 121
91 5 27 67 77
92 5 14 46 91
93 1 16 38 50
94 8 47 4o
95 4 43 17
96 4 4 12
97 2 10 12
98 4 7
99 2 4 3
100 1 5


3
3 3
4 3
3 1
1 7 8
5 1 24
2 7 7 35
3 13 11 71
2 14 36 87
51 43 101
6 1 57 84 161
9 1 7 6 68 115 172
8 4 9 6 81 139 269
17 4 15 16 89 172 211
33 9 .11 40 42 98 310 328
66 36 30 40 83 108 226 221
139 61 75 93 163 185 225 262
168 190 141 135 137 182 212 209
257 323 253 334 282 289 324 215
262 349 302 310 240 270 235 128
213 322 352 372 290 314 212 154
153 258 329 237 318 246 171 94
189 209 270 246 292 172 142 112
150 142 170 150 193 182 128 120
215 217 219 219 183 147 152 141
160 172 162 104 139 129 102 94
210 162 216 120 133 108 97 92
133 187 185 146 121 113 111 73
155 142 172 184 135 121 77 67
152 196 179 168 129 96 78 63
143 155 162 155 141 95 67 52
146 168 .120 124 137 85 78 50
127 107 144 139 139 102 53 24
94 99 104 104 107 83 37 25
127 68 78 126 97 64 27 8
89 39 28 71 74 50 10 6
97 36 12 46 46 40 14 1
61 19 3 15 34 17 3 1
50 16 13 10 13 1
42 10 14 10 3
22 6 10 5
13 9 4 4
8 4
2 3


I-






Table 5.- Hourly march of temperature and biotemperature for an average day during each month of the year at
Finca Los Murcielagos, Guatemalaka)


January
Hour Temp. Biotemp.u )


February
Temp. Biotemp.


Degrees centigrade


March
Temp. Blotcmp.


April
Temp. Biotemp.


May June
Temp. Biotemp. Temp. Biotemp.


20.6
20.6
20.4
20.2
20.0
19.9
20.5
22.1
24.5
26.1
27.1
27.8
28.'5
28.7
28.8
28.2
26.9
25.3
23.8
23.0
22.4
21.9
21.4
21.1


22.8
22.4
22.1
21.9
21.7
21,8
21.8
23.2
25.2
25.4 27.2
18.3 28.4
10.9 29.6
6.5 30.4
3.2 31..4
1.1 32.1
-1.8 32.0
2.3 31.5
11.8 30.1
20.1 28.6
?2'.3 27,1
26.3
25.5
24.8
5 2.1
21,5 23.4


20.3 25.1 17.1 26.4 15.6 27.1


20.1
19.9
19.9
19.7
19.5
19. 5
20.0
21.6
23.8
25.0
26 .0
26.0
27.0
27.2
27.3
26.7
25.4
23.5
22.3
21.7
21.2
20.7
20.5
20.3


25.0
24.3
21.2
17.8
15.1
14.7
18.7
23.3


23.8
23.4
23.1
22.9
22.7
22.9
24.6
26.6
23.1 28.3
18.7 29.2
8.5 30.3
1.5 30.9
-6.4 31.8
-7.5 32.5
-10.7 32.5
-7.5 31.6
5.6 30.1
19.4 28.7
24.6 27,3
26.8
26.2
25.7
23.7 24.9
23.0 24.3


21.5
21.2
21.1
20.9
20.8
20.8
22.0
24.1
26.3
27.5
28.8
29.5
30.1
30.4
30.6
29,9
28.4
26.7
25.2
24.5
23.8
23.2
22.5
21.9


23.5
21.9
19.4
15.5
14.9
13.6
13.2
20.4
23.4


23.9
23.7
23.5
23.3
23.2
23,4
24.6
26.2
27.
28.2
29.3
2).9
30.8
31.1
31.2
30.6
29.3
27.8
26.5
25.9
25.3
24.9
24.5
24.2


25.3
25.3
25.0
17.8
11.1
-0.2
-3.4
-3.8
3.1
11,4
22.4


21.9
15.3
6.2
-2.3
-8.9
-10.8
-12.7
-9. 4
6.6
18.4
26.4

25.8
25.2


Mean 22.7 20.6 23.7


15.3 26.6 18.8


- --------~----YII ___


_ --- -IW--^--LI I I _~








Table 5.- Cont'd.


July
Hour Temp. Biotemp.


24.2
24.0
23.9
23.7
23.5
23.7
24.8
26.2
27.3
28.1
29.0
29.5
30.1
30.5
30.5
30.0
29.0
27.5
26.4
25.8
25.3
21.9
24.6
24.14


26.5
25.9
20.7
15.5
4.3
-1.2
-3-4
3.3
19.4
26.6
26.0
25.3


August
Temp. Biotemp.


23.8
23.7
23.5
23.4
23.3
23.4
24.7
26.2
27.6
28.5
29.3
30.0
30.6
31.1
31 .3
30.7
29.6
27.9
26.3
25.5
24.9
24.4
24.1
23.8


24.1
26.2
27.1
21.7
13.2
4.5
-0.9
-3.5
-6.3
.16
. 10.6
25.6


Septc
Temp. B


23.8
23.6
23.4
23.3
23.1
23.2
24.2
25.8.
27.4
28.4
29.4
30.2
31.0
31.4
31.4
30.8
29 1
27.3
26.1
25.5
25.0
24.6
24.3
24.0


Degrees Centigrade
ember October
.iotemp. Temp. Biotemp.


24,9
18.3
15.3
2.1
-4.4
-5.0
-5.7
-2.9
13.7
24.8


22.7
22.5
22.3
22.1
22.0
22.1
23.3
24.9
26.7
2'7.7
28.7
29..1
29.6
29.7
29.5
28.9
27.4
25.8
24.7
24.0
23.2
22.9
22.8
22.5


25.5
18.
11,7
8.3
6,6
4.9
56,
8.8
18.2
23.8


No'venber
Temo. Biotenic.


21.4
21. 2
21.1
21.0
20.8
20.7
21.4
23.1
25.3
26.3
27,3
27.9
23. 3
28.3


26.0
24,4
23-5
22.9
22.5
22.0
21.7
21.4


24.4
25.0
19.7


12.9
13.6
18. 4
2;.6


December
Temp. Biotemp.


20.2
19.6
19.5
19.3
19.2
19.2
20.0
21.8
23.7
25.0
25.7
2 .3
26.7
26,8
26.5
25.9
24.4
22.8
21.9
21.3
20.9
20.6
20.3
26.2


24.1
23.1
20.6
19.1
17.2
19.5
22.8


Mean 26.5 20.1 26.6 18.1 26.5 17.5 25.2 18.9 23.9 20.9 22.4 20.9


/


(a)Calculated from hourly temperature recorded during the period July, 1962 through July, 1967.
(b)Biotemperatures equal to actual temperatures are not listed.


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


- --




- 38 -


inland from the lake.

The mean annual hourly marches of ambient and biotemperatures were

calculated and are graphically presented in Figure 4. The biotempera-

ture is equal to, or approximates, the ambient temperature between

the hours of 1900 and 0600 on consecutive days. During the morning

daylight hours, the biotemperature rapidly falls. Beginning between

1500 and 1600 hours, there is an equally rapid rise. The onset,

termination and magnitudehave been shown (Table 5) to vary with the

month of the year.


Precipitation


The Observatorio Nacional (196h) indicates an annual rainfall

of around 2000 mm around the central portion of the lake area, increas-

ing to 3000 mm at the western and eastern extremities. This magnitude

and pattern is in general agreement with observations recorded since

1961 along the northern shore of the lake. Although the available

data enable estimates to be made, Riehl (1954) and others suggest a

minimum of 40 years of continuous records is needed before a serious

rainfall analysis can be attempted.

Based on the data collected from 1961 through 1967 at Finca Los

Murcielagos, the average annual precipitation-is calculated to be

2004 mm. During most of this 6-year period, precipitation was recorded

using a Belfort recording intensity rain gage. During periods when

the Belfort instrument was not in operation, only daily or weekly

totals were recorded. The number of days (complete 24-hour records)

for which intensity records are available are presented in Appendix,

Table B by month and year. The precipitation data were reduced to




- 39 -


I
I
I


-- TEMPERATURE (A1MB3IE

-- -BIOTEMPERATURE


I

t

I






\


/
/
/
I
i
I
I
I
I
I

I
I
I
I
I
I

I
I
I
I


I
I
I
I


I


*\ I
I


2 3 4 5 6
AM


7 8 9 10 II 12 13 14 15 16
I
HOURS


I I I I I I I
18 19 20 21 22 23 241
PM I


Figure 4.-


Mean annual daily march of temperature and biotemperature
averaged from hourly temperatures recorded from July, 1962
through July, 1967 at Finca Los Murcielagos, Guatemala


w

LI-

Lii
0.
Li


1




- 40 -


hourly totals, put on punch cards and sumr.iarizcd on the IBM 360

computer. In all cases, simple means are used to express parametric

values.

The distribution of rainfall throughout the year is presented

in Table 6. Rainfall is generally seasonal in nature with the driest

period corresponding with the winter months and the wet season with

the summer months. February and July are the driest (35.6 mm) and

wettest (338.4 mmL) months, respectively. A characteristic of Central

American rainfall is a bi-modal maximum-rainfall peak in July and

September (Portig, 1965). August, though wet, has less precipitation

than either of the adjacent months. At Lake Izabal, this period coin-

cides with the caniculas (dog days), a distinct week to 10-day period

of dry, cloudless days. The maximum (4941.3 mm) and minimum (14.7 mm)

monthly totals occurred in July, 1961, and March, 1965, respectively.

Based on information obtained from the local population, it is suggested

that the long-term mean annual precipitation is probably higher than

the calculated 6-year mean of 2004 mm.

On the north side of Lake Izabal, starting at the delta of the

Rio Tunico (Figure 1), the rainfall increases both eastwardly and

westwardly approaching 3000 mm per year at around San Felipe and El

Estor, respectively (Observatorio Nacional, 1964). The central delta

portion, where Finca Los Murcielagos is located, appears to lie in a

rain shadow formed by the Sierra de Santa Cruz, which obstructs the

prevailing tradevinds. Much of the precipitation in this area is

derived from rain showers which originate over the Sierra de las Minas

and i ove northward. The higher mean precipitation toward the west is







Table 6.- Monthly precipitation at Finca Los Murcielagos, Guatemala


Precipitation (anm)
Month 1961 1962 1963 19bT 1963 1966 1957 Mean

January NR() 97.8 85.9 45.0 132.3 NR 116.3 95.5

February NR 46.2 52.1 20.8 24-1 NR 35.1 35.7

March NR 112.0 158.5 47.2 14.7 100.6 90.4 87.2

April NR 258.6 38.4 94.5 32.3 45.7 80.5 91.7

May NR 9.2 131.1 208.8 127.0 179.1 183,0 153.9
S-'
June 225.6 323.3 227.8 234.4 197.6 315.5 271.8 256.6

July 494.3 385.1 285.5 224.9 327.2 293.6 290.1 331-5

August 310.6 329.2 132.3 176.0 215.4 243.8 232.4 234.2

September 310.6 261.1 394.0 95.5 271.5 312.7 292.6 276.9

October 212.3 266.2 183.6 78.2 NR 204.2 295.4 206.7

November 101.4 172.2 121.2 230.4 NR 62.8 138.7 146.1

December 97.8 44.5 23.2 253.5 NR 39.1 66.8 88.3
Total 2390.4 1 .6 1729.2 2093.1 2004.3(b)
t(aNot recorded.
b) Calculated from monthly means.








due to orographic rains which originate in the lower portion of the

Polochic Valley. The eastern reaches of Lake Izabal are in the un-

obstructed path of the tradewinds. The monthly precipitation totals

recorded at Las Dantas, located about 6 km west of El Estor, are listed

in Table 7. It should be noted that these few data reflect only a

slightly higher annual precipitation than was recorded at Finca Los

Murcielagos during the years 1963, 1964 and 1967. The annual total is

also significantly. lower than estima-tes made for the northwestern

portion of the lake.

In general, temperatures and precipitation are related. The

higher temperatures occur toward the end of the dry season, reaching

a maximum in May prior to the beginning of the rainy season. During

the months of bigh rainfall, the temperatures tend to levcl off at

around 26.5 C, but begin a steady decrease in October.

Characteristics of the average rein-day are presented in Table 8.

For computational analysis, a rain-day is defined as one in which there

is at least 1.0 mm of recorded precipitation. These data are computed

as simple means based on the available 24-hour corrplete records

(Appendix, Table B).

On an annual basis, 45.5% of the days (166 days) experience at

least 1.0 mm of precipitation. On a monthly basis, this ranges from

a low of 21.5% (6 days) in March to a high of 84.5% (26 days) in July,

the months of lowest and highest total rainfall, respectively. There

is a general positive correlation between the per cent of rain-days

and the total monthly precipitation. The mean precipitation rain-day-1

shows a positive correlation, which also reflects monthly changes


- 42 -




- 43 -


Table 7.- Monthly precipitation at Las Dantas, Guatemala(a)


January

February

March

April

May

June

July

August

September

October

November

December


1963
mm

157.5

58.4

16o.o

48.3

50.8

137.2

491.5

208.3

233.7

281.9

83.8

15.2


1964
mm

22.9

15.2

54.6

40.6

215.9

368.3

542.3

181.6

208.3

135.9

115.6

139.7


1965
mm

177.8

46.2

0.0

42.7

264.2

424.4

326.9

299.0

351.5

385.6

211.8

80.3


1966
mm

32.5

53.3

62.2

23.4

172,7

396.7

292.1

254.0

500.4

227.3

52.6

17.8


1967(b)
mm

26.7

73.2

89.9

39.9

76.2

206.0

462.8

310.6

378.5

236.2

118.1

106.9


Total 1926.6 2040.9 2610,4 2085.0 2125.0 2157.7


Taken from weather records maintained by E
(b Nickel Corporation.
SInterpreted from consecutive 7-day totals..


xmibal, International.


Average
mm

83.5

49.3

73.3

39.0

156.0

306.5

423.1

250.7

334.5

253.4

116.4

72.0


m


--


"Y


U




- 44 -


Table 8.- Average rain-day characteristics, summarized by month, for
Finca Los Murcielagos, Guatemala


Month

January

February

March

April

May

June

July

August

September

October

November

December


Rain Rain Rain


days
(a)

35.4

24.7

21.5

22.8

34.2

68.0

84.5

-64.9

62.9

56.0

40.7

35.4


Characteristics of the average rain-day


mm.
(b)

8.23

5.02

13.44

13.13

14.07

15.24

15.62

11.22

13.79

12.34

10.74

9.17


% of rain between hours of:


0600- 1200- 1800-


hours 0000-
(c) 0600

4 42.65

4 47.47

4 33.78

4 26.36

4 41.85

4 27.21

4 40.58

4 30.17

4 35.08

4 22.81

4 29.37

4 26.95


Annual(d) 45.5 12.64

(a)Per cent of days in month
hour.
(b)Mean rainfall.
SCMean number of hours with
d'eighted means.


4 32.88 13.13 15.90 38.13

with a minimum of 1 mm precipitation in one


rain + 1 mm.


1200


18o0


2400


17.46

23.85

26.21

29.90

14.19

5.00

9.59

15.34

5.70

18.60

14.11

21.10


15.10

14.44

20.31

16.08

11.68

17.55

12.44

8.85

13.56

30.95

16.23

16.74


24.79

14.14

19.70

27.66

32.28

50.24

37.39

45.64

45.66

27.64

40.29

35.21


--


I-




- 45 -


in the intensity of individual showers. It is notable that there is a

mean of 4 hours of rain-day-1 during each month, a further indication

of the variation in intensity.

During the average rain-day, about 71.0o of the total precipita-

tion occurs at night between the hours of 1800 and 0600. During the

vet season, May through November, 72.3Th occurs at night, compared with

59.7% during the dry season, December through April.

The frequencies of hours with precipitation of specified amounts

are summarized by month for Finca Los Murcielagos (Table 9). These

are weighted to compensate for missing records and are expressed as a

5-year.rate. (It should be noted that these data do not estimate in-

dividual shower intensities.) In general, over 45% of the total pre-

cipitation occurs at less than 1.1 nm hr-' and over 93% occurs in

hours with less than 10.1 mm. There is a characteristic trend between

hourly.precipitation and monthly precipitation (i.e., season). The

maximum 1-hour rates occur during the peak of the wet season; a higher

proportion of the total rain occurs in 1-hour class intervals.


Life Zone Classification


Numerous attempts have been made in an effort to categorize and

classify units of vegetation based on one or more abiotic and/or biotic

parameters. The usefulness of any vegetation classification system is

measured by its simplicity, reproducibility among users and universal

applicability. In addition, the criteria used to distinguish catego-

ries should be capable of being readily understood by non-users. It

is the author's opinion that the system developed by Holdridge (1967)

best meets these requirements.




- 46 -


Table 9.- Frequency of hours with precipitation of specified amounts,
summarized by month to provide a 5-year ratela) for Finca Los
Murcielagos, Guatemala



mm rain
per one-hour Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total

0.1 1.0 64 51 54 49 59 127 156 114 121 120 75 69 1058

1.1 2.0 22 17 22 20 20 72 61 44 44 46 29 24 421

2.1 5.0 25 14 34 23 27 73 78 39 49 47 30 24 463

5.1 -10.0 7 3 14 18 15 48 31 24 27 22 13 9 231

10.1 15.0 .1 4 6 7 12 15 7 10 8 4 2 76

15.1 20.0 1 2 4 8 11 3 10 6 1 1 47

20.1 25.0 1 2 2 3 3 2 3 2 18

25.1 30.0 2 2 1 3 1 2 2 13

31.1 35.0 1 3 1 5

35.1 40.0 1 1 2 1 5

40.1 45.0 1 1

45.1 50.0 1 1 2

50.1 60.0 1 1 2

60.1 70.0

70.1 80.0

80.1 90.0

90.1 100.0 1


Total 120 '85 131 120 136 349 364 237 266 252 155 128 2343
(a)
Corrected for missing intensity data.




- 47 -


Holdridge's (1967) World Life Zones or Plant Formations are

defined by the mean annual values of temperature, precipitation and a

measure of atmospheric moisture, the potential evapotranspiration ratio.

Because the system allows the determination of any one of these three

parameters based on the other two, Life Zones can be determined from

annual temperature and precipitation data for a given area. Calcula-

tion and use of the biotemperature, theoretically take into account

the limiting effects of both low and high temperature on plant growth

processes.

The mean annual biotemperature at Finca Los Murcielagos is cal-

culated to be 18.7 C, using the raw data summarized in Tables 3 and 5.

This value, together vith the mean annual precipitation (2004 mm),

locates the study area in the transition between the Subtropical Moist

Forest and Subtropical Wet Forest Life Zone classifications. The term,

subtropical moist or vet, should not be confused with other definitions

or connotations. The suggested higher estimate for the mean annual pre-

cipitation would shift the exact location out of the transition wholly

into.the Subtropical Wet Forest Life Zone.

Holdridge (1967) has derived and explained a simple method for

calculating the potential evapotranspiration (PET) from mature forest

in a zonal climate with a zonal soil. He proposes the biotemperature

times 58.93 as an approximation of this theoretical parameter. Thus,

at'Finca Los Murcielagos, the so-calculated annual PET equals 1102 mm;

the amount of water which potentially can be lost by evaporation and

transpiration on an annual basis. The PET ratio, calculated as the

PET (1102 mm) divided by the annual precipitation (2004 mm), is 0.55.




- 48 -


Thus, at Finca Los Murcielagos, on an annual basis, an estimated 902

.mm, or 45% of the precipitation, is lost to runoff and subsurface

drainage, assuming a static balance in the soil and biomass. Runoff

occurs during the months of May through November.

J. Tosi (reported in Ewel and Madriz, 1968) has used the basic

parameters of Holdridge in devising a method for calculating the monthly

water balance. Evel and Snedaker (unpublished data) have used this

method to calculate the water balance for Finca Los Murcielagos. A

similar value for runoff wvs determined (913 mm) and occurs during

the months of May through November. The remaining months, previously

suggested to constitute the dry season, are not effectively dry (i.e.,

moisture limiting), with the possible exception of February. For

these calculations, an assumed field capacity of 200 mm was used.

The probable lack of an effective water deficit during the "dry

season" agrees with observations and descriptive information from local

inhabitants. Deciduous vegetation occurs primarily on the limestone

foothills and soils with high proportions of sand and gravel, as stated

previously. Three soil pits dug within a kilometer of the lake shore

in the study area in April, 1965, entered the water table at around

1.5 m. This evidence suggests that during months with a precipitation

deficit (i.e., a PET ratio greater than 1) the proximity of the lake

and the high-water table may preclude the occurrence of a critical

water deficit.











TRANSMITTED RADIATION IN VEGETATION


Statement and Methods


The solar radiation incident upon vegetation is important to

many diverse, but related, processes such as photosynthesis, evapo-

transpiration, and temperature maintenance. Vegetation and other

organisms in the lower strata below the canopy similarly respond to

radiant energy, but are relegated to an entirely different radiation

environment (Anderson, 1964a,b; Evans, reported in Bainbridge, Evans

and Rackham, 1966). The subcanopy radiation is a function of the re-

flection, absorption, and transmission characteristics of the overlying

canopy.

Reifsnyder and Lull (1965) and Vezina and Boulter (1966), citing

theirs and other vorks, estimate that dense forest vegetation in full

leaf absorbs 75 to 90 of the incident solar radiation. Conversely, 10

to 25% is reflected. Individual leaves absorb a much lower per cent, but

Miller (1955 and in Reifsnyder and Lull, 1965) suggests that the overall

structural organization of a forest serves as an effective system for

radiation capture. The proportion of the incident shortwave radiation

that is reflected is called the albedo. Published tables of the albedos

of natural and artificial surfaces indicate that vegetation, and forest

cover in particular, reflects relatively low proportions of the incident

radiation in both the total and visible spectrum (Geiger, 1965; Sellers,

1965; and Chang, 1968).


- 49 -




- 50 -


The spectral properties of photosynthetic surfaces are important

with respect to albedo, absorption and transmission. They not only

describe the surface characteristics, but also permit interpretation

of the radiation environment below the exposed surfaces.

S The Dutch Committee on Plant Irradiation (Wassink, 1953) defined

eight bands in the radiation spectrum in relation to plant physiological

responses. Five of these bands are considered here. The radiation in

Band 1, encompassing most of the infrared region between 4.00 and 1.00c,

which is absorbed, is converted directly into heat without directly

affecting biochemical processes. Radiation in Band 2, between 1.00 and

0.72-L, is chiefly responsible for the specific elongation of plants

and also contributes to the heat load. Band 3 radiation, 0.72 to 0.61A,

is strongly absorbed by chlorophyll and converted to chemical energy

by photosynthesis. Photoperiodism is strongly influenced by this

spectral region. The radiation between 0.61 and 0.51i<, Band 4, is

strongly reflected by the chlorophyll pigments resulting in a low

photosynthetic effectiveness. Band 5 defines the spectral region, 0.51

to 0.40A, which exhibits a high radiation absorption by the chlorophyll

and various yellow pigments. Photosynthesis in the blue-violet portion

Of the spectrum is strongest in this band. Physiological responses to

the intensity and spectral distribution of internal (subcanopy) radia-

tion have been discussed by Cummings (1963) and Daubenmire (1959).

A general consensus is found among research reports concerning

the reflection of incident radiation of varying wavelengths from dif-

ferent types of vegetation cover (Krinov, in Reifsnyder and Lull, 1965;

Gates, Tribbals and Kreith, 1965; Gates and Tantraporn, 1952; Yocum,




- 51 -


Allen and Lemon, 1961; Ewel, personal communication). Reflection is

lowest in the near ultraviolet and visible up to about 0.5/x. It is

higher between 0.5 and 0,6/U, coinciding with Band 4, but again drops

off approaching the near infrared at about 0.7j... At around 0.7jk,

the reflectivity increases dramatically and levels off at greater than

40,. When the reflection and transmission are subtracted from the

total incident radiation, the difference is assumed to be the absorption

(Yocum, Allen and Lemon, 1961). In general, reflection equals trans-

mission, except at shorter wavelengths where reflection may be somewhat

higher.

Published reports indicate a large variation in the spectral

transmission under various temperate forest types and atmospheric con-

ditions. Reasons for the variation are numerous. Anderson (1964b)

cites the difficulty of obtaining a representative sample under sunlight

versus diffuse light conditions, complicated by the presence of sunflecks.

Logan and Peterson (1964) mention differences in spatial distribution

of light quantity and quality resulting from differences in time and

season and stand characteristics. The presence and proportions of the

principal and accessory pigments also affect reflection and absorption

characteristics (Meyer, Anderson and Bohning, 1960). Selective reflec-

tion and absorption, as a function of leaf size and configuration

(Spurr, 1964), canopy structure (Miller, in Reifsnyder and Lull, 1965),

and crown closure (Vezina and Pech, 1964), have been cited as influencing

the quantity and quality of the transmission. Anderson's (1964a,b)

reviews and discussions cover many of the aspects of internal forest

radiation.




- 52 -


Previous attempts to estimate internal radiation intensity levels

in'lowland tropical forests employed the use of instruments differentially

sensitive to varying portions of the visible spectrum. Thus, comparisons

would be tenuous even if the measurements were made under similar

forest/time/latitudinal conditions. In spite of the reported inconsist-

encies in the literature, intensities reported in per cent of the external

radiation rarely exceed 1% (reviewed in Schulz, 1960). Values which

exceed 1% have included both "sunflecks" and radiation extending into

the near-infrared wavelengths (Schulz, 1960 and Evans, 1939). Fewer

attempts have been made to describe radiation intensity in specific

regions of the spectrum. The most notable of these for tropical forests

is the work of Evans (1939) in Southern Nigeria.

The gross incident radiation and spectral distribution between

0.4 and l.l1p were determined with an ISCO spectroradiometer on July

26, 1967. Four readings each were made in the clearing and nearby forest

(ca 32 years of age) between the hours of 1140 and 1250. During this

period, the sky was completely free of clouds or haze, the ambient

air temperature was 90 C and the relative humidity ranged between 76

and 82%. Readings were taken progressively at 0.025 A intervals during

each sequence and corrected to 1200 hours TST using the method describe.

by List (1966). These data, expressed in ly min-l1, are shown in

Figure 5. The incident radiation at 40 cm in the forest is also shown

in per cent of the external radiation.



Sunflecks refer to the unobstructed light penetrating through openings
in the overlying canopy.




- 53 -


Results and Discussion


Spectral Composition Under a Forest Canopy

The gross incident radiation (Figure 5) shows a higher intensity

in the visible range, 0.4 to 0.7.4, than in the near infrared. The

intensity spectrum is in general agreement with the results of other

researchers (Yocum, Allen and Lemon, 1961 and Lemon, 1964). The carbon

dioxide absorption band is particularly noticeable in the .9 to 1.0,u.

region. The lower intensity in the near infrared is due to absorption

by both water vapor and carbon dioxide (Sellers, 1965).

The transmission of radiation through the forest, expressed as a

per cent of external radiation, indicates almost complete absorption

and reflection (99.995%) in the 0.4 to 0.7A. band. In comparison, there

is a relatively high percentage (ca 13,) transmission in the near infra-

red. The higher transmission in the near infrared has been observed in

annual crops (Yocum, Allen and Lemon, 1961), temperate forests

(Baumagartner, in Reifsnyder and Lull, 1965) and tropical forest (Kline,

et al., 1967).

The actual spectral intensities at the low transmission wave-

lengths are shown in the bar graph in Figure 6. The spectral bands

defined by the Dutch Committee on Plant Irradiation (Wassink, 1953)

are shown below the graph. The greatest proportionalabsorption occurs

in Band 3, the region of strongest chlorophyll and carotene absorption.

The overall per cent transmission is apparently lower by 2 to 3 magni-

tudes in comparison to general light levels reported in the literature

(Richards, 1964; Reifsnyder and Lull, 1965 and Geiger, 1965). This,




- 54 -


TOTAL I' IC!C ::T


2.0-


.ADIAT0 -, 3/2/7, 1200 TST


.4;


.0


.8 .9
.3 .9


1I
1.0


P2RCE"V[~~~ TiAS S!> TH3O~ i OFr T


- .05-- ------


Figure 5.- Spectral distribution of the incident radiation between
0.5 and 1.1i-, and the per cent transmission through a
32-year-old forest at Finca Los Murcielagos, Guateinala





.0110-


.0100-


.0090-


.0080-


.0070-


.0060-


m'


t.


------ -----,
i
-I
I


.0050-


.0040-


.0030

.40 .45 .50 .55 .60 .65 .70


BAND #5


RAND #4


BAND #3


Figure 6.-


Spectral distribution of the transmitted radiation between 0.5 and 1.lj,
and in bands 5 to 3 (see text), in a 32-year-old forest at Finca Los
Murcielagos, Guatemala


-~ I \-- I


--- ---1
i
t.-- -



i
~I


I




. 56 -


in fact, is a result of measuring intensities at wavelength with half

bandwidths of 0.15 my between 0.4 and 0.7p.


Variation in Intensity Under Fallow Vegetation

The per cent of the incident solar radiation that is transmitted

through vegetation to the ground below is.a function of the kind and

amount of intervening vegetation. Presented in Table 10 are the trans-

mission values, in per cent of full sunlight, for fallow vegetation of

selected ages. These data were collected during periods of the summer

wet seasons of 1965, 1966 and 1967 using Gunn-Bellani radiation inte-

grators. The mean values are weighted according to the number of days

composing the monthly averages. The spectral transmission characteris-

tics of the glass-jacket surrounding the integrator are not precisely

known, and no data are available which correlate instrument response

with the spectral variation and intensity of the radiation environment

under vegetation (Shaw and McComb, 1959 and Vezina, 1961).

The per cent transmission values evidence a dramatic decrease in

radiation transmission by vegetation in fallows of 1, 2 and 3 years of

age, respectively. This agrees with the observed increase in foliar

density during these first years of second-growth vegetation develop-

ment. In vegetation of ages 6 to 14 years, the per cent transmission

appears to be higher than expected under-the apparent closed canopy

conditions of these fallows. The per cent value determined under forest

conditions agrees with literature values for closed canopy tropical

forest (Richards, 1964). For those fallow-ages where the transmitted

radiation was measured in consecutive months, May-June, June-July and

July-August, the average transmission in the second month is consistently




- 57 -


Table 10.- Radiation at soil surface under fallow vegetation and
forest expressed as a per cent of total incident solar
radiation(a)


(Per cent)


Fallow Ages
(Years)


Site


May June


July August (weighted)


36.9 33.3


9.7


35.3


9.3


5.3

7.4


6.8

7.6


1C. 0

3.7


9.5

5.4

7.4

6.8

7.6


10.0


measuredd with Gunn-Bellani Radiation Integrators,


Forest


I


----





. 58 -


lower then the first. This conforms with observations on the local

phenology. The latter part of May and the months of June are charac-

terized by a flush of new growth which continues during the following

months. This rapid increase in May-June is particularly evident as

many of the component species, especially in older fallows, are decid-

uous during the winter dry season.

Site #5 was studied in consecutive years, 1965 and 1966, when

the vegetation was 2 and 3-years-old, respectively. During both periods,

the radiation integrator was located in the same prepared hole. The

reduction in per cent transmission is considered to be the result of

changes in the flora and physiognomy which resulted in an apparent

decrease in the number, size and duration of sunflecks.

Transmission was measured at site #3 during the month of July

and at the same location when the site was 8-years-old in 1965 and

10-years-old in 1967. The increase in the per cent transmission suggests

an opening of the overlying vegetation canopy which continues through

the 14th year of succession.













PRECIPITATION THROUGHFALL


Statement and Methods

The energy balance of an ecosystem is closely tied to the

hydrologic cycle through evaporation and transpiration, The primary

input into the hydrologic cycle of terrestrial ecosystems is generally

in the form of precipitation. Losses from the system are by

evaporation, transpiration, surface run-off and subsurface drainage.

Hydrologists have determined the general magnitudes and Inter-

relationships of these factors in temperate terrestrial ecosystems,

but by comparison, little is known about tropical-forest hydrology

except that within broad limits the processes are the same.

One of the factors related to the hydrologic cycle, which has

undergone considerable study in temperate areas, is the ratio of

precipitation Interception versus throughfall under varying forest

conditions. The earlier literature is reviewed in Kittredge (1948),

though Helvey and Patric (1965) suggest that many of these studies

suffered from limitations in the sampling or measurement techniques.

For the purposes of this study, precipitation throughfall is

that portion of the gross precipitation which falls freely to the

litter or soil surface through openings in the canopy and as leaf

drip. The remaining portion is intercepted by tire vegetation. By

this definition, stemflow, or the portion channeled downward along

stem, vines, etc., is necessarily included in the interception term.


- 59 -








Some intercepted precipitation may be lost immediately through

evaporation or temporarily stored in the canopy subsequent to later

evaporation. Evaporation of wvter directly from the intercepting aer-

ial vegetation does not represent a net loss in the water economy as

the resulting increased water content of the air and the reduced satura-

tion deficit decreases transpiration and the evaporation of previously

existing free water. Direct uptake into plant tissues has been suggest-

ed (Stone and Schachori, 1954; Stone, et al., 1956; and Slatyer, 1956)

though it is probably quite small in terms of annual rainfall. The

bromeliads and other groups which similarly collect water in leaf

axils probably take up a significant amount.

Based on a considerable amount of data. in the literature, Rahkmanov

(1962) has arrived at a reasonable estimate of the per cent precipita-

tion intercepted by broad-leaved forests in temperate latitudes. He

suggests an annual mean of between 20 and 22%.

In general, similar data for tropical forest indicate a higher

percentage is intercepted by this type of vegetation. Vaughan and Wiehe

(1947) report 30.5% interception in the upland climax forest of Mauritius.

This agrees with the figure of 35.8% by tropical forest in Uganda given

by Hopkins (1960). Somewhat higher values of 58.4h (Odum, in Kline,

et al., 1967) and 57% (Clegg, 1963) were determined for the "Rain

Forest" on Mt. El Yunque, Puerto Rico. The values that Freise (in

Richards, 1952) obtained in "Tropical Rain Forest" in Brazil, varying

from 65 to 70%, have been critically cited by Richards (1952) as

being at variance with the results of both temperature and tropical

studies. Based on data obtained in the Darien, Panama, Golley,


- 6o -




- 61 -


McGinnis and Clements (1968), made the assumption that 15% of the first

3.8 mm of rain during each storm day is intercepted by tropical forest

and lost by evaporation; the rest reaches the forest floor. Then, using

the average rainfall data from 3 locations covering a 221-day period

from May 2 to December 31, 1967, a 17% interception value was estimated.

The majority of rain-days (67.05%) had precipitation of less than 3.8 mm.

In comparison with temperate studies, these few data preclude the

estimation of a mean interception percentage for tropical forest, such

as presented by Rahkmanov (1962) or Helvey and Patric (1965). In addi-

tion to the normal variation, errors are possibly associated with sampling

under a single tree (Odum, in Kline, et al., 1967), applicability of

using base data from a distant control point (Vaughan and Wiehe, 1947),

location of the control gage in the turbulent air layer next to the can-

opy (Freise, in Richards, 1952), and insufficient direct observations

(Golley, McGinnis and Clements, 1968).

The ratio of interception to throughfall has also been consistently

shown to be a function of the precipitation intensity and duration.

Reports of tropical forest studies relating the ratio to intensity are

inconclusive and, at times, conflicting.

During the vet months of June, July and August in 1965 and

1966, efforts were made to determine the per cent throughfall in

mature-secondary forest and successional fallows of ages 6, 8 and

14 years within the study area at Finca Los Murcielagos. Through-

fall was collected in rain gages at 5 and 6 locations within each site

for periods extending up to 8 weeks. The gages were checked daily

except during those periods when there was no recordable precipitation.




- 62 -


Two types of rain gages were used: Tru-View direct-reading gages, and

standard weather-bureau type catch basins. The orifice sizes were 58

and 184 cm2, respectively, and according to Huff (1955), differences

in size should not result in a loss of accuracy. Readings made with

these gages were then compared with the continuous intensity records

of the Belfort recording intensity rain gage for the intervals between

checks. As both a greater number of showers and a greater percentage

of the total amount of rain falls at night in the study area, the

throughfall gages were checked as early in the morning as possible

to reduce errors resulting from direct evaporation from the gages.

On several occasions when rain would fall throughout the day, the

gages would be checked at the end of the rainy period when all leaf

drip had ceased. This technique was followed to ensure accurate com-

parisons with the intensity record in the clearing.

The gages were laid out in a mechanical pattern at ground level,

and in the case of the Tru-View gages, on short posts to hold them

erect. Once the sampling pattern was established, the gages were left

in position throughout the sampling season and, in the event of repeat-

ed annual observations, the gages were returned to the same points.


Results and Discussion


Factors Affecting Throughfall

Personal observations and published reports suggested that

throughfall may be influenced by, at least, 4 independent precipita-

tion-related variables. These are: (A) total rain hours; (B) hours

of drying weather between showers; (C) maximum one-hour intensity; and

(D) total precipitation. The values for each of these parameters were




- 63 -


determined from the intensity records for each period between con-

secutive throughfall-gage checks. Rain hours (A) are those with at

least 0.1 mm precipitation. Likewise, (B) is the total number of

hours without rain during each period, occurring between the hours

of 0800 and 2000. This parameter is used as an estimate of the

saturation capacity (Fenman, 1963) of the intercepting vegetation.

The tacit assumption is that during the wet season the amount of

water retained by, and evaporated from, the intercepting vegetation

is related to the period of drying between discrete showers. The

maximum one-hour intensity (C) was calculated as the maximum amount

of precipitation that occurred within any 60-minute period. The

total precipitation (D) represents the total amount recorded between

throughfall-gage checks.

For each age-site, multiple regression equations relating the

observed throughfall with the variables A through D were estimated.

Computation was performed on the University of Florida IBM 360

computer using a stepwise regression program.7 In addition to the

linear terms, the quadratics and first-order interactions were entered

as independent variables. The regression equations were determined

in such a way that only those variables which make the greatest reduc-

tion in the error sum of squares are included in the equation. For

use as a measure of the goodness-of-fit of the equations, the coeffi-

cient of multiple correlation (R) was calculated. The closer the value

100 R2 is to 100, the better the fit of the equation.



BMD02R Stepwise Regression Version of May 2, 1966. Health
Sciences Computing Facility, U. C. L. A.







It should be noted that the multiple regression analysis was

not used with the intent of creating a predictive model. Instead,

it is employed as a tool to aid in the interpretation of causes for

throughfall variation among the four sites. Predictive models for

2 sites (14 year old and mature secondary forest) were calculated

for comparison using the linear regression of precipitation through-

fall on total precipitation. The mature secondary forest model is

compared with a similar model for eastern United States hardwood

forests reported by Helvey and Patric (1965).

The relationships between precipitation throughfall and each

of the independent variables are shown in Table 11. The variable,

hours of drying weather between showers (B), was found to be consist-

ently insignificant in reducing the errorsum of squares in each analysis,

and therefore, does not appear in the equations. For each equation,

there is a high R value indicating a relatively good fit of the data.

As each R value is greater than 0.8, a lower limit suggested by

Snedecor (1964), it may be assumed then, that the included variables

account for most of the observed variation in throughfall among sites.

The relationship of throughfall in 6-year-old vegetation to the

three precipitation variables is described by equation (1) in Table 11.

The most important term in the equation is the interaction A*C (total

rain hours x maximum one-hour intensity), which accounts for over 821

of the observed variation. The extremely high linear correlation

(.954) between maximum one-hour intensity (C) and total precipitation

(D) however, suggests that the relationship may be fortuitous (Snedecor,

1964). In equations (2) and (4), (D) total precipitation, is the most


- D4 -









Table 11.-


Regression equations of precipitation throughfall on 3 precipitation-related variables (dura-
tion, intensity and amount) for 4 forest fallows of known ages at Finca Los Murcielagos,
Guatemala


Fallow ages
(years)

6


Coefficient of
Sdetermination(a)

0.9074


0.8831


0.9007


0.9535.


Regression equation

4(b) = 1.796 0.890A(c)+ 0.127AC(d) + 0.735D(e) 0.O15CD


^ = -4.125 + 1.073A + 1.283C 0.017C2 + 0.316D


Y = -2.364 1.734A 0.292A2 4 0.668D 0.004D2 0.088AD

A 0.212 .63D
Y = 0.212 + o.636D + 0.011o D


(a)Per cent of the variation of Y attributable to the regression equation.

(b)Estimate of precipitation throughfall in mm for a 24-hour period.

(c)Number of hours in sample period with precipitation greater than 0.1 mm.

(d)Maximum 1-hour precipitation during sample period.

(e)Total precipitation in mm for sample period.


Equation

(1)





- 66 -


important variable, and in each case exhibits a low linear correlation

with (C), the maximum one-hour intensity.

Equation (2), describing throughfall in 8-year-old vegetation,

also includes the three independent variables, and (D), total precipita-

tion, accounts for 71% of the variation. In addition to (D), both the

duration (A) and intensity (C) tend to increase throughfall, but at

lower rates for increasing values of (C).

Throughfall at the 14-year-old site is described by equation (3),

in which total precipitation (D) is responsible for 87% of the varia-

tion. The inclusion of just two variables permits the determination
A
of a family of curves (Figure 7). In this procedure, Y is estimated

for total precipitation (D) values of from 5 to 50 mm, while duration

(A) is held constant at 1, 2, 3 and 5 hours. Thus, for each number of

hours of rain, throughfall increases with total precipitation (D), but

the rate of increase becomes slightly smaller for the larger values of (D).

Precipitation throughfall at the forest site is almost entirely

a function (equation 4) of tha total precipitation. This variable, (D),

accounts for over 94% of the observed variation. Shown in Figure 8,

is a family of curves describing the relationship between intensity

and amount on the throughfall. In this example, the maximum one-hour
A
intensity (C) is held constant at 1, 10, 20 and 30 mm/hr and Y is

solved for varying amounts of total precipitation. In general, through-

fall increases linearly with increases in total precipitation (D) for

a given maximum one-hour-intensity. The slope of the lines increases

with higher intensities.




- 67.-


50-


45-


40-


35-


E
E 30-


-J
^ 25-


S20-
2:
n-
-r

15 -


10-


5-


0-
(


Figure 7.-


I I I I I


10 15 20 25 30 35 40 45 50

D-TOTAL PRECIPITATION (mm/24hrs)


Relationship between precipitation throughfall and
(A) hours with precipitation, and (D) total precip-
itation in a 14-year-old fallow at Finca Los Mur-
cielagos, Guatemala


I I
O 5















50-


45


40-


- 35-
E C = 30 mm/hr
-J 30-
-J

C 25- C=20mrn/hr
0

- 20-


15-


10- C= 10 rnm/hr


5-

C= 1mm/hr
0- 1 I I I Il I I II---
0 5 10 15 20 25 30 35 40 45 50
D-TOTAL PRECIPITATION (mm/24hrs)




Figure 8.- Relationship between precipitation throughfall and
(D) maximum one-hour intensity, and (D) total pre-
cipitation in a 32-year-old forest at Finca Los
Murcielagos, Guatemala




- 69 -


Predictive Models Based on Total Precipitation

With the exception of the anomolous relationship (equatio:i 1)

at.the 6-year-old site, it is apparent that with increasing age of

the vegetation, the per cent of the variation accounted for by total

precipitation (D) increases from 71% at 8 years to 87% at 14 years

to over 94% at the forest site. These latter two sites were selected

to serve as comparative models because of their larger sample sizes,

higher R values for (D) and closer similarities in age with other

forest studies reported in the literature. The models are based on

the linear regression of throughfall on total precipitation.

In order to place interception and throughfall on a comparable

basis, throughfall was converted to per cent of the total precipita-

tion (D). The difference between the per cent throughfall and 100%

is considered to be interception, irregardless of its subsequent fate.

The model for the 14-year-old fallow, or what may be considered

early secondary forest, is best expressed by the regression equation,
A
Y = 100 o(.98D 1.628)/D (5)

where Y is the per cent throughfall and (D) is the total precipita-

tion during an approximate 24-hour period. The model for a late

secondary forest (ca 32 years old) is

Y% = 100 0.859D 0.980)/D (6)

The curves derived from these two equations are shown in Figure 9. In

each example, there is a rapid change in the per cent throughfall for

total precipitation amounts between zero and 15 mm. For increasingly

larger values of (D), the rate of increase in throughfall diminishes.

There is little difference in throughfall between the early and late

secondary forests up to about 5 mm. For precipitation totals greater





- 70 -


100-

90-- -
90- /

so--

80-


14-year-old fallow
60 Y 100 o .984D 1.628)/D]

SA_ ^32-year-old forest
50- A -'
SO- /Y =100 ,0.859D 0.988)/Dj
0
S40-

,- I
W 30- 1
w (

- 20-

I
10-



0 5 10 15 20 25 30 35 40 45 50
TOTAL PRECIPITATION (mmr/24hrs)



Figure 9.- Regression of precipitation throughfall on (D) total
precipitation (expressed in per cent) for a 14-year-
old fallow and a 32-year-old forest at Finca Los
Murcielagos, Guatemala





* 71 -


than 15 mm, the early secondary forest exhibits about 9% more through-

fail.

The investigations of throughfall indicate that the amount of

precipitation is the most important parameter associated with the

quantity of rain falling through the canopy. The association is weak-

est when the total is small, thus allowing intensity and duration to

have greater influence. The association is also better defined in

older-age vegetation (14 and ca 32 years) than in younger fallows (6

and 8 years).

The 6-year-old fallow study produced results which can not be

effectively interpreted. This is thought to be, in part, a result of

the structure of the overlying canopy. An abundance of subcanopy vines,

standing-dead stems, and large, succulent understory leaves create

numerous flow paths for intercepted rainfall. Together with large

variations in leaf-drip and free-fall patterns, it would be reasonable

to assume that six collecting devices does not represent an adequate

sampling technique.

Similarily, the greater uniformity in results from the older

sites suggests a more homogeneous canopy structure and fewer subcanopy

deflecting surfaces. It is interesting to note that not only is through-

fall greater at the 14-year-old site than in the forest, but that

ground level radiation is also higher; 10.0% of incident solar radia-

tion versus 3.9% (Table 10), respectively. The comparisons indicate

a more open canopy in the younger-aged site. It is, however, impossible

to say whether this is a function of vegetation age or site quality.




S- 72 -


In 1965, Helvey and Patric reported the results-of their efforts

td combine, through regression analysis, all available data on through-

fall in the hardwood forests of the eastern United States. For compar-

ative purposes, their regression equation for summer rains was convert-

ed to,

Y% 100 [(0.901D 0.03:L)/] (7)

where Y is the per cent throughfall and D is the total precipitation

per "storm." Equations (6), late secondary forest, Guatemala, and (7),

hardwood forests, United States, are plotted in figure 10. If the

variation due to sampling technique is assumed negligible, then the

tropical forest permits less throughfall, particularly for values of

(D), total precipitation, less than 15 rmm. Also, differences in the

two curves suggest the tropical forest has a greater canopy saturation

capacity (20 mm versus <5 ma) and a higher stem flow once that capac-

ity is filled.





- 73 -


I






Eastern U. S. Forest
Y =.100 0.901D 0.031)/j

Guatemslan Lowland Forest
Y 100 o(0.859D 0.988)/D


I I I I I I
5 10 15 20 25 30


I I I
35 40 45 50


TOTAL PRECIPITATION (mm)

aEquation from Helvey and Patric (1965) converted to per cent.


Figure 10.-


Comparison of per cent throughfall versus total
precipitation during growing season for mature
mixed hardwood forests in the eastern United
States and eastern lowland Guatemala


100-


80-


70-


J 60-
Ii
0
50-
o

40-
z
w
S30
0.

20-


I0-













BIOMASS AND ELEMENT INITINTORY


Statement and Methods


One of the more obvious changes that occurs with maturation of

a tropical forest ecosystem is the hyperbolic increase in above-ground

standing crop. This is assumed to be of great importance in agricul-

ture because during the subsequent oxidation and decomposition of the

cleared pre-crop fallow vegetation, nutrient elements become available

to the potential benefit of the planted crop species; more abundant

vegetation should contribute a greater quantity of nutrients. Various

efforts have been made to assess the amount and element composition

of fallow vegetation (cited in Table 12), and element compositions of

fallow species (Bartholomew, Meyer and Laudelot, 1953; Nye, 1958; and

Snedaker and Gamble, 1969).

Presented in Table 12 is a partial summarization of some of the

major investigations on standing crop of tropical fallow and forest

vegetation. It is difficult to draw age-dependent conclusions from

the total biomass data due to regional variations and differences in

sampling techniques among investigators. Nevertheless, certain gross

trends are evident. In general, total biomass increases with the age

of the fallow. The values reported by Bartholomew, Meyer and Laudelot

(1953) for 2-,5-,b-and 18-year-old fallows in the Central Congo Basin

represent major deviations in a suggested age-biomass correlation.

Considered separately, the net biomass increase over time is evident,


- 74 -









Table 12. Biomass standing crop and leaf compartment macro-element inventories in selected tropical
lowland fallows and mature forests


Vegetation
ago
(rs) Location


1 Izabal,
Guatemala

1 Izabal,
Guatemala

2 Belgian
Congo

2 Darien,
Panama

2 Darien,
Panama

2 Guarin,
Colombnia

4 Guarin,
Colombia

4 Darien,
Panama

5 Belgian
Congo


Vegetation
type


Hel iconla
fallow

Mixed spp.
fallow

Mixed spp.
fallow

Mixed spo.
fallow(d)

Mixed spp.
fallow(e)

Mixed spp.
fallow

Mixed spp.


Mixed spp.
fallow(

Mixed spp.
fallow


Total
biomass(a) Leaf
K(fms m"2) BRiomass


779


874


1323


1302


2436


1534


4839


3794


7668


compartrmnt (gms m-2)
j K Ca ,


779(b) 9.1


874(c) 11.7


0.5 14.2 2.6 6.8


0.9 10.0 5.5 6.0


787 10.5 1.2 10.2 ( 7.7 )


- 0.4 4.2 15.7 1.3


296


260


495


594


Reference

Popenoe (unpublished)


Tergas (1965)


Bartholomew, et al.
(1953)

Golley, et al.,
(1968)


- 0.3 5.0 3.5 1.3 Golley, et al.,
(196)

Ewel (in Gamble, et
al., 1968)

Ewel (in Gamble, et
al., 1968)


- 0.7 8.3 8.0 1.8


563 12.5 0.7 7.9 ( 7,8 )


Golley, et al., (1968)


Bartholomew, at al,,
(1953)









Table 12. cont'd.


Vegetation
ago
(vrs) Location


6


6


8


ca 18


ca 20


40-50


(h)


(h)


(h)


(h)


Darien,
Panama

Benin, South
Nigeria

Belgian
Congo

Belgian
Congo

Kumasi ,
Ghana

Kade,
Ghana

Massa Me ,
Ivory Coast

Okomu,
Nigeria

Darien,
Panama

Darien,
Panama


Total
biomass(a)


Leaf compartment (ams m-2)


(rmns m"2) liomass


Vegetation
type

Mixed spp.
fallow(d)

Acioa barteri
fallow

Mixed spp.
fallow1

Mixed spp.
fallow

Mixed spp.
fallow

Mixed spp.
fallow

Mixed spp.
forest

Mixed spp.
forest

Mixed spp.
forest(d

Mixed spp.
forest(e)


4245


4609


12168


12108


9366


26582


20000(


22700(j)


37053


26351


N P IK Ca _


Reference


655


419(f


538


644


446


2025(g)


1000(


1200(j)


1217


794


0.7 7.8 17,2 2.5 Golley, et al., (1968)


5.2 0.4 2.8 2.9 2.8 Nye and Greenland
(1960)

12.0 0.7 7.9 ( 8.7 ) Bartholomew, _t al.,
(1953)

14.3 0.8 8.0 ( 7.6 ) Bartholomew,'ot al.,
(1953)

11,2 0.6 3.8 6.9 2.1 Nye and Greenland
(1960)

38.1 25.4 15.4 38.5 5.3 Greenland and Kowal
(1960)

S Ogawa, et al., (1961)


Ogawa, et al., (1961)


1.2 15.5 19.2 4.8 Golley, et al., (1968)


0.3 11.5 17.3 1.7 Golley, et al., (1968)









Table 12. cont'd.


Vegetation
age
(vrs)


Total
Vegetation biomass(a)
Location type (pms m-2)

Northwestern Gallery forest 29060(k)
Thailand (immature)

Darien, Gallery forest 117658
Panama Priorla
copalfera


Leaf compartment (gms m"2)
Biomass N P K Ca M_ i


1950(k) 51.8 -


1218


Reference


- Ogawa, et al., (1961)


- 0.7 17.2 14.8 3.8


Golley, et al., (1968)


(a)Above-ground biomass excluding litter.
(b)Includes an estimated 16% (125 gms m-2) wood,
(C)Includes an estimated 16% (140 gms m"2) wood and 87 gms litter (Ewel, 1968).
(d)Determined:during wet season.
(e)Determined during dry season.
()Includes twigs.
(P)Includes stemwood up to 5 cm in diameter.
(h)Natural forest not considered to have arisen from an agricultural fallow.
(i)Allometric determinations by Ogawa, et Ea.. (1961) applied to physiognomic date in Aubreville (1938).
(j)Allometric determination by Ogawa, Ct al., (1961) applied to physiogncmic data in Jones (1956).
(k)Allometric determination by Ogawa, et al., (1961).




- 78 -


except for the similarity between the 8- and 18-year-old fallow vegeta-

tion. The dry-season biomass for 2-year-old fallow in the Darien,

Panama (Golley, McGinnis and Clement, 1968) is over 53% greater than

the vet-season biomass. As stated, it is probably a better estimate

of a 3-year-old fallow if 1-year intervals are used. The value reported

by Ewel (in Gamble, Snedaker, et al., 1968) for a 4-year fallow near

Guarin, Colombia, is thought to be an overestimation as the result of

the inclusion of an atypically large tree (Ochrcma sp.).

The total biomass in the oldervaged forest examples ranges be-

tween 20 x 103 and 118 x 103 gms m-2. Ogawa, Yoda and Kira (1961)

suggest that the plant biomass of well-grown forests of the world,

irrespective of climatic zone, ranges between 20 x 103 and 35 x 103

gms m-2. The gallery forest in the Darien, Panama, is about three

times the suggested upper limit. Ovington (1962) cites other examples

which exceed the limits of Ogawa, Yoda and Kira (1961), although none

approach the value for the gallery forest in Panama.

Whereas the total above-ground biomass increases over a relative-

ly long period of time to the environmental carrying capacity, the

component leaf biomass quickly reaches a high level and appears then

to remain at a steady-state level to over 20 years of age (Table 12).

The mean + 1 std dev dry-weight leaf biomass for these fallows is 526

+ 137 gis m-2. The older tropical forest values are higher (1230 gms

m-2), but the questionable validity of including allcmetrically determined

values and gallery forest examples precludes any definitive conclusions.

Temperate hardwood and conifer forest biomass studies have been

reviewed by Ovington (1962). By using his data for 19 temperate hard-

wood forests, a value of 286 + 20 gms m-2 for the leaf compartment




. 79 -


(dry weight) was calculated. These forests, both natural and planted,

range inage from 20 to 200 years. This value is considerably lower

than for either of the two tropical vegetation types and evidences

no change with age. It is interesting to note, however, that the 23

temperate conifer forest examples, ranging in age from 16 to 93 years,

have a calculated mean of 931 + 158 gnms m-2 leaves and indicate a

possible positive linear correlation with age.

The leaf compartment is of additional interest because it is a

major contributor of organic matter and nutrient elements to the litter

compartment. Other components of vegetative litter fall include fruits,

flowers, stems, twigs, vines, wood, etc. Evel (personal communication)

has shown that the annual litter production in second-growth fallows

of ages 1 to 14 years at Finca Los Murcielagos is a function of the

fallow vegetation age. The relationship is significant at the 1%

level and is best expressed by the equation,
A
Y 5.698 + 0.643X 0.015X2 (8)

where Y is the estimate of the annual litter production in T ha-1

and X is the age in years of the fallow vegetation. It is suggested

that this equation underestimates actual annual production due to de-

composition of litter in the collection baskets during intervals be-

tween checks (Ewel, 1968).

Equation (8) indicates an annual litter production of about 400

gms m-2 during the first year of fallow succession, which increases

to a calculated maximum of 1060 gms m-2 at about 21 years of age and

then decreases. Ewel (1968) assumed the annual litter production.of

900 gms m-2 in the forest represents a steady-state level and concluded









the minimum age for the forest to be 32 years. If the leaf compartment

biomass is indeed constant, or nearly so, over the first 10 years of suc-

cessional regrowth, the Ewel's data indicates an increasing gross pro-

duction and turnover rate.

The major objective of the successional biomass studies at Finca

Los Murcielagos was to characterize changes over time in three above-

ground biomass compartments and in the compositional element inventor-

ies of the leaf compartment. For expediency in relating the leaf com-

partment to litter production, the former was defined to include fruits

and flowers, small twigs, and herbaceous vines;that portion of the

living biomass most likely to become a litter compartment input within

one year. Fruits and flowers form a very small proportion of the total

biomass. Other studies have shown it to be less than 2.5% of the leaf

compartment biomass in tropical fallow vegetation (Golley, McGinnis and

Clements, 1968; Ewel, in Gamble, Snedaker, et al., 1968). Small twigs

include buds, petioles, the rachis of compound leaves and the smallest

twigs of less than about 0.5 cm possessing leaves. It is doubtful

under normal conditions that these components, including herbaceous

vines, exceed 10% of the successional leaf biomass on a dry-weight basis.

Odum (1960) defines litter to include standing dead plants. In

the two previous studies at Finca Los Murcielagos, Tergas (1965)

included both surface litter and standing dead plants in the above-

ground biomass, whereas Ewel (1968) did not include the latter in de-

terminations of the standing crop of litter. In this study, the

standing dead component was considered to be a distinct transition

compartment, or pathway, linking the living biomass to the litter




- 81 -


compartment. Therefore, it vas not included in the biomass sampling.

Biomass standing-crop harvests were made on 143 different 9 m2

ample plots, representing fallow vegetation ages of 1- through 10-years.

A minimum of 4 plots were harvested and sampled at each age-site. From

a randomly selected starting point, four 3 x 3 m plots were delineated

mechanically and additional plots were similarly laid out as needed.

All plot corners were marked with stakes and the vegetation cleared

away from around plot perimeters. The aggregate of all plants rooted

within the plot was considered to constitute the sample. Vines, form-

ing a continuous horizontal strata in the canopy, were cut where they

intersected the vertical planes of the plot boundaries.

Prior to. the harvest sampling, a cursory floristic inventory8

was made on the sample plots. This consisted of determining the rela-

tive frequency of occurrence of species within a block of 4 plots.

Identification of species was made in the field using pre-assigned

code names and tentative identifications. Positive identifications

have since been corroborated for many of the species from prepared

herbarium material handled through the University of Florida Herbarium.

Uncorroborated identifications are so indicated. The frequency data

vere combined with other observations on species presence in various

fallow ages.

Each plot was harvested by hand using local labor instructed in

the correct separation of wood and leaf compartments. Laborers were

personally supervised during each plot harvest. Wood and leaf



Seedlings were ignored due to the difficulties involved in identifi-
cation.




- 82 -


compartment material were respectively placed in tared coffee bags,

weighed on a Chantilion milk scale and the vet weights recorded. Sub-

samples of the wood compartment were retained for the determination of

moisture content. The leaf compartment material was finely chopped

with machetes, thoroughly mixed and a 2 kg grab-sample taken for sub-

sequent moisture and element analyses. Both the wood and leaf sub-

samples were respectively placed in tared cloth bags and weighed on a

triple-beam balance. All steps between the initial harvesting and

final sub-sample weighing were accomplished as quickly as possible to

avoid significant moisture losses. To prevent deterioration prior to

arrival at Florida, all sub-samples were air dried, weather permitting,

or oven dried at 40 to 50 C. Samples were subjected to fumigation

with CH3Br upon entry into the United States.

Composite soil samples, representing five cores each, were taken

from depths of 0-5, 5-20 and 20-40 cm in each plot as part of a sepa-

rate study partially reported elsewhere (Gamble, Snedaker, et al.,

1968, and Snedaker and Gamble, 1969). After each plot harvest, the

remaining plant material was evenly redistributed over the plot.

A partial element analysis of the leaf compartment material

from 35 sample plots, representing fallow vegetation of 7 ages, was

made in the Tropical Soils and Central Analytical Laboratories at the

University of Florida. Quality control measures included the use of

selected referee samples and repetitive analysis by the laboratories.

Plant sub-samples were oven dried at 70 C to constant weight, and

moisture content percentages were used to correct the biomass standing crop

weights to a dry-weight basis. Prior to analysis, each sub-sample was




- 83 -


ground in a Wiley mill until it passed through a number 20 mesh brass

screen,and then stored in liquid-tight paper food containers.

Selected sub-samples were analyzed for N using the Jackson (1962)

Kjeldahl method. Ammonia produced in the digestion was distilled into

a 4% H13B03 solution with a mixed indicator of bromcresol green and

methyl red. Standard HCl acid was used in the titration.

Plant tissue extracts were prepared by ashing at 500 C and adding

an excess of 25% HNO3. After evaporating the acid and re-ashing, an

excess of concentrated HCl was added and the solution'evaporated to

dryness. The final product was taken up in 10 N HC1, filtered through

a number 42 nhatman filter paper and diluted to 100 ml, yielding a 0.1

N HC1 solution. The extract was stored in glass bottles preparatory to

analysis.

P was determined in the extract using the Truog (1930) method

involving a (NH4)2MoO4 H2S04 solution and SnCl2. Analysis was made

on a Bausch and Lomb Spectronic 20 colorimeter with a red filter, IP

40 phototube and a wavelength of 700 my.

K was determined with a Beckman B Flame Spectrophotometer, using

a propane-air flame, at 771 my. The elements, Ca, Sr, Mg, Mn, Fe, Cu,

Zn and Al,were determined by atomic absorption (Perkin Elmer) follow-

ing the recommended procedures.


Results and Discussion


Floristic Composition of Fallow Vegetation

Presented in Table 13 is a relative frequency listing of the plant

species comprising 4 subjectively defined age classes of vegetation.




-84 -


Table 13. Relative frequency of occurrence of selected plant
species in vegetation of 4 general age classes at
Finca Los Murclelagos, Guatemala


Relative frequency of occurrence(a)
Scientific nare(b) 1-2 yrs. 3-6 rs. 7+ vrs Forest(c)

Abutilon sp. (F) 1 2 0 0
Acacia Donnellina (F) 1 1 1 0
Aeciphila e!ata (ii) 0 3 2 0
Aliberta eclu Is (ii) 0 0 1 1
Alloohvilus cn::-ina (H) 1 2 2 0
Alterent-erna sp. (H) 1 1 0 0
Andira inr.,;is (F) 0 0 0 1
Aohelandra c'"encana (:1) 2 1 0 0
Ascleoias cr--s_: vica (H) 1 0 0 0
Asoidoseorr'a ".~a] ocaroon (F) 0 0 0 1
Boerhavia rc-ns (C1) 3 0 0 0
Bunch-nsia, ]W -c-Oat- (H) 0 0 1 0
Bursera si '-u-a (:) 1 2 2 0
Bvrsoni-a crassinolia (H) 1 2 2 0
Bvttneria 'acllanta (n) 2 3 1 0
Calathea 1uteta (Hi) 0 2 2 1
Casearia sp. (I;) 0 1 1 1
Cecro)ia so. (F) 1 2 2 0
Centrose-' il- ':.'iori (H) 1 0 0 0
Chrvsobaln-'is Icaco (i) 0 0 1 1
Cinura nalurocn (i) 1 0 0 0
Cissa:-~elos iartira (H) 3 1 0 0
Cissus sicvoiYs (H) 3 1 0 0
Cochlos"er-":u:i vitifol ia (H) 2 2 0 0
Conosteia xPe!nensis (H) 1 2 0 0
Costus sanc in-.u (0) 0 1 1 0
Croton 'labiilus (E) 0 1 2 1
Cuphoa calonn'lla (iI) 3 0 0 0
Cyperus spp. (1;) 1 0 0 0
Dalberfia bro:-nei (H) 0 0 1 1
Desnodium:- spp. (I) 1 0 0 0
Divitaria horizontalis (Hi) 1 0 0 0
Diosuvros sp. (1i) 0 0 0 1
Entada pol ystachya 0 0 0 1
Erythrina ex icn a (H) 0 1 1 0
Euwenia spp. (r) 2 2 1 1
Eunatorium aff. crocodili m (H) 3 0 0 0
Genioa caruto (F) 2 3 3 0
Hanelia patons (H) 0 0 1 1
Hamelia rovirosae (H) 0 1 1 1
Heliconia ia isntha (H) 3 2 1 0
Hyotih verticillata (H) 1 0 0 0
Inca sp. (F) 0 0 1 1
Iresine colosia (H) 2 1 0 0





- 85 -


Table 13. cont'd.


Relative frcouency of occurrence
Scientific nare 1-2 yrs 3-6 _vrs. 7+ _yrs. Forest

Laciacis rhizoophra (H) 2 1 0 0
Lantana cainra (i) 3 1 0 0
Leptochloa virnat- (H) 1 0 0 0
Lucuma izahaic n sis (F) 0 0 0 1
Lycianthcs ]r-- t (H) 2 1 0 0
Lvyodirn hceto.dr-.c::- (H) 1 2 1 0
Nandcvvl l',a su 'bs' i ttata (H) 1 0 0 0
Norordica c haraptia (1) 3 0 0 0
Nectrardra sp. (i.) 0 0 1 1
Nourol.na loba-t (H) 2 0 0 0
Ocoten sp. (:-) 0 1 2 1
Oronto~neero & l-1bf:--ui (H) 0 0 1 1
Orbl-vna cn't1e: (F) 2 2 2 0
OreopanaxC c:-)itat:ui (H) 0 0 0 1
Panlicn spp. (iH) 1 0 0 0
raspalu- spp. (H) 1 0 0 0
Passiflora corinacnc (H) 1 1 0 0
Pavonia rnsa (i) 1 0 0 0
Phor.cdendr.r sp. (!i) 0 0 0 1
Pinor aern:inosib ccW1E (n) 0 2 3 0
Fiper aurit:n (-) 3 1 0 0
Pithecelloo!, i lococa'- (H.) 0 0 0 1
Pithecollobnin:i s',.n (U) 0 0 0 1
Pithec l -- .- ---oc n () 0 0 1 1
Pluiuaria prit -'nvi (Cr) 1 2 3 0
Fosoucria latifol i, (2) 1 1 2 1
Priva 1on-'i'.ca-, -" (i) 1 0 0 0
Psychvtria spp, (i) 1 2 1 0
Serjania -rosii () 3 1 0 0
Sida spp. (;i) 2 0 0 0
oSjilax Yolis (H) 0 1 1 0
Solanun ochrncco-frruineui (HI) 2 1 0 0
Spigelia hi-u :.oltirnn (H) 3 2 1 0
Stachytarpheta cavce.encnsis 1 0 0 0
Sternade-ia Do:nnell-S:0itii (F) 0 0 0 1
Sti:.;aphyllo oillipt.ct. (i0) 3 0 0 0
Stro:ianthie luten (-1) 0 1 1 1
Svietsnia iacro~h-11la (F) 0 0 0 1
Tabebuia pontnapyil.. (F) 0 0 0 1
Tabernaeo.ont.:na al.ebia (H) 0 0 1 1
tabcrnae-;ntna cr'--itha (0) 0 1 1 0
Tabcrpac:-ont-'na ch.-' sce-ora (H) 1 2 1 0
Teco-a sctas (I) 0 1 2 0
Tephrosia sp. (0 0 1 1 0
Ternstroemian tcoe::'--ote (H) 0 1 1 0
Tetra)cteiis schi.d'c- (H 1 1 1 0
Thevetia nhouai (H) 1 2 3 0




- 86 -


Table 13. cont'd.


Relative freouencv of occurrence
Scientific name 1-2 yrs. 3-6 yrs. 7- yrs. Forest

Trema micrantha var. floridana (H) 3 3 1 0
Tri.onla ra7.a C:) 1 3 3 0
Triu -c-ta i:i-- ha (H) 1 0 0 0
Vitex k'lonvii (r,) 0 0 1 1
Waltherica a:. ricana (H) 3 1 0 0
'ledolia aca,-iilconsis (H) 0 0 0 0
Wedelia trilo',t (I) 0 0 1 0
Zexr,:onia scandccs (II) 3 0 0 0



(a)Relative frequency of occurrence represented by .(0) not observed,
(1) present, (2) frequent and (3) abundant. Forest species were
recorded as present (1) or not observed (0).
(b)Identification made in herbaria (;i) or in field (F).
(C)Early mature secondary forest (p32 yrs.).





- 87 -


The 4 age-classes are thought to be indicative of a uniformity among

the component ages with respect to certain environmental and physiognomic

characteristics. The basis for these age-categories is described in

the following chapter. For each age-class, each species was given a

number corresponding to its relative frequency of occurrence within the

class. Zero (0) indicates that the species has not been observed,

whereas a one (1) indicates at least two individuals have been observed

in fallow vegetation. Individuals of species which occurred on at

least 25% of biomass sample plots, and commonly appear on the fallow

sites are designated with a two (2). Species given a numeric value of

three (3) occurred on at least 50% of the sample plots or have been

observed to be a ubiquitous component of fallows of a given age-class.

These values are meant to express only the relative frequency of the

observed and recorded species. The greater number of species recorded

for the earlier age-classes reflects the emphasis placed on the earliest

appearing successional species and are not necessarily interpreted as a

measure of species diversity. Also, the construction of age-classes of

dissimilar interval size, and the use of equal sized sampling plots in

all ages are recognized for their influence on comparisons among age-

classes.

Within the limits imposed by these deficiencies, a simple inter-

pretation of the data can be made. Inter age-class comparisons showing

the per cent of observed species common to_2 age-classes were calculated

using the formula

% Similarity = 2X x 100 (9)
Y + Z





- 88 -


where X is the number of species occurring in 2 age-classes, and Y and

Z are the total number of observed species in each respective age-class.

The results of the 6 comparisons of similarity are shown in Figure 11.

In general, the greater the age difference between vegetation units,

the less species there are in common, based on a successional oriented

flora. Contiguous age-classes show the highest degree of similarity.


Biomass Standing Crop

Summarized in Figures 12 through 15 are the results of the bio-

mass standing-crop harvests. The data used in the following calcula-

tions are presented in Appendix, Tables C-1 to C-10. All biomass values

are expressed on an oven-dry basis in gms m-2. Because the second-

growth regeneration begins immediately after clearing and burning, pre-

paratory to cropping, and the final release time is unknown for most

sites, time zero is not included in the regression analysis. For pur-

poses of this study, second-growth regeneration is assumed to begin

normally'following the last weeding in July. As all biomass harvests

were made around July in each of the study years, 1 year-age-intervals

are used as the independent variable.

For each of the designated compartments- total biomass, wood and

leaf- regression equations were estimated relating the standing crop

to age. The hypothesis was that
A
Y = bo + blX + b2X2 (10)
A
here Y is the estimated standing crop in gms m-2 and X is the age in

years. The step-wise multiple regression analysis was employed such

.that the quadratic term (b2X2) was included only if it significantly












VEGETATION AGE CLASSES


1-2 YEARS


S3-G YeA4 3


7 YEARS
l________


>32 YEARS


68%


S 72%


4t5%/


38 %


9%0


Figure 11.- Comparison of species similarity among 4 age classes.
value indicates the observed number of species common
(linked by arrow) out of the total number of observed
age classes


The percentage
to 2 age-classes
species in both


2% -_.




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