Structure and Function in Agroecosystem Design and Management - Chapter 6 doc

CHAPTER 6
Biological Interaction in Tropical
Grassland Ecosystems
Panjab Singh and S.D. Upadhyaya
CONTENTS
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Nature of Tropical Grasslands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Successional Levels. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Diverse Grassland Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Biodiversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Species Diversity in the World . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Community Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Ecosystem Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Structure of Tropical Grassland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Abiotic Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Biotic Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Production Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Primary Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Secondary Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Biological Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Biophysical Interactions at the Ecosystem Level: Exploratory
Studies at Iseilema Grasslands of Ujjain, India . . . . . . . . . . . . . . . 127
Interspecific and Intraspecific Interactions. . . . . . . . . . . . . . . . . . . . . 131
Biophysical Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Interaction of Trees and Grasses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Aboveground Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Belowground Interactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
113
0-8493-0904-2/01/$0.00+$.50
© 2001 by CRC Press LLC

920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 113
114 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Grass-Legume Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Trees-Grass-Livestock Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Tree/Grass-Legume-Animal Interactions. . . . . . . . . . . . . . . . . . . . . . 136
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
INTRODUCTION
The grassland biome is characterized by grasses and their relatives where
the dominant life forms are mixed with herbaceous plants. Grassland ecosys-
tems consist of many interacting environmental forces, local combinations of
organisms, and the impacts of use by an increasing number of people. These
systems remain primarily under the control of overall environment, although
use and management of grassland ecosystems alter populations of organ-
isms, change the rate of physical and biological inputs, and account for about
25% of earth’s natural vegetation. Grassland ecosystem components include
soil, vegetation, populations, communities, and animals. Most of the exten-
sive areas of existing natural grassland have undergone changes through
man-tree-grass-animal interactions. Significant impact from grazing and fire
has been noticed. Plants are often adapted to fast, scattered fires that burn the
tops of plants but leave seeds, roots, or other resistant structures intact.
Examples include the tall grass prairie of the U.S. and Canada, the steppes of
Central Asia, and the plains of Africa. Because these areas are often suitable
for cultivation or livestock grazing, a great deal of this biome around the
world has been highly modified, often for many centuries or millennia.
The existence of grassland, i.e., the great bread baskets of the world, and
grazing animals extends back into the geological history (Box et al., 1969).
The grasslands have been one of the most precious of natural wealth since
times immemorial to man, which is supported by fossil records of grasses
observed in the cretaceous, or even earlier when flowering plants were

spreading throughout the biosphere. The precipitation-evaporation ratio and
precipitation-seasonality ratio are important biophysical factors in produc-
ing different types of grasslands and in the delineation of the grasslands.
Grasslands occur over a wide range of mean annual temperatures, occurring
in near tropical situations as well as extremely cold climates, having been
classified as steppes, prairies, and savannas, and temperate, semi-arid,
desert, alpine, and tundra grasslands, depending on their environment and
the vegetational characteristics at their place of occurrence.
One of the main aims of the international biological program (IBP) has
been the evaluation of the terrestrial productivity, the main theme having
been the synthesis of the grassland ecosystem to examine the “biological
basis of productivity in human welfare.” The synthesis of grassland ecosys-
tem analysis usually involves various statistical and mathematical models.
According to Van Dyne et al. (1978), grassland ecosystems are dynamic and
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 114
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 115
not static. In the grassland ecosystem, we see various dynamic phenomena,
such as changes in the biomass of plants and animals and phenological pro-
gression, as well as the less noticeable but still significant changes occurring
underground. In fact, these latter changes are more important when the
impact on the system is considered, such as changes in soil-water-energy, the
exuberance and extinction of microbial populations, the growth and vanish-
ing of roots, and other such related processes. Having taken notice of the
myriad changes taking place in response to the seemingly probabilistic
changes leading to a complexity, one needs to view the whole process as a
total system (Van Dyne et al., 1978) in view of biological interactions. In this
chapter, an analysis is made of research results obtained on the main interac-
tions identified in tropical grassland ecosystems, and their potential signifi-
cant impact is discussed.
NATURE OF TROPICAL GRASSLANDS

Tropical grasslands are seral in nature, attaining a status of disclimax at
many places, due to recurring biotic operations, such as grazing, fire, and
scrapping. They owe their origin either to deforestation or to shifting culti-
vation by nomadics, with the species composition of these grasslands vary-
ing with the intensity of grazing and harvesting.
The important functions of the grassland ecosystems are the dynamics of
organic matter and the production processes. Odum (1971) asserted that the
most important functional properties of ecosystems are energy flow, biogeo-
chemical cycles, and biological regulation. A major portion of the energy fixed
by the photosynthetic canopy of green plants ultimately finds its way into the
detritus component (Macfadyen, 1963). A considerable amount of information
is available about organic matter production and the processes associated with
it in different grassland ecosystems of the world, under varying climatic con-
ditions. Singh and Yadava (1974), Sims et al., (1978), Sims and Singh (1971,
1978a, 1978b, and 1978c) have presented illuminating accounts of the biomass
structure, productivity, and energy compartmental transfers, as well as the
accumulation and disappearance of organic matter in grazing land ecosystems.
Bokhari and Singh (1975), Billore and Mall (1976), Pandey (1975),
Upadhyaya (1979), and Paliwal and Karunaichamy (1999) have adopted a
modeling approach for the evaluation of the uptake, transfer and release of
the system state variables. Yadav and Singh (1977) have described a thorough
legend of the grasslands of India, while others (Coupland, 1979) have ade-
quately dealt with the structure and function of the grasslands of India and
the world, including an illustrative account of the decomposer kinetics in the
grazing land ecosystems. A survey of this literature points out that although
much information is available on production dynamics and the aspects of the
grazing land ecosystems, there is a wide lacuna in our understanding of the
biological interaction in tropical grassland ecosystems.
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 115
116 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT

ORIGIN
Fossil record shows that tropical grasslands originated as long ago as 6 to
12 million years. The environment remained in its pristine purity and conge-
niality as man stayed in the hunting and gathering stage. However, man
entered the pastoral age and domesticated animals and then gradually
passed from the nomadic stage to settled cultivation. Grasslands were impor-
tant to man before plants were ever domesticated. In the late 1800s the impor-
tance of grasslands and the grass plant were recognized. The great “bread
baskets” of the world exist on soils developed under centuries of grassland
cover. Grasslands in tropics have mainly originated from the destruction of
permanent woody vegetation and are thus bio-edaphic sub-climaxes.
Tropical and subtropical grasslands are located in the plains and mountains
within 28°N and 30°S of the equator (Thomas, 1978). This land mass of trop-
ics and subtropics accounts for 38% of the earth’s surface and 45% of the
world’s population (FAO, 1995). The extent of tropical grasslands and live-
stock population is summarized in Table 6.1, which illustrates the livestock
dependence on grasslands. The number of livestock has increased, and at the
same time the area of grasslands has decreased around the world (except in
Brazil), indicating intensification of grassland usage by livestock. It is esti-
mated that over 90% of the feed for livestock on a world-wide basis comes
from grasslands/rangelands. With continued human population growth,
there will be increased demand for milk and meat, resulting in even more
intensive grassland utilization. Greater intensity of grassland utilization will
require more knowledge of the functional ecology and biological interactions
in grassland ecosystems.
Successional Levels
Every living being is surrounded by materials and forces that constitute
its environment and through which it meets its needs. Nothing can escape its
environment, no animal or plant can live completely sealed off from the
world, and all living things must make exchanges with their environment in

terms of energy, matter, and waste elimination. All living beings are interde-
pendent and must absorb energy, termed as natural resources, more or less
continuously to fuel their life process. The grasslands are renewable natural
resources and are one of a number of seral phases of vegetation. Their struc-
ture is dynamic rather than static. One ecological association follows upon
and grows in consequence of its predecessor in a well-marked and orderly
sequence. One association therefore acts as a nursery to its immediate suc-
cessor. This series of orderly sequence from the first to the last is referred to
as the sere. The successional levels of tropical grasslands are characteristic
phases of the sere which may thus end at a subclimax rather than at its
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 116
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 117
Table 6.1 Land Area, Permanent Pastures, and Livestock Population of the
Tropical and Subtropical Countries*
Land area Permanent pastures Livestock*
(M ha) (M ha) (M)
Region 1979 1994 1979 1994 1989 1994
World Total 13040.9 13045.4 3265.0 3395.2 4164.3 4204.3
Africa 2963.5 2963.5 892.7 883.5 609.5 641.3
Asia 2679.0 2679.0 686.4 792.1 1697.0 1880.9
Brazil 845.6 845.6 170.1 185.0 224.3 236.2
Australia 764.4 764.4 436.3 414.5 191.6 149.9
India 297.3 297.3 112.1 111.4 440.4 454.5
Sudan 237.6 237.6 98.0 110.1 59.0 65.1
Indonesia 181.0 181.0 12.0 11.8 38.8 44.3
Chad 125.9 125.9 45.0 45.0 10.1 11.1
South Africa 122.1 122.1 81.4 81.3 53.0 50.3
Ethopia 110.1 110.1 45.4 44.9 — 77.8
Venezuela 88.2 88.2 17.1 17.8 — —
Pakistan 77.0 77.0 5.0 5.0 101.1 117.2

Nigeria (Kenya) 56.9 56.9 21.3 21.3 55.2 64.4
Cameroon 46.5 46.5 2.0 2.0 12.8 13.9
Nepal 13.6 13.6 1.8 2.0 — —
Bangladesh 13.0 13.0 0.6 0.6 — —
Sri Lanka 6.4 6.4 0.4 0.4 3.2 3.3
Bhutan 4.7 4.7 0.2 0.2 0.6 0.7
Pacific Islands (Fiji) 1.8 1.8 0.1 0.1 0.5 0.7
* Livestock numbers include horses, mules, asses, cattle, buffaloes, camels, pigs, sheep, and
goats
* Based on FAO Production Yearbook data, 1995
climax, e.g., grassland of arid and semi-arid tropics (low rainfall areas).
Monsoonal grasslands in the tropics are the stabilized successional stages of
vegetation. In areas of higher rainfall, the successional levels terminate in for-
est as a climax stage. Here, the biological interaction determines the charac-
ter of vegetation and also the successional level of the ecosystem. The grazing
animals (biotic pressure) maintain the successional level of grasslands
(Barnard and Frankel, 1964). Grasslands are maintained as such due to bio-
edaphic pressures. Similarly the use of fire has also been a very important fea-
ture associated with development of tropical grasslands. Besides these, the
most important constraint affecting grassland is its extreme fragility. This
means that the landscape, vegetation, and soil cover degrade much more
quickly than in more favored habitats; fragility affects the biological system
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 117
118 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
and hence sustainability or, in ecological parlance, homeostasis—the ten-
dency of a biological system to resist change and remain at a stage of dynamic
equilibrium or relative consistency. This is because a grassland ecosystem is
capable of self regulation due to biological interactions as a law of nature.
Diverse Grassland Communities
Most developing countries are in the tropics, where grasslands are the

major feed resources (over 40%) for livestock rearing. Due to enormous biotic
activities, the grassland communities have undergone significant changes.
The tropical and subtropical grasslands of the southern hemisphere are rep-
resented by savannas with low vegetation and scattered trees, while steppes
in Asia are generally grassy and without trees. Africa is covered with more
than one third grassland of Acacia-based savannas. The savannas in Australia
are dominated by Eucalyptus and Acacia both equally. In India, Burma, and
Indonesia, grassland savannas occur in the tropical rain forests. Bamboo-
based savannas are common in India. Most of the Japanese grasslands repre-
sent semi-natural grasslands created and maintained by man. Around the
world, the grassland communities consist of 22% high grass savannas, 31%
tall grass savannas, 13% tall grass prairies, 10% short grass prairies, 18%
grasslands and savannas, and 6% mountain grasslands, (Shantz, 1954;
Whyte, 1960). Tropical grasslands of India are rich in biodiversity and also
diverse heterogeneity in nature because of the great variation in climate, soil,
and physiography. Dabadghao and Shankarnarayan (1973) have identified
five major grass covers of India—Sehima-Dichanthium, Dichanthium-
Cenchrus-Lasiurus, Phragmites-Saccharum-Imperata, Themeda-Arundinella and
Temperate Alpine distributed in elevation from 150 to 2100 m and rainfall
ranges from 100 to 3750 mm. Over 40% of the total geographical area of India
is available for grazing by over 400 million livestock under diverse grassland
communities. The grazing pressure is very high, 1–4 ACU/ha, against the
normal 0.2–0.5 ACU/ha in the arid and semi-arid areas of India (Shankar and
Gupta, 1992).
BIODIVERSITY
The variety of all life forms—the different plants, animals, and microor-
ganisms, the genes they contain, and the ecosystems of which they form a
part—is termed biological diversity or biodiversity (Wilson, 1992). Grassland
biodiversity is not a fixed entity, but constantly changing; it is increased by
genetic change and evolutionary processes and reduced by extinction and

habitat degradation. The concept emphasizes the interrelatedness of biome
and biological interactions. Grassland biodiversity is also a limited and a
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 118
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 119
Table 6.2 Tribes and Genera of the Family Gramineae (Grasses)
Tribe Genera
Andropogoneae Andropogon, Bothriochloa,Chrysopogon, Colix, Cymbopogon,
Dichanthium, Hemarthria, Heteropogon, Hyparthenia,
Hyperthelia, Imperata, Ischaemum, Iseilema, Lasiurus,
Saccharu, Sehima, Sorghum, Themeda, Trachypogon,
Tripsacum, Vetiveria, Vossia, Zea
Aristidae Aristida
Arundineae Phragmites
Arundinelleae Loudetia, Tristachya
Chlorideae Asterbla, Chloris, Cynodon, Enteropogon
Eragrostideae Dactyloctenium, Diplachne, Eleusine, Eragrostis,Triodia
Oryzeae Leersia, Oryza
Paniceae Acroceras, Anthephora, Axonopus, Brachiaria, Cenchrus,
Digitaria, Echinochloa, Eriochloa, Hymenachne, Melinis,
Panicum, Paspalidium, Paspelum, Pennisetum, Setaria,
Spinifex, Stenotaphrum, Tricholaena, Urochloa
Sporoboleae Sporobolus
Zoysieae Leptothrium
perishable natural resource. It has three components, namely: species diver-
sity, community diversity, and ecosystem diversity.
Species Diversity in the World
The strong impact of climate throughout the world also manifests itself
in marked species diversity in world grasslands. The flora of grasslands, in
general, is dominated by therophytes and cryptophytes (Singh and Yadav,
1974). The preponderance of therophytes results from a strong periodicity in

biotope and biocoenosis. The loss of a species reduces species diversity and
threatens the functioning of ecological communities.
Grassland is one of a number of serial phases of vegetation (grass, shrub,
and trees), which has dynamic rather than static structure. Many of the large
tropical grasslands from west to east are dominated by the species of tribes:
Paniaceae characterized by high temperature and low rainfall, Andropogoncae
characterized by rainfall varying from 125 to 2250 mm and distribution
closely related to temperature. They are abundant in the tropical savannas of
India, Africa, and South America. Eragrostideae tribe is distributed abun-
dantly where yearly winter temperature is above 10°C and rainfall is about
1000 mm (Skerman and Riveros, 1990). There are ten common groups of
tribes (Table 6.2) found in tropical grasslands, which are unevenly distrib-
uted in world grasslands (Figure 6.1).
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 119
120 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Figure 6.1 Percentage distribution of tribes/grass in World Grassland Ecosystem.
Indian tropical grasslands consist of 245 genera and 1256 species of
grasses (Bor, 1960); out of these, 139 species are reported to be endemic
(Mehra and Magoon, 1974). Indian grassland legumes consist of 167 genera
and about 1150 species, including cultivated, introduced as wild species
(Singh and Morrison, 1998).
Community Diversity
The International Biological Programme (IBP) analyzed world grassland
communities, including natural grasslands, tundras, deserts, savannas,
prairies, steppes, and other grasslands derived from forests, and cautioned
about change in communities due to biological interactions. Man has modi-
fied grassland communities for intensification of animal and plant produc-
tivity through prudent use of fire, conversion to croplands, introduction of
new herbivores, replacement of native grasses/legumes by exotics, deliber-
ate incorporation of trees, etc. Permanent pastures occupy approximately

25% of the earth’s land area (Table 6.1): 3395 million hectares of permanent
pastures of the world provide forage and habitat for some 4204 million live-
stock. In the tropical and subtropical regions of the world, approximately
23% is grazing land communities (‘t Mannetje, 1978), mostly savannas with
varying proportions of trees and shrubs. Many of the large grassland com-
munities are climax formations determined by soil and climate; others are of
more recent origin and have replaced forest communities destroyed mainly
by cutting and fire, and these have been maintained largely through grazing
animals (Barnard and Frankel, 1964). Hence, fire and grazing have been
very important features associated with the community diversity. Natural
communities converted into grasslands are greatly influenced by biological
interactions.
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 120
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 121
Ecosystem Diversity
Ecosystem or ecological diversity of grasslands is changing day by day.
Many of the world’s original grasslands have been largely converted to crop-
lands or to seeded pastures, although these regions carry large numbers of
grazing animals. Similarly, many of the world’s original forests have been
converted to grasslands. Many desert areas are also utilized seasonally for
grazing. Collectively, about 40% of the earth’s ecosystem with normal spec-
trum of tribes and genera of the family Gramineae (grasses) is used by grazing
animals. The four main elements of grassland ecosystem, namely abiotic sub-
stances, producer organisms, consumers, and decomposer organisms, have
great diversity in world grasslands. Living organisms (plants, animals, and
microorganisms) are taken as a whole while studying interactions with the
nonliving environment in the ecosystem. It is mostly an open system compris-
ing plants, animals, organic residues, atmospheric gases, water, and minerals
that are involved together in the flow of energy and circulation of matter. A
conceptual model of organic matter storage, flow, and biological interactions

which help in nutrient cycling and CO
2
fertilization is shown in Figure 6.2. The
boxes in the figure represent organic matter accumulation, and the arrows
show pathways of transfer from one sink to another. Alphabetical symbols (u:
uptake; t: transfer; r: release) denote biophysical or biological functions of inter-
actions. The biochemical and physical factors include sunlight, rainfall, soil
nutrients, and climate. A grassland ecosystem is inherently “leaky”: at a mini-
mum, energy and nutrients move in and out. More likely, individual organisms
move in and out as well. Within each grassland ecosystem, there are a myriad
of well-defined groups of living organisms—producers (plants), consumers
(animals), and decomposers (bacteria and fungi)—interacting with each other.
Interactions of herbiovores, carnivores, and decomposers provide many routes
of nutrient transfer and release, describing the quantities of minerals in the var-
ious pools such as the soil, litter, and urine. A common type of interaction
amongst different tropic levels and total quantity of mineral flow from source
to sink are depicted in Figure 6.2. Detailed analysis of mineral/energy reserves
describes the system organization and provides a base for the study of mineral
cycling/energy flow through the system and the biological groups responsible
for transformations which will facilitate the grassland management in a sus-
tainable manner. The annual cycle of plant biomass accumulations and litter
decomposition has received much attention. With the development of concepts
of ecosystem structure and function, many grassland ecologists assorted the
carbon fixation by grasses and its later circulation in the ecosystems.
STRUCTURE OF TROPICAL GRASSLAND
Tropical natural grasslands structurally and physiognomically are char-
acterized by mixed herbaceous plants (dominated by grasses), trees, and a
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 121
122 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Figure 6.2 Box and arrow diagram of ecosystem level model of mineral cycling and

energy flow in grassland ecosystem to study the impact of biological
interactions (r-release, t-transfer, u-uptake).
low plant cover of non-woody species. Unstable grasslands representing
disclimax have been derived after the destruction of forests and are main-
tained due to regular biotic interference. Such vegetation is normally termed
savanna (Moore, 1970). In the course of time, the grasslands have undergone
significant changes, due to the human population pressure, in terms of
declining area, carrying capacity, and productivity. Structure and function of
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 122
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 123
tropical grasslands need to be improved by enhancing the potential of bio-
logical interactions and matter cycling.
Abiotic Characteristics
Tropical grasslands are characterized by a climate that shows distinct wet
and dry (cyclical) seasonal patterns. Mean annual precipitation in tropical
and subtropical grasslands usually ranges 600–1500 mm and the length of
active growing season ranges 120–190 days (rainy season). Temperature
becomes the controlling factor for biological interactions in a tropical grass-
land ecosystem. The mean monthly variation of temperature between the
warmer and colder seasons in a tropical area is 5°C; for every 100 m increase
in elevation there is a decline of 0.8°C in the mean annual temperature. The
soils of tropical grassland are highly leached, and there is rapid decay (due to
high temperatures) with low levels of humus accumulation with reddish or
yellowish color. Abundant groups of microorganisms in tropical grassland
soils are bacteria, actinomycetes, and fungi. A number of different types of
decomposer organisms are recognized on a functional basis (Clark and Paul,
1970).
In order to consider abiotic characteristics from an ecological point of
view, Walter (1973) proposed the climate diagram which gives information
concerning the mean temperature, precipitation, relative humidity, and arid

seasons. Based on the climate diagram, different ecological zones of tropical
and subtropical grasslands are abiotically characterized as semihumid, semi-
arid, subarid, euarid, and perarid grasslands, depending on hot/wet/dry
season duration.
Biotic Characteristics
If the grasslands are to be maintained as seral stages of ecological devel-
opment, the biotic components (producer, consumers, and decomposers)
must interact with each other, at the expense of solar energy, into a form in
which they are to be reused. Producers, consumers, and decomposers are
well organized grazing and detritus food webs.
A biotic model (Figure 6.3) depicting biophagic and saprophagic path-
ways describes the food web, and utilization of biological interactions and car-
bon cycling in tropical interactions will lead to a sustainability of grazing land
resources. Producers in tropical grasslands are mainly graminoides (grasses
and sedges) of Andropogoneae, Paniceae, and Cyperaceae groups which often
furnish 90% contribution. The fauna consists of invertebrate and vertebrate
predators, small herbivores, and very few carnivores. Microarthropods and
microbes (mostly bacteria) comprise the group of developers which help in
operating the detritus food web. In addition to the role of reducers and
decomposers, the microbes also play a vital role in biological nitrogen fixation.
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 123
124 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Biophagic path
of CO
2
Producers
(Autotroph)
Consumer l
(Herbivores)
Consumer ll

(Carnivores)
Decomposer
(Microorganism)
AT M
Co
2
release
Saprophagic
path of Co
2
R
1
R
2
R
3
Figure 6.3 Grazing (Biophagic pathway) and detritus (Saprophagic pathway) food
web in a tropical grassland ecosystem. R
1
, R
2
, and R
3
represent the res-
piratory losses from trophic levels.
Production Strategy
The grassland ecosystem contains a complex mixture of carbon compo-
nents in a continuous state of creation, transformation, and decomposition.
This dynamic state is maintained through the ability of grasses (C3 and C4),
forbs, shrubs, and trees in grassland to capture the solar radiation and utilize

it to transform carbon dioxide (and water) into organic molecules of rich
diversity. This interaction between the living (plants) and nonliving (abiotic)
environments is known as biophysical interaction. Many interesting interac-
tions between animals within the grassland system form feedback loops
related to the food chain (Figure 6.2). Animals (herbivores and carnivores) of
the tropical grasslands vary from the lowly insects (invertebrates) to magnif-
icent vertebrates. Ants and termites are often abundant among the verte-
brates, and the large mammal herbivores dominate, with predators making
only a small contribution to the average annual biomass. Where autotrophs
are measured in thousands of kilogram biomass per hectare, the annual
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 124
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 125
standing crops are in the order of hundreds of grams per hectare. The
average biomass of invertebrates below ground may be ten times that which
is above ground. Based on studies of several types of grassland ecosystems.
Wiegert and Evans (1967) concluded that (1) stable natural grasslands can be
utilized to an extent approaching 0.3–0.45 range of herbivore ingestion/net
primary production ratios determined for managed grasslands; (2) the level
of utilization with the presence of large livestock, and (3) ecosystems domi-
nated by invertebrates may be exploited at very low levels, but secondary
productivity is high when calculated per unit standing crop basis. Where net
primary productivity is very high, the secondary productivity of invertebrate
herbivore populations may be greater, on a per unit area basis, than that of
the large mammal herbivore population. In tropical grassland ecosystems,
the large animal biomass is higher as compared to temperate grassland
ecosystems. Man’s domestic animals make up the greatest part of the large
animal biomass in the developing tropics. Much of the natural grasslands
have been replaced by man-managed rangelands and are deteriorating day
by day because of erosion, recurring drought, and abusive grazing. Biotic
operations also change the production strategy of tropical grasslands favor-

ably. It is thus essential to promote studies on biological interactions and
cause-effect relationships operating among biotic organisms and abiotic
environmental variables (Shiyomi, 1997).
Primary Productivity
Synthesizing information and analyzing data on standing state biomass,
energy flow, nutrient cycling, and primary productivity are of immense value
for understanding biological interactions. Figure 6.2 simulates the grassland
ecosystem-level model used to explore the interactions of producers, herbi-
vores, carnivores, and decomposers within an environment. The biological
and biophysical interactions of elevated CO
2
and elevated solar radiation
change grassland production, decomposition rates, and nutrient uptake, and
transfer and release functions. It is one of several large models that derive the
interactions in the ecosystem level model (ELM). By computing different
components and functions of ELM and their relationship with driving
variables (i.e., daily precipitation, weekly max/min temperatures, wind
speed, relative humidity, monthly mean soil temperature) and state variables
(i.e., soil and inorganic ammonium, nitrate data, and growth parameters),
one can predict grassland ecosystem dynamics that could be attended by
changes in temperature, elevated CO
2
concentrations, changes in precipita-
tion, and ultimately changes in grassland productivity. Energy fixed by the
producer component as total net primary production is dissipated to herbi-
vores—carnivores—decomposer via litter and root decomposition. The pro-
ducers also absorb nutrients from soil and incorporate these in their biomass.
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 125
126 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
These nutrients are then transferred through dry matter to consumers and

decomposers and ultimately are released to soil through biological phenom-
ena. Primary productivity, nutrient status, and turnover have been studied
by various workers in tropical grasslands (Singh, 1976; Billore and Mall, 1976;
Singh and Yadav, 1974; Karunaichamy and Paliwal, 1995; Paliwal and
Karunaichamy 1999) and temperate grasslands (Bokhari and Singh, 1975;
and Sims and Singh; 1978).
Secondary Productivity
Tropical grasslands constitute a significant community type in the energy
economy of the biosphere. Grassland systems are managed primarily for the
development of plant materials for the production of livestock (cattle, sheep,
goats, and other herbivores) utilized by man as food or byproducts. Secondary
productivity of grasslands, defined as the calorific equivalent of consumer
protoplasm produced per unit time, is dependent on primary productivity of
the system and also on the assimilation/ingestion (a/i) and/or produc-
tion/assimilation (ps/a) efficiency of the consumers. Consequently, the com-
munity with the highest primary productivity possesses the capabilities for
the greatest secondary productivity. The degree of utilization of primary pro-
duction by herbivores (grazing pathway) and its further ingestion by carni-
vores and decomposer (detritus pathway) are two different forms of energy
flows (Figure 6.3). Based on the population of consumers, the a/i efficiency
and ps/a efficiency may vary independently with each other. Odum et al.
(1962) noted that grasslands can be utilized sustainably to an extent of herbi-
vore ingestion/net primary production ratio of 0.3 to 0.45. Surprisingly,
information on the sustainable utilization of tropical grasslands with special
reference to optimization of consumers’ interactions is meager. Studies are
needed on complex, indirect biological interactions in tropical grassland
ecosystems.
BIOLOGICAL INTERACTIONS
The literal meaning of “interaction” is reciprocal action or influence on
each other. In grassland ecosystems, various types of grasses (different tribes

and species) are grown in close proximity to one another and also to other
herbaceous plants (other than the gramineae family), and in some cases (such
as in savannas) with woody perennials. Invertebrates (including arthropods
and microbes) and vertebrates (including livestock) also live together in
grasslands. Various interactions take place between the species (plants and
animals) and within the species through the media of soil and microclimate
and may exert favorable or adverse effects on each other and also on envi-
ronment. Nair (1993), Ong and Huxley (1997), and Rao et al. (1997) discussed
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 126
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 127
various tree-crop and biophysical interactions in agroforestry. These interac-
tions take place both above- and belowground and include a complete set of
systems relating to radiation exchange, water and nutrient budget, and car-
bon budget. Tropical grasslands have all kinds of interactions categorically
defined as (1) physical environment affecting biological environment and
vice versa, (2) interspecific and intraspecific interaction, (3) biophysical inter-
action, (4) tree/grass interaction, (5) grass/legume interaction, and (6)
tree/grass-livestock interaction.
Biophysical Interactions at the Ecosystem Level: Exploratory Studies
at Iseilema Grasslands of Ujjain, India
The organic matter budget has been intensively studied in Iseilema
grassland community at Ujjain, located tropically (23°11Ј N and 75°43Ј E) in
India. The grasslands of Ujjain are tropical and are seral in nature, owing
their origin to the biotic perturbations. The driving variables of the site
included precipitation (AR), humidity (ARH), temperature (AT), and solar
radiation (ASR). The climate, essentially monsoonic, is characterized by three
distinct seasons in a year—rainy, winter, and summer seasons. The climatic
data of the area revealed the average annual rainfall during the last 10 years
to be 928 mm, while the mean minimum and mean maximum air tempera-
ture during the same period ranged from 24°C to 32°C and 12°C to 16°C,

respectively. Likewise, the mean relative humidity of the area under investi-
gation was found to be 41.12% at 4.30 p.m. while 70.25% at 8.30 a.m. The
monsoonic climate of the area is paralleled by dry subhumid and megather-
mal conditions. The hydrological processes model revealed a little water sur-
plus during the period of the investigation. Biotic state variables revealed a
significant positive correlation between these variables and the total viable
microbial populations in the grazing lands under study. Having observed
during the different seasons significant variations in the edaphic variables,
such as soil moisture, bulk density, pH, organic matter, and high amounts of
organic phosphate content of the soil, it is concluded that the soils of the pres-
ent study exhibit a greater degree of fertility.
As a sequel to a detailed study of the organic matter dynamics and other
aspects in these grazing lands, a brief investigation of the community struc-
ture and floristic composition of these grasslands was made. It revealed that
there were 42 species: 30 were grass species (most were Andropogoneae and
Paniceae), 5 were legumes, and 7 were species of forbs. Iseilema had the high-
est important value index among the grasses, while Indigofera had the high-
est important value index among the legumes.
The above- and belowground vegetational productive patterns consti-
tuted the input parameters, and the output variables comprised root decom-
position, litter decomposition and the total soil respiration. The production
dynamics patterns were evaluated by taking the biomass estimation and the
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 127
128 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
harvest data essentially into cognizance. The vegetation matrix was divided
into the compartments—live green (LG), standing dead (SD), litter (L), and
belowground (BG)—for synthesizing and modeling vegetation dynamics. By
a seasonal evaluation of the biomass, it was found to be maximum during the
rainy season. The values of the standing dead and the litter compartments
were abundant during the summer season, while the belowground value

was abundant during the winter season. The evaluation of the production
dynamics in the grasslands under study revealed that the aboveground net
production was maximum (612.28 gm
Ϫ2
) during the rainy season while the
belowground net productivity was maximum (762.34 gm
Ϫ2
) during the win-
ter season. The total net production value was highest (988.03 gm
Ϫ2
) during
the winter season. It was further observed that the canopy turnover rate was
at its maximum (0.89) during the rainy season, while the root turnover rate
was maximal during the rainy and winter seasons (0.36 and 0.28, respec-
tively). A box and arrow diagram (Figure 6.4) revealed the production
dynamics of the Iseilema grassland ecosystem at Ujjain (India) where the total
disappearance value of the organic matter for the year was 1086.44 gm
Ϫ2
yr
Ϫ1
,
while the net accumulation value of the organic matter had been found to be
1137.25 gm
Ϫ2
yr
Ϫ1
. Also, the production dynamics values and the biomass val-
ues were primarily correlated with environmental variables such as air tem-
perature and rainfall. An analysis of the transfer matrix of the process model
revealed greater insight into the mathematical modeling of the transfer rates.

The maximum amount of the organic matter input was observed during the
rainy season in the present grazing land ecosystem. An analysis of the system
transfer functions had revealed that the input and output ratio was high
(1.96) during the summer season for the organic matter transfer from the TNP
to the TD, while the ratio was lower during the rainy and winter season. An
open, time varying transfer coefficient model having differential equations of
an algebraic nature (based on the summation value of income minus loss of
internal variable) as employed to evaluate the transfer values in the system
compartments; the results revealed that more organic matter was transferred
during the rainy season than in other seasons (Gupta and Singh, 1977). It had
been inferred by the present investigation that organic matter output func-
tions included the litter decomposition and root decomposition, as well as
total soil respiration parameters. The litter decomposition study showed
higher rates of litter disappearance, in terms of reduced weight, in September
and that the litter decomposed very fast at 5 cm soil depths in the litter bags
buried during the rainy season. The various chemical constituents of the
decomposing litter had a positive correlation with concentration of the nitro-
gen, water soluble sugars, and phosphorus, while a negative correlation was
found between these parameters and the lignin, cellulose, and organic carbon
content. It was further noticed that the decomposition rates for the litter and
roots were 0.3 and 0.8 (upper 10 cm soil depth), respectively. The studies
reveal that the maximum percentage of roots (72.88%) was found in the
upper 10 cm soil depth. It has also been observed that increased soil depth
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 128
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 129
203.6 33 70
149354
ARHASRATAR
ANP
870

TNP
1698
L
656
R
2722
1960
572
S
SR
1121
NR
17.6
PR
0.80
TD
1086
PU
7.3
NU
24.7
1218
TSR 2339
RR
828 BNP
Figure 6.4 Figure showing the biophysical interaction and functions of the iseilema
tropical grazingland detritus sub system at Ujjain, India. Boxes and
arrows represent the status and flow of organic matter functions.
TNP-Total Net Production (g, m
Ϫ2

) PR-Phosphorus Release (g, p, m
Ϫ2
),
ANP-Aboveground Net Production (g, m
Ϫ2
) PU-Phosphorus Uptake (g, p, m
Ϫ2
)
BNP-Belowground Net Production (g, m;g
Ϫ2
) NR-Nitrogen Release (g, p, m;
Ϫ2
), and
TD-Total Disappearance (g, m
Ϫ2
) NU-Nitrogen Uptake (g, p, m
Ϫ2
)
L-Litter (g, CO
2
, m
Ϫ2
, yr
Ϫ1
) Large, lined arrows represents the
SR-Soil Respiration (g, CO
2
, m;g
Ϫ2
, yr

Ϫ1
) driving variable of the system i.e.
R-Root (g, CO
2
, m
Ϫ2
, yr
Ϫ1
) AR - Average Rainfall (mm), AT -
RR-Root Respiration (g, CO
2
, m
Ϫ2
, yr
Ϫ1
) Average (maximum) temperature (°C),
TSR-Total Soil Respiration ASR-Average (K Cal m;g
Ϫ2
, month;
(g, CO
2
, m
Ϫ3
, yr
Ϫ1
), g
Ϫ1
) Solar Radiation, ARH-Average
Relative Humidity (% value, recorded
at morning hours)

S-Soil
enhances root biomass reduction. Observations on the root decomposition
reveal that the litter bags buried during the rainy season decomposed very
fast, at upper 10 cm soil depth (85.33% weight loss during 365 days), with the
mean relative decomposition with 0.00198 g.g.
Ϫ1
day
Ϫ1
and K value (decom-
position constant) of 0.32 at the same soil depth. The root decomposition
rates were positively correlated with concentrations of nitrogen, water solu-
ble sugars, and minerals and negatively correlated with concentrations of
carbon, lignin, cellulose, and C/N ratio. Abiotic variables such as rainfall, soil
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 129
130 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
moisture percentage and root water content showed a positive correlation
with the root decomposition rates. In addition, the total viable microbial
population showed a positive correlation with root decomposition rates. A
multiple regression equation explains 46% variability in root respiration due
to variation in root biomass and root moisture index. It was noted that in
the total CO
2
output, as a result of net respiration by soil, 56% CO
2
was
contributed by the roots, while the remaining percent of CO
2
was contributed
by the litter and soil microorganisms. Soil respiration was significantly
correlated with soil water, rainfall, temperature, and microbial population

in the soil (Upadhyaya et al., 1981; Upadhyaya and Singh, 1981). Soil
respiration was positively correlated with the litter weight loss value. A
diurnal analysis of CO
2
evolution rates from the soil revealed that the soil
respiration was high at night in the summer and high during the day in
the winter. It was noted that litter and root biomass showed a positive corre-
lation with the total soil respiration along with the litter and root water
(Coleman, 1973 a, b).
The nitrogen and phosphorus cycling and their seasonality were evalu-
ated in the grazing land study, and it was found that the input and output
variables were significantly correlated with the temperature and rainfall.
From those studies, it was concluded that the total nitrogen and phosphorus
uptake of about 71.3% of nitrogen and 64.48% of phosphorus was released
into the soil, while 28.7% of nitrogen and 35.5% of phosphorus was usually
immobilized in the system. An analysis of the microbial turnover of nitrogen
in the system revealed that with higher nitrogen content of litter, the micro-
bial activity significantly increased, thereby enhancing the rate of the litter
decomposition. A model had been developed to explain the dynamic struc-
ture of detritus carbon in the Iseilema grazing land community at Ujjain, India
(Figure 6.4). The equilibrium time for organic matter accumulation in this
ecosystem was found to be 8.3 years. Input and output rates had been calcu-
lated quantitatively and a balance sheet was proposed. All the values had
been represented in terms of g CO
2
m
Ϫ2
hr
Ϫ1
. It was concluded that 96% CO

2
output (2398 g CO
2
m
Ϫ2
hr
Ϫ1
) of total carbon input (2530 g CO
2
m
Ϫ2
hr
Ϫ1
) was
found in the present grazing land community. The organic matter turnover
rate was 0.55 (calculating by value of K, i.e., decomposition constant), which
was high compared to the temperate grazing lands. The decomposing roots
were separately developed as a “seasonal soil core model,” illustrating the
maximum root organic matter flow (input and output) during the rainy sea-
son at the upper 10 cm soil depth, with the 62% CO
2
output of the total car-
bon input values. Standing crops of the soil microbes were responsible for the
microbial turnover of the detritus carbon which varied from 11 ϫ 10
6
to
256 ϫ 10
6}
counts per gram of dry soil. A summarization of data revealed that
the values of input and output variables of organic matter in the Iseilema graz-

ing land ecosystem, on a year-long basis, were found to be 253 g CO
2
m
Ϫ2
yr
Ϫ1
,
of which 92% of the total organic matter was output, and the mean detritus
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 130
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 131
was found to be 0.68 kg C m
Ϫ2
. A critical evaluation of the total system mod-
els, in the form summarized in Figure 6.4, indicated generalized flow pat-
terns of the matter in the Iseilema grazing land ecosystem and optimization of
biophysical interactions of abiotic and biotic components of the ecosystem.
The study revealed that biotic and abiotic variables interacted with each
other and greatly influenced the detritus structure and function and also the
productive potentials of tropical grasslands in a sustainable manner.
Interspecific and Intraspecific Interactions
The phenology and growth of plants are governed by the environment,
but at the same time plant species can alter the environment. The nature of
the interactions within and among species, therefore, concerns the ways in
which a plant can influence its associates by changing the environment,
either directly by addition or subtraction (e.g., of nutrients) or indirectly (e.g.,
by encouraging bioagents). The “response and effect” principle (Goldberg
and Werner, 1983) states that the plant and its environment modify one
another in such a way that the environment causes a response in plant func-
tion and growth, and that response in turn affects the environment. In an
environment where one species of grass is with other species, grasses are

with legumes, or trees are with grasses, there are a number of possible out-
comes of the interactions among them. Grasses/legumes and trees may affect
the environment in a negative or positive way with respect to each other. Cole
(1949) and Goodall (1952) were instrumental in detecting the statistical asso-
ciation between pairs of species naturally occurring in plant communities. By
computing the association index, the amount of co-occurrence can be quanti-
fied. Williamson (1972) described the fitness of the species in the environ-
ment; for instance species “A” may increase (ϩ), decrease (Ϫ), or have no
effect (0) on the fitness of an individual of species “B.” Schoener (1988)
defined five resultant possible interactions, which are summarized in Table
6.3. The productivity of grassland can be defined by using the knowledge of
interactions categorically defined in Table 3 for land use practices of man-
managed tropical grasslands, which aims to encourage a favorable combina-
tion of the biological interactions (competition, predation, mutualism,
commenalism).
Biophysical Interactions
The four basic biophysical elements affecting grassland productivity are
solar radiation, water, nitrogen, and certain other nutrients, particularly
phosphorus and potassium. How each of these contributes to the
aboveground biomass yield (i.e., primary and secondary productivity) is
important. Through computing association analysis, land equivalent ratio
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 131
132 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Table 6.3 Positive (؉) and Negative (؊) Interactions of Tree-Grass
Association
Positive (
؉)
Aboveground herbage biomass of the grass is increased by 25% under the canopy
Trees shelter crop from wind
Trees as a check to wind erosion

Rainfall is intercepted by the tree canopy
Soil infiltration is higher—reduced evapotranspiration
Soil surface temperature reduced 6–8°C under canopy
Nitrogen mineralization is double under the canopy
Microbial biomass is higher under the canopy; CO
2
evolution is higher and leads to
high soil productivity
Root/shoot ratio of grass decreases under tree canopy
Tree inhibits pests of grasses
Nutrient retrieval by tree roots to grasses supplying fodder
Soil under the canopy is drier during the rainy season but wetter during the dry
season
Negative (؊)
Solar radiation is reduced 30–60% from the canopy
Phosphorus and calcium in the soil decrease in shade
Produces low dry matter content but higher water content in shade
Leaves become etiolated under shade
Computed from available literature on TCI
(LER) calculations, canonical variate analysis, spatial heterogeneity tests,
and diversity/dominance stability indices much attention has been paid to
evaluating the positive and/or negative effect of interaction of abiotic and
biotic components of grassland ecosystems. Many workers have also exam-
ined the basis of positive or negative interactions between aboveground
species “A” and “B” for physical resources. Now it is necessary to develop
and use concepts of biological interactions for optimization of biological
productivity.
The two most likely mechanisms of plant competitions are exploitation
(resource competition) and interference (allelopathic competition). Plant-
plant interactions are not direct effects of one plant on another, but indirect

effects on the environment acting via the response. So far, interspecific resource
competition, changes in the density, or biomass of a plant species are likely to
affect the availabilities of various resources in the environment and thus influ-
ence the growth of other species indirectly. As the number of individuals and
diversity of the species increases, the quantum of resources is lowered because
of a higher rate of resource consumption; hence, for the sustainable use of
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 132
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 133
resources aboveground (light) and belowground (water and nutrients), bio-
logical interaction studies are prerequisites. Mechanistic and process-based
uptake transfer and release functions of ecosystem level components (Figure
6.2) and interlinked resource use efficiency must be worked out for under-
standing the mechanisms of component-component interactions.
Interaction of Trees and Grasses
The interaction of trees and grasses is a kind of intergeneric interaction in
which microclimate and soil are the two important aspects through which
interactions are effected. Trees are grown in close proximity to pasture (called
silvipasture/savanna, trees in natural grasslands), where the biophysical and
biological interactions take place both aboveground and belowground.
Among those interactions, some are negative (4Ϫ) and some are positive
(12ϩ) interactions relating to physical resources (summarized in Table 6.3).
The relationship between the tree layer and grass cover has been studied by
many authors (Huxley, 1983; Kennard and Walver, 1973; Scanlan, 1992;
Belsky, 1992; Belsky et al., 1993) covering a wide range of aspects. Belsky et
al. (1989) established positive and negative interactions between grassland
trees in one of the studies in semi-arid savanna in Kenya.
Tree-grass is an example of the competition (Ϫ) and facilitation (ϩ) types
of interaction which most often occur together. Trees may compete with
grasses for light, tending to reduce grass yield (dry matter) through shading
while simultaneously increasing soil organic matter, hence soil moisture con-

tent, and the availability of nutrients for the grasses through decomposition
of tree leaf litter. It is the sum of negative and positive biological interactions.
Aboveground Interactions
Interspecific and intergeneric aboveground interactions are discussed
mainly in relation to light (solar radiation) and litter fall. It is a general
assumption that when plants are grown together in a community, they will
affect each other and there will be an interference or competition (Harper,
1961). According to Beets (1982), competition results from the reaction of one
plant/species on the environment (abiotic factors) as well as the effect of the
modified environmental factors upon its competitors. In tropical countries,
the aboveground interaction’s net results are positive due to plenty of sun-
light, favorable temperature, and litter biomass (and its amelioration in the
soil). However, C4 tropical grasses are more sensitive to light interception
compared to C3 grasses and legumes (Ludlow, 1978). Evans et al. (1992)
noted that generally in legumes and grasses, high levels of shade will encour-
age plants to become more etiolated, growing taller in an effort to gain better
access to available light. Also, leaves become larger and thinner, decreasing
the density and digestible fraction, which is one reason why cattle prefer to
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 133
134 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Table 6.4 Interspecific and Intraspecific Categories of Biological Interactions
Net results denoted by the signs (
؉), (؊), or
Interaction (0) for each of two species or two
⇓ Species ⇒ components (abiotic-biotic) of the ecosystem
Postive
Mutualism (ϩ, ϩ) Symmetric, reciprocal enhancement
of fitness.
Commensalism (
ϩ, 0) Asymmetric, only one individual

benefits from the association.
Negative
Competition— (Ϫ, Ϫ) Harmfully reduces the fitness of both
harmful the groups.
Competition— (
Ϫ, 0) Asymmetric, reduces the fitness of
asymmetric only one group. Other group is
unresponsive.
Predation (
Ϫ, ϩ) Preying of one group on another,
results in enhanced fitness of the
predator at the expense of other.
graze in open grasslands. Shade-tolerant grasses and legumes have positive
interaction (ϩ) in relation to quality and quantity of tropical grasslands.
Current research (up to 1999) on aboveground interaction in tropical grass-
land can be summarized by saying that nitrogen-deficient tropical grasses
showed enhanced growth/yield under shade (mostly of leguminous trees)
due to more favorable microclimate at the soil surface/litter interface, and
soil moisture particularly appeared to be responsible for improved litter
breakdown and possibly soil mineralization activity.
Belowground Interactions
Studies on root interactions between species have received less attention
because of the difficulties of studying root growth. Recent advances in tech-
nologies such as the miniaturization of video cameras to enable image analy-
sis and the use of NMR imaging are helpful aids which overcome some of the
limitations of rhizotroms. Campbell (1989) and Campbell and Dawson (1991)
studied the importance of root interaction for grasses and trees, and also the
belowground competition, and concluded that higher rooting densities pro-
moted competition for nutrients due to decreasing interroot distances.
Woody perennials have deeper root systems which absorb nutrients from

weathering rock/subsoil zone and are returned to soil via litter. Tree-grass
root/belowground interaction is a good example of commensalism’s having
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 134
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 135
a positive interaction between species called the “facilitative production
principle” where the environment of one species is modified positively by
another species (woody perennial) so that the first species benefits from the
presence of the second.
Grass-Legume Interactions
Only 12–15 genera of tropical grasses (family Graminae/poaceae) are
widely associated with more than 100 genera of legume plants of
Papilionaceae and woody species of the family Mimosoideae. Some herbaceous
legumes are creeping types and root at the nodes of stolons but can also react
to sunlight by climbing on the associated grasses. Most legume nodules on
the root systems may develop nitrogen-fixing bacteria of the genus Rhizobium
from a symbiotic relationship with the plants. Large amounts of nitrogen
(50–500 kg N/ha/year; Young, 1989, and Shelton, 1990) may be fixed
through the action of root nodules; hence, the legume has a mutualistic (ϩ,ϩ)
interaction with grasses. Good grazing management is needed to maintain a
fair proportion of them in a grassland. Rao and Giller (1993) reported that
the nitrogen fixed by tropical legumes is transferred to associate grasses.
Besides this, because of their high protein content, legumes improve tropical
grasslands, thus having a direct bearing on the level of animal production
(grass-legume-animal interaction; Jones, 1972).
Trees-Grass-Livestock Interactions
Associations among livestock, grasses, and trees are intense. The live-
stock component of the grassland may be herds and flocks grazing and
browsing in the vicinity of grazing lands and a mutually beneficial associa-
tion (fodder-grassland manuring). In man-managed grounds or a silvipas-
toral system, trees or shrubs (collectively called trub) are combined with

livestock and pasture production on the same land management unit. In this
system, trees play an interactive role in animal production by providing shade
and fodder to livestock and enhancing grassland productivity. Livestock
forms a major component of grasslands contributes 30–40% of the agricultural
GDP in west African countries. In India, 196 million cattle and 80 million buf-
falo, which account for 15% and 52% of world totals of these animals, are used
for milk and draft power (India, ranking first in milk production by produc-
ing 74 million tonnes of milk). Thus, tree fodder and browsing systems in
tropical grasslands are more common in drier areas. According to one esti-
mate (FAO, 1995), shrubs and trees in silvipastoral production systems con-
stitute the basic feed resource for more than 3811.6 million livestock in the
tropics and subtropics, out of the 4204 million head of livestock in the world
(Table 6.4). The importance of the pasture-cattle-coconut system in southeast
Asia and the Pacific (Reynolds, 1995), silvipastoral systems in Africa
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 135
136 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
(Le Houerou, 1980), silvipasture in India (Singh and Roy, 1998), pasture
under cashew plantation in Kenya (Goldson, 1981), pasture under Caribbean
pine in Fiji (Bell, 1981), and pasture under guava in Costa Rica (Somarriba
and Lega, 1991; Somarriba, 1988a,b) has been extensively discussed. Trees
and livestock—a productive co-existence—have also been discussed at
length by Moore (1993).
Tree/Grass-Legume-Animal Interactions
It is practically axiomatic that tropical grassland productivity, if not
managed in a sustainable manner, will lead to the degradation of biological
and biophysical natural resources in general, and grassland soils in particu-
lar. Soil carbon and nutrients (particularly nitrogen) are major determinants
of sustainability of grazing lands. As the biological interactions, discussed in
earlier pages of this chapter, are a complex natural phenomenon, they
should not be taken in isolation; it is an integrated complex system in which

not only the three major components (trees, grass, animal) are included, but
soil microbes and arthropods also play vital roles in carbon circulation. Both
plants (grasses and trees) and animals provide inputs of organic matter to
soils. The amount of litter in semi-arid grasslands is usually more than
3 t/ha (Klemmedson, 1989) which undergoes rapid decomposition and
inherently adds higher levels of organic matter in a tropical situation (Juo
and Payne, 1993). The relative importance of litter, root and manure (cow
dung and urine) as inputs of organic matter varies between grassland types.
Figure 6.5 illustrates the flow of carbon in a livestock-based tropical grass-
land ecosystem where trees/grass and animals interact with each other.
Generally, all the root material (about 40% of the total herbage) and 10–30%
of aboveground phytomass may be recycled in a grazing land system in the
form of litter. The rate of breakdown of litter and other organic debris deter-
mines and depends on the populations and interactions of organisms in the
soil. This breakdown also determines the extent to which minerals taken up
by grass/trees are released from their organic residues and made available
for decomposition. There are numerous mathematical models of organic
matter decomposition. Almost all assume that the rate of decomposition
decreases with time and increases with favorable biophysical variables, par-
ticularly soil temperature and soil water, and progressively broken down by
soil organisms (Singh and Gupta, 1977; Upadhyaya and Singh, 1981).
Decomposition also varies with plant type and age of litter. The specific
properties of litter from different species and the generally exponential form
of litter decay lead to different values of decomposition from different tropi-
cal grasslands. Because of the importance of temperature in determining
decomposition, the rate declines from tropical to temperature grasslands.
Plant species, water, temperature, and microbes are not only factors affecting
the rate of litter decomposition but also that of carbon flow and mineral
920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 136
BIOLOGICAL INTERACTION IN TROPICAL GRASSLAND ECOSYSTEMS 137

ASSIMILATION
BY PLANTS
CARBON DIOXIDE
IN ATMOSPHERE
NUTRIENT UPTAKE
FROM DEEPER LAYER
FOOD FOR
ANIMALS
COW DUNG
+ URINE
TLU
SUNDRY
GASES
ANIMAL
RESPIRATION
SOIL
RESPIRATION
PLANTS
RESPIRATION
BROWSE+
CUT AND CARRY
FODDER
GRASS +
LEGUME
DECOMPOSITION
TREES
MICROBES
N-FIXATION
ROOT
RESPIRATION

LITTER
NUTRIENT RELEASE
Figure 6.5 Tree/grass-legume-animal interaction in relation to organic matter
turnover in a typical tropical grassland ecosystem. Arrows represent the
flow of carbon.TLUϭ Tropical Livestock Unit Le. 250 kg live weight of ani-
mal.
cycling. Manipulation, management, and proper utilization of biological
interactions play a vital role and are the researchable issues that need to be
further studied.
Schlesinger (1977) pointed out that the amount of carbon in biosphere
detritus is enormous but poorly evaluated. Likewise, the amount of CO
2
;=
released from the detritus is great but remains poorly evaluated in terms of
the CO
2
atmospheric balance. For grassland ecosystems, data on detritus
accumulation and turnover are extremely limited. Patterns of detritus carbon
accumulation vary from tropical to temperate grasslands (Table 6.5). The
mean detrital values of temperate grasslands are higher (19.90 kg C m
Ϫ2
) than
those of tropical grasslands (2.30, kg C m
Ϫ2
). The low values in tropical grass-
lands presumably reflect high rates of decomposition and soil respiration
(Singh and Upadhyaya, 1999) subsequently released higher amounts of car-
bon which compensate for the high productivity and litter fall in tropical as
compared to temperate grasslands. This ecosystem function of tropical grass-
lands is positively correlated with the biophysical interactions. Soil microbes

920103_CRC20_0904_CH06 1/13/01 10:51 AM Page 137