Black & Ong
Black & Ong
Black & Ong
Abstract
The resource capture approach developed by John Monteith has been applied in studies of a wide variety of plant species
and cropping systems in the tropics over the past 18 years. The purpose of this review is to highlight the progress made
and the new challenges which lie ahead. The foundation for this approach was the establishment of ‘response surfaces’ for
the development and growth of tropical crops using controlled-environment facilities. The concepts of light interception and
thermal time developed were then used to investigate the mechanisms responsible for overyielding in intercropping systems
and genotypic differences in the drought adaptation of crops in the semi-arid tropics. The most significant achievements were
in the understanding of temporal and spatial complementarity in intercropping and agroforestry systems and the development
of plant growth models. More recently, the same concepts have been extended to the capture of below-ground resources in
agroforestry systems and rain forests. The most serious remaining challenge is to extend this approach to studies of complex
multispecies systems in the humid tropics. © 2000 Elsevier Science B.V. All rights reserved.
Keywords: Agroforestry; Intercropping; Resource capture; Light; Water; Temperature
0168-1923/00/$ – see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 8 - 1 9 2 3 ( 0 0 ) 0 0 1 4 5 - 3
26 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
by showing that accumulated dry matter production focus on recent attempts to apply these concepts
of a wide range of crops and orchards in Britain was to agroforestry. Finally, we highlight some of the
linearly related to accumulated intercepted solar radi- progress made and the new challenges ahead.
ation (Monteith, 1977a). The second was the applica-
tion of the thermal time concept to describe the effects
of temperature on crop development (Monteith, 1979). 2. Principles of radiation interception and use
These concepts were tested in the tropics in close
collaboration with the International Crop Research The capture of radiation and its use in dry mat-
Institute for the Semi-Arid Tropics (ICRISAT), Hy- ter production depends on the fraction of the incident
derabad, India, where JLM was Director of the Re- photosynthetically active radiation (PAR) that is in-
source Management Programme between 1987 and tercepted and the efficiency with which it is used for
1991, to assist in the more practical problem of es- dry matter production. Intercepted radiation (Si ) is of-
tablishing genotype–environment interactions for mil- ten estimated as the difference between the quantity
let and groundnut (Williams, 2000) and determine the of incident radiation (S) and that transmitted through
mechanisms responsible for overyielding in intercrop- the canopy to the soil (St ). However, this approach has
ping systems (Marshall and Willey, 1983; see also re- inherent technical and theoretical difficulties since is
views by Fukai and Trenbath, 1993; Ong and Black, does not account for the reflection of incident radiation
1994; Azam-Ali, 1995). Today these concepts have from the canopy surface (typically 5–20% depending
been applied throughout the tropics and subtropics and on surface characteristics and moisture content), or for
such studies are no longer confined to international radiation intercepted by non-photosynthetic canopy el-
centres; for example, Vijaya Kumar et al. (1996) ap- ements. As a result, interception by photosynthetically
plied these concepts to castor beans at the Central Re- competent tissues may be greatly overestimated, par-
search Institute for Dryland Agriculture, Hyderabad, ticularly for canopies which are senescing or contain
India. JLM’s resource capture concept has also been numerous woody structural elements. Corrections for
applied in forestry research (Landsberg, 1986; Can- these errors have often been ignored when estimating
nell et al., 1987), crop simulation models (Ritchie and Si and photosynthetic efficiency.
Otter, 1985; Jones and Kiniry, 1986), as a tool for The quantity of radiation intercepted depends on
biomass prediction in agronomic research (Sinclair the amount received by the canopy, its size and dura-
et al., 1992) and in remote sensing (Christensen and tion and fractional interception (f ). The seasonal time
Goudriaan, 1993). course of f, defined here as Si /S, varies greatly de-
Recently, the same approaches have been used to pending on canopy architecture and the phenology of
unravel the more complex interactions between trees the vegetation involved; thus, f increases more rapidly
and crops in agroforestry systems in studies of both in cereals such as sorghum (Sorghum bicolor) than in
above- and below-ground interactions (Ong et al., legumes such as groundnut (Arachis hypgaea), reflect-
1991; Ong and Black, 1994). Initial findings indicated ing their differing rates of leaf initiation and expan-
that there are critical differences between intercrop- sion (Squire, 1990). The variation in f between crops
ping and agroforestry which are apparently linked to is generally smaller than that in green leaf area index,
the relative importance of below-ground interactions partly because the extinction coefficient for radiation
(Ong et al., 1996). Even more formidable difficulties (k) is often larger in species whose canopy expands
were encountered when these essentially agronomic slowly; maximum f values may therefore differ little
plot size findings were extrapolated to farm and land- between crops grown under non-limiting conditions.
scape levels since the extensive lateral growth of tree Mean f values calculated over the duration of the crop
roots may lead to their extension across farm bound- (f¯) are generally lower in short-duration cereals (ca.
aries or between adjacent experimental plots (Hauser, 0.5) and legumes (ca. 0.15) than in perennial species
1993; Rau et al., 1993). (ca. 0.9), largely because of the differing duration of
In this review, we describe how John Monteith’s ground-cover (Squire, 1990).
concepts of resource capture and thermal time have The theory and experimental approaches required
been applied in tropical agricultural research and to quantify radiation interception by monocultures
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 27
containing uniform plants are now well-established In 1981, John Monteith (Monteith, 1981) suggested
and have been widely used in both temperate and that the basic principle underlying resource capture
tropical environments during the past three decades in mixed communities is that the complementarity or
(Monteith et al., 1994). These approaches assume that competitiveness of interactions between species will
the canopy is a heterogeneous arrangement of ran- depend on their ability to capture and use the most
domly distributed leaves, and that the spatial variation essential limiting growth resources effectively. This
in radiation transmission by the canopy is limited. theory has been the cornerstone of numerous stud-
However, these conditions are clearly not met in inter- ies of resource partitioning between the component
crops or agroforestry systems because of the extensive species of intercrops and agroforestry systems by his
horizontal and vertical variation in canopy structure team at ICRISAT and other groups. Thus, the capture
introduced by the intimate mixture of species with of the limiting resource (i.e. light, water or nutrients)
differing planting dates and arrangement, heights and will depend on the number, surface area, distribution
maturity dates. Canopy architecture is also constantly and effectiveness of the individual elements of the
changing in mixed cropping systems because of the canopy or root system of the monocrop or mixture
differing growth rates and canopy durations of the being examined.
component species. For example, compact legumes One of the earliest detailed studies of resource par-
growing adjacent to taller cereals initially experience titioning in agroforestry systems was that described
greater competition than in the equivalent monocrop by Monteith et al. (1991) and Corlett et al., 1992a,b.
because of the faster growth of the cereal, but subse- In this study in India involving a Leucaena leuco-
quently experience less competition for much of the cephala/millet (Pennisetum americanum) alley crop-
reproductive phase due to the earlier harvest of the ping system, tube solarimeters of a design originally
cereal component. developed under JLM’s guidance (Green and Deuchar,
The methodological problems involved in charac- 1985) were used to determine radiation interception
terising the spatial and temporal variation in radiation by both species (Table 1). Total intercepted radia-
interception are much greater in mixed communities tion during the 1986 rainy season was substantially
than in monocultures, and the partitioning of radia- greater in the alley cropping system than in either of
tion interception (and also water uptake) between the the monocrops, primarily because the presence of leu-
components of such systems has provided a major caena increased fractional interception during the early
challenge. Detailed reviews of radiation interception stages of the growing season, while the millet provided
in intercropping and agroforestry systems are given a more complete ground cover across the alleys during
by Keating and Carberry (1993) and Ong et al. (1996). the later stages of the season. Biomass production was
This spatial variability is relatively modest and ca- also substantially greater in this treatment. The alley
pable of resolution in annual row intercrops, as has millet intercepted 48% less radiation than monocrop
been demonstrated by several studies (e.g. Marshall millet due to its lower population on a system area ba-
and Willey, 1983; Azam-Ali et al., 1990), but is much sis and the shading effect of the leucaena. However,
greater in agroforestry systems, particularly those in- the reduction in biomass production by alley-cropped
corporating large overstorey trees. An additional com- millet was smaller than that in radiation interception
plexity is the extended time scale and ever-changing due to an increase in conversion efficiency (e), possi-
relationship between the tree and crop components. bly because the light saturation of photosynthesis as-
In annual intercrops, the growing season is normally sociated with drought occurred less frequently under
confined to a period of 100–150 days, whereas the partial shade. The conversion coefficient, defined here
trees in agroforestry systems are often active through- as the quantity of biomass produced per unit of inter-
out the year. The size of the trees and their influence cepted radiation (g MJ−1 ), provides a measure of the
on associated crops may also change rapidly from ‘efficiency’ with which the captured radiation is used
year to year. For example, in a recent trial at ICRAFs to produce new plant material; the alternative term ra-
Machakos Research Station in Kenya, Grevillea diation use efficiency (RUE) is also commonly used.
robusta trees reached a height of 6–8 m within 5 years In 1986, the hedges were regularly pruned to a
of planting. height of 70 cm to minimise shading of the millet,
28 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
Table 1
defined above (Monteith et al., 1991). These authors
Intercepted (total) solar radiation, above-ground biomass produc-
tion and biomass production per unit of intercepted radiation (e) suggested that this philosophy was applicable both to
(modified from Corlett et al., 1992b) monocrops and to the components of intercropping
Intercepted Biomass e (g MJ−1 )
and agroforestry systems, provided fi and e could be
radiation (t ha−1 ) determined for each component. They also proposed
(MJ m−2 ) that, if fractional interception by the whole system was
Rainy season (July– recorded, an overall system average value for e could
August 1986) be obtained and system performance analysed in terms
Sole millet 581 4.7 0.81 of the efficiency of radiation capture and utilisation to
Alley millet 300 3.1 1.03
produce biomass. The use of the acquired resources
Sole L. leucocephala 520 4.0 0.77
Alley L. leucocephala 510 4.0 0.77 was therefore assumed to depend on the conversion
Total alley system 810 7.1 0.81 coefficient of the species involved and environmental
Dry season (September
influences such as drought. A major advantage of ex-
1986–June 1987) pressing productivity in these terms is to emphasise
Sole L. leucocephala 1270 1.5 0.12 the apparent conservativeness of e under many con-
Alley L. leucocephala 1160 1.7 0.15 ditions. This approach has subsequently been widely
Year (July 1986– adopted in studies of resource partitioning in inter-
August 1987) cropping and agroforestry.
Total alley system 1970 8.8 0.45 Numerous studies of annual crops, and some with
Rainy season (July– perennial species, have demonstrated the existence
August 1987) of close correlations between dry matter production
Sole millet 504 5.0 0.98 and cumulative intercepted radiation. For example,
Alley millet 180 0.9 0.50
Sole L. leucocephala 861 7.1 0.82
Stirling et al. (1990) examined the impact of artifi-
Alley L. leucocephala 748 6.4 0.86 cial shade imposed on groundnut (Arachis hypogaea)
Total alley system 928 7.3 0.79 between the onset of peg initiation or pod-filling and
final harvest using bamboo screens. A close linear
correlation between above-ground biomass and cu-
but in 1987 they were allowed to grow unchecked, as mulative intercepted radiation was found in all treat-
is reflected by the substantially higher radiation in- ments, although the quantity of biomass produced
terception by leucaena in all treatments and the 64% per unit of intercepted radiation was substantially
reduction in interception by the alley-cropped millet greater when shading was imposed from peg initia-
relative to monocrop millet. The intense shading of tion onwards (Fig. 1). In the absence of stress, e is
the former was accompanied by a sharp reduction in often conservative, typically ranging between 1.0 and
e and an 82% reduction in above-ground biomass. 1.5 g MJ−1 for C3 species in temperate environments
These results reinforce the conclusion reached in other (Monteith and Elston, 1983; Russell et al., 1988),
studies that, although leucaena may be a successful 1.5–1.7 g MJ−1 for tropical C3 species (Kiniry et al.,
component of agroforestry systems in the humid trop- 1989; Monteith, 1990) and up to 2.5 g MJ−1 for trop-
ics, it is too competitive for successful adoption in ical C4 cereals under favourable conditions (Squire,
water-limited semi-arid regions (Singh et al., 1989; 1990). However, the work of Stirling et al. (1990)
Monteith et al., 1991). showed that e may vary substantially within a single
When growth is not limited by water or nutrient sup- season between 0.98 g MJ−1 in the unshaded control
plies, the quantity of biomass produced by monocrops and 2.36 g MJ−1 in crops shaded from peg initiation
is limited primarily by the quantity of radiation cap- onwards. Thus, plants in the latter treatment inter-
tured. JLM suggested that seasonal biomass accumu- cepted approximately one-quarter of the radiation
lation for a given species may be expressed as the received by the unshaded control, but converted this
time integral of the product Sfi e, where S denotes inci- to dry matter 2.4 times more efficiently.
dent radiation, fi the fractional interception on a given In contrast, in a study of millet grown in a Leucaena
day, and e the conversion coefficient for radiation as leucocephala alley-cropping system in India, Corlett
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 29
the line-planted and dispersed systems. However, the argued that the concept of radiation use efficiency
seasonal mean e value for monocrop pigeonpea was is over-simplistic, cannot improve our understanding
lower than that for line-planted or dispersed pigeon- of crop growth and is of limited value in predicting
pea (0.42, 0.54 and 0.56 g MJ−1 , respectively), with yield, and concluded that there was little evidence
the result that its substantial interception advantage that incident radiation is a critical limiting factor de-
was not reflected by a similar biomass advantage. termining crop growth under normal field conditions.
Monocrop groundnut intercepted 10–40% more radia- In support of this argument, Demetriades-Shah et al.
tion than line-planted or dispersed groundnut in 1989, (1992) presented data to show that the growth of
although this was again not matched by a similar broiler chickens could be correlated with estimates of
increase in biomass. radiation interception derived from the size and popu-
The population of the monocrop pigeonpea was re- lation of birds, even though the two variables were not
duced to 0.5 plants m−2 in 1990 because the initial causally related. They also contested the validity of
population was believed to exceed the optimum for establishing correlations between cumulative values
maximum productivity during the second year, when for biomass and radiation on the grounds that plot-
competition between neighbouring trees was expected ting cumulative values for randomly selected pairs of
to cause extensive mortality; this reduction also estab- numbers could produce apparently significant corre-
lished identical populations in all treatments. Intercep- lations, even though there was no linkage between the
tion by the 2-year-old monocrop pigeonpea was 70% numbers involved. They re-evaluated several experi-
greater than that for line-planted pigeonpea, but al- mental datasets and concluded that analysis of crop
most 50% lower than that for the dispersed pigeonpea, growth in terms of cumulative intercepted radiation
a pattern reflected by biomass production (Ong et al., and the conversion efficiency of solar energy during
1996). The combination of greater radiation capture dry matter production should be approached with
and lower e in the dispersed pigeonpea probably re- caution. A major plank in their argument was that
sulted from its bushy structure, which would have in- photosynthesis, and hence crop growth rate, depends
creased the fraction of radiation intercepted by stems on numerous soil, atmospheric and biological factors
and branches. The much greater total interception by of which radiation is only one component. They sug-
the dispersed system in 1990 was primarily due to the gested that good correlations will always be found be-
pigeonpea which captured over 80% of the incident ra- tween radiation interception and any growing object,
diation, although its substantial interception advantage even when radiation is not the limiting variable, and
was not reflected by increased biomass production be- so a close correlation between crop growth and radi-
cause of its relatively low e value. These observations ation interception may be expected even when light
are analogous to the leucaena/millet system described is not a major limiting factor. Thus, although solar
by Corlett et al. (1992b), in which the less efficient tree energy may be the most fundamental natural resource
component captured most of the incident radiation dur- for crop growth from a physical viewpoint, from a bi-
ing the second year, to the detriment of crop growth. ological viewpoint it is no more important than water,
The observed variability in experimentally deter- nutrients, CO2 or any other essential commodity. As a
mined e values contrasts with earlier views that e is result, analysis of crop growth in terms of its radiation
highly conservative except during severe water stress conversion coefficient may be inappropriate when
(Azam-Ali et al., 1989), but complies with more re- variables other than radiation are the primary limiting
cent suggestions that the assumption of a constant factor.
value within species or cultivars may be erroneous Vijaya Kumar et al. (1996) provided experimental
(Demetriades-Shah et al., 1992, 1994). Indeed, these support for this view when they showed that the con-
papers advanced some controversial views which version coefficient for rainfed castor beans (Ricinus
sparked considerable lively debate. These authors communis) was less stable than previously suggested.
comprehensively criticised the concept that biomass The values obtained varied from year to year and were
accumulation may be linked directly with cumu- influenced by sowing date, decreasing with lateness
lative intercepted radiation, and that meaningful e of planting within the range 0.79–1.10 g MJ−1 ; val-
values may be derived from such correlations. They ues recorded prior to flowering were more stable than
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 31
those obtained after flowering began. These variations environmental factors is likely to meet with lim-
were positively correlated with atmospheric satura- ited success. A reanalysis of data first published by
tion deficit (D) and wind speed and negatively corre- Kanemasu et al. (1990) was used to support this
lated with water availability and temperature. In view argument. JLM concluded that there is a sound
of these results, Vijaya Kumar et al. suggested that mechanistic basis for observations that biomass ac-
it would be necessary to incorporate the influence of cumulation often bears an almost linear relationship
weather conditions into the algorithms used to calcu- to intercepted radiation since much of the radiation
late biomass in crop simulation models in order to pre- captured by annual crops is intercepted by young
dict effects on growth and yield. Other authors have leaves with a photosynthetic efficiency that is almost
previously described the influence of physical vari- constant over a wide range of irradiances, stressed
ables such as atmospheric saturation deficit (Stockle or senescent leaves spend much of their time on
and Kiniry, 1990), temperature (Hammer and Vander- the linear phase of the photosynthetic light response
lip, 1989), water stress (Ong and Monteith, 1985) and curve, and there is often little seasonal variation in
biological factors such as phenology (Spitters, 1990; the average point at which leaves operate on the
Giauffret et al., 1991), CO2 fixation pathway and fol- light response curve. In his original discussion of
iar nitrogen content (Sinclair and Horie, 1989) on e. the concept conversion coefficients, JLM (1977a)
In a spirited defence of the validity, generality and presented the results of a theoretical analysis which
robustness of correlations between intercepted radia- demonstrated that e would be expected to be directly
tion and growth and the conservativeness of e, JLM related to the CO2 exchange rates of leaves, while
concluded that few of the arguments advanced by Sinclair and Horie (1989) subsequently described the
Demetriades-Shah et al. (1992) were invincible, but steps involved in the theoretical derivation of daily
that their paper would provoke crop scientists to con- e values. Their analysis demonstrated the close link-
sider more carefully how errors involved in measur- age between e and the light-saturated rate of CO2
ing intercepted radiation may be minimised and how exchange, the CO2 fixation pathway and respiratory
responses to stress may be interpreted in terms of sea- characteristics.
sonal changes in f and e. JLM stressed that, while some JLM’s arguments were supported by Kiniry (1994)
early papers concerned with the relationship between and Arkebauer et al. (1994), who suggested that
radiation capture and dry matter production, including Demetriades-Shah et al. (1992) had overlooked the
his own 1977a contribution, suggested that e was con- fact that many environmental stresses which limit
sistent between species within major groups and was growth act through physiological pathways directly
apparently less affected by stress than f, many subse- involving the photosynthetic process and its products.
quent reports showed that e is not invariably conserva- Arkebauer et al. (1994) advanced strong arguments to
tive, but may vary depending on growth stage, water show that e cannot be expected to be constant, even
and nutrient availability and the incidence of disease within a single species or genotype, in the face of
(Green, 1987; Garcia et al., 1988; Steinmetz et al., changes in other environmental variables. They sug-
1990; Bastiaans and Kropff, 1993). JLM suggested gested that ‘the concept of RUE is therefore a very
that, contrary to the view of Demetriades-Shah et al. powerful tool in understanding crop growth and pre-
(1992), this did not invalidate the concept but instead dicting crop yield’ since factors which affect canopy
highlighted the need to test and improve methodology and leaf photosynthesis and crop biomass accumula-
as new information becomes available. tion have a well defined influence on the value of e
In their original paper, Demetriades-Shah et al. expressed by the crop.
suggested that relative growth rate (RGR) or net Arkebauer et al. (1994) also dismissed criticisms
assimilation rate (NAR) may be preferable to us- that the reported variability in e is symptomatic of
ing cumulative radiation interception and conversion inherent inadequacies in the concept by arguing that
coefficients when analysing crop growth. However, these resulted largely from an inconsistent definition
Monteith (1994) argued that, since RGR and NAR of e and lack of consideration of its physiological
both decline systematically during crop growth, any basis. The definition of e involves three separate fac-
attempt to correlate these variables with climatic or tors: firstly, the type and energy content of the carbon
32 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
involved, i.e. net CO2 uptake by the canopy, total Modelling crop yields under water stress condi-
above-ground dry matter production, or total plant tions is important in understanding their growth un-
dry matter including roots and storage organs; sec- der normal field conditions. As indicated previously,
ondly, the way in which radiation is characterised i.e. e is influenced by stress factors because of its depen-
total incident solar radiation, intercepted shortwave dence on the gas exchange characteristics of leaves,
radiation, intercepted PAR or absorbed PAR; finally, which are in turn related to soil moisture status. Simple
the time scale over which e is calculated is extremely models in which the impact of soil water deficits has
important and may range from instantaneous through been related directly to e, crop growth and yield have
hourly, daily or weekly estimates to the seasonal time been used successfully to simulate growth in soybean
scale. Because widely differing definitions of e have (Muchow and Sinclair, 1991), maize (Muchow and
been adopted, the values obtained may be expected to Sinclair, 1991) and wheat (Amir and Sinclair, 1991b).
show substantial variation, contributing to the contro- For instance, Sinclair et al. (1992) used a soybean
versy over its conservatism. Thus, Begue et al. (1991) model incorporating e as a function of soil water con-
found that, when e was calculated for monocrop mil- tent to compare yields across years and locations with
let over 5 day intervals, the values declined during differing rainfall in Argentina.
the season from ca. 5.6 to 0.5 g MJ−1 PAR when Remotely-sensed, as opposed to ground-level mea-
expressed on a total above-ground basis, but the vari- surements, of annual PAR interception or even simple
ation was reduced when e was calculated for specific accumulated values for spectral vegetation indices
developmental stages, ranging from 5.0 g MJ−1 dur- have been correlated with annual biomass production
ing tillering to 2.3 g MJ−1 during maturation. Ong for a range of ecosystems including crops (Daughtry
and Monteith (1985) also summarised values calcu- et al., 1992) and semi-arid grasslands (Prince and
lated over extended periods and showed that e was Tucker, 1986), and similar approaches have been
highly conservative for three well watered pearl millet applied at the continental (Goward et al., 1985) or
crops, ranging from 2.15 to 2.37 g MJ−1 during the global scale (Potter et al., 1993; Ruimy et al., 1994).
pre-anthesis period; the values for two water-limited This approach was adapted by Goetz and Prince
crops were 2.0 and 1.5 g MJ−1 . The seasonal values (1996) who used a simplified canopy radiative transfer
were lower, ranging from 1.26 to 1.49 g MJ−1 for well model in combination with high resolution LAND-
watered crops and 1.14–1.17 g MJ−1 for water-limited SAT greeness vegetation index images to estimate
stands. annual interception of PAR by stands of quaking
Adverse environmental conditions may therefore aspen (Populus tremuloides) and black spruce (Picea
reduce e because of their adverse effect on photo- mariana) in Northeastern Minnesota. The relation-
synthetic activity. Although Demetriades-Shah et al. ship between estimated annual PAR interception and
(1992) pointed out that stress is commonly encoun- measured above-ground biomass production was then
tered in the field, Arkebauer et al. (1994) empha- used to calculate e. Annual PAR interception was
sised that, far from invalidating the approach, a major generally higher and more consistent between stands
strength of e is that it can be used to quantify the im- of aspen (600–1100 MJ m−2 per year) than in black
pact of stress factors by comparing the observed values spruce (100–1100 MJ m−2 per year), for which the
with those obtained under unstressed conditions. They values varied widely irrespective of age. Stand age
also suggested that, if radiation interception can be es- influenced the relationship between cumulative PAR
timated from models of canopy development, the cu- interception and biomass production in aspen, with
mulative product of intercepted radiation and e calcu- a strong correlation being found in young stands
lated on a daily basis would enable maximum growth and a weak correlation in mature stands, an effect
and yield potentials to be estimated for specific crops attributed to increasing maintenance respiration de-
and environments. This approach has proved useful in mands as the ratio of foliar to total biomass decreased
assessing the yield potential of various annual crops with age. The e values for aspen (0.44–1.29 g MJ−1
including maize (Zea mays; Jones and Kiniry, 1986), with a mean of 0.92 g MJ−1 ) were lower than those
soybean (Glycine max; Spaeth et al., 1987) and wheat for black spruce (0.17–0.89 g MJ−1 with a mean of
(Triticum aestivum; Amir and Sinclair, 1991a). 0.49 g MJ−1 ).
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 33
3. Principles of water uptake and use in the transpiring area are achieved over periods of
days or weeks through a combination of premature
The resource capture principles pioneered by JLM senescence and reduced production and/or expansion
may also be applied to water by breaking its utilisa- of new leaves and shoots. Irrespective of how they
tion down into ‘capture’ and ‘conversion efficiency’ are achieved, reductions in transpiration are generally
components. As for light, the quantity of dry matter accompanied by decreased assimilation and growth.
produced (W) depends on the quantity of water cap- Transpiration is influenced by the canopy and aero-
tured and the ‘efficiency’ with which it is used to pro- dynamic conductances (gc and ga ) or their reciprocal
duce dry matter. The ratio of dry matter production to resistances (rs and ra ). gc takes account of the phys-
water transpired, expressed on a unit leaf area or land iological and morphological attributes of the canopy
area basis, is known as the water use ratio (ew ), a term since it is usually calculated as the product of leaf area
equivalent to the conversion coefficient for radiation index (L) and leaf conductance (gl ) for the various lay-
(e). Using this analogy, dry matter production may be ers within the canopy (Azam-Ali, 1983; Black et al.,
expressed as W=ew 6EW , where 6EW represents cu- 1985). Leaf sheaths, pods, panicles and other green or-
mulative transpiration; as for light, W is often linearly gans must be included in such calculations since they
related to the quantity of water transpired, indicating contribute up to 30% of gc and canopy transpiration,
that ew is conservative (de Wit, 1958; Azam-Ali, 1983; and 15% of all water transpired during the cropping
Connor et al., 1985; Cooper et al., 1987). This rela- cycle (Batchelor and Roberts, 1983). Water use is fre-
tionship depends on the close linkage between CO2 quently controlled by regulation of canopy size rather
and water vapour fluxes which results from the role than leaf conductance during sustained drought (Ong
of stomata in regulating the exchange of both gases. et al., 1996); thus gc commonly determines transpira-
However, atmospheric saturation deficit (D) may exert tion from open, stressed or senescent canopies, while
a strong modifying influence on ew , as is considered ga may be limiting in dense canopies, particularly at
further below. low windspeeds.
Actively growing vegetation which is well supplied The balance between transpiration and absorption
with water exerts little control over water use and depends on both soil and atmospheric conditions.
transpires at rates determined by the prevailing evap- Well-watered vegetation generally transpires at rates
orative demand. The maximum rate of water use for close to the prevailing potential evaporation. Studies
prescribed environmental conditions has been termed at ICRISAT by members of JLM’s group showed that
potential evaporation (Eo ), which defines the upper in monsoon climates with Eo values of ca. 5 mm per
limit to the actual evapotranspiration (Et ), a term rarely day this requires mean absorption rates of 1.7–2.5 g
used by JLM; Et is smaller than Eo when ground cover H2 O m−1 root per day for groundnut or millet stands
is incomplete, the soil is dry or stress factors enforce with rooting densities of 2–3 km m−2 of land area
stomatal closure. Although Eo is determined primar- (Gregory and Reddy, 1982). In this and similar studies
ily by atmospheric conditions, Et is influenced by of rice (Oriza sativa; Yoshida and Hasegawa, 1982)
attributes of the vegetation which limit transpiration. and cassava (Manihot esculenta; Aresta and Fukai,
These may be subdivided into plant characteristics 1984), absorption was probably limited by Eo rather
which influence transpiration and rooting characteris- than rooting or soil characteristics, whereas the con-
tics which influence absorption, processes integrated verse applies in dry soil where absorption depends
by feed-forward and feed-back linkages. Control of more closely on the size and distribution of the root
water use at the canopy level involves both short- and system and soil water content than on atmospheric
long-term regulatory mechanisms, i.e. those which conditions. For example, Squire (1990) reported
protect against transient or more prolonged periods that transpiration from groundnut and millet grow-
of stress. Short-term responses operating over periods ing on drying soil was only 2–4 mm per day, even
of minutes or hours include stomatal closure to limit though total root lengths were greater than under well
transpiration and leaf movements to decrease radia- watered conditions. Thus, millet growing on stored
tion interception, leaf temperature and the leaf-to-air water in Niger required a mean absorption rate of
vapour pressure gradient. Longer term adjustments 0.65 g H2 O m−1 root per day over a total root length of
34 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
5.9 km m−2 to support a transpiration rate of 3.8 mm of two-species agroforestry systems which offer
per day; mean absorption rates were therefore up to promise.
four times lower than in moist soil. In some instances, Several studies have attempted to partition wa-
it may be difficult to establish whether transpiration is ter use in agroforestry systems by determining total
limited primarily by above- or below-ground factors. community water use using the soil water balance
For example, the simultaneous senescence of leaves approach, despite the practical limitations outlined
and roots during grain filling in determinate crops below, and measuring transpiration by one of its
affects both leaf area index and root length at a time components (e.g. the leucaena/millet and perennial
when the functional efficiency of these organs is also pigeonpea/groundnut systems described by Corlett
declining; in such cases it is difficult to disentangle et al., 1992a,b and Ong et al. (1996). In the water bal-
the causes and effects underlying observed changes ance approach (Wallace (1996) and Ong et al. (1996)
in water use. for reviews), total community water use is determined
Partitioning of water use between the components indirectly from the balance of all other components
of intercropping and agroforestry systems has posed of the water balance, namely precipitation, inter-
a continuing challenge. Three broad approaches may ception losses from the tree and crop canopies, soil
be used to determine water use and ew for the compo- evaporation, deep drainage and changes in soil wa-
nents of multispecies communities: (1) transpiration ter content within the rooting zone. Several of these
by each component is measured separately; (2) total terms are difficult or laborious to determine since
community water use and transpiration by one of the they exhibit extensive spatial and temporal variation
components are measured, leaving transpiration by the within multispecies communities, introducing sub-
other component to be calculated as the difference; stantial uncertainties into estimates of the combined
(3) transpiration may be estimated using transpiration transpiration of trees and crops. Soil water content is
models based on radiation interception by each com- generally determined using the neutron probe or time
ponent. domain reflectometry (TDR) approaches, while tree
Approach 1 is preferable since the values for each or crop transpiration may be measured using diffusion
component are determined separately and so are porometry, chamber systems, deuterium labelling or
subject only to the errors inherent in the techniques sap flow techniques. Porometry and chamber systems
involved. In approach 2, the estimates for each com- allow daily or seasonal time courses of water use to
ponent are not statistically independent and the values be constructed, but the measurements are discontin-
for the component derived by difference are influenced uous and labour-intensive; these approaches are also
by two sets of errors, those for community water use unusable when the foliage is wet, a major problem in
and those for the component whose transpiration was monsoon climates. Deuterium labelling avoids these
actually determined. However, until recently, option problems and permits transpiration to be measured
2 was widely regarded as the only realistic approach over periods of several days, but is relatively ex-
because the available methods were technically too pensive and requires specialised analytical facilities.
demanding, labour-intensive or expensive for routine However, the advent of inexpensive and reliable sap
season-long measurements. Approach 3 assumes that flow techniques suitable for field use has allowed con-
the Penman–Monteith equation (see below) may be tinuous, non-destructive measurements to be made
modified to calculate water use by each component for both trees and established crops over extended
provided its radiation interception can be measured or periods, as is discussed below.
estimated; transpiration may then be calculated using A major problem with the water balance approach
single or dual source transpiration models. Substantial is that it may be difficult to monitor soil water con-
progress has recently been made in this area. For ex- tent throughout the rooting zone due to the deep root-
ample, Shuttleworth and Wallace (1985) adapted the ing habit of trees and technical difficulties resulting
Penman–Monteith equation to allow for the interac- from the presence of stony or compacted horizons.
tive energy fluxes between the soil and sparse crops, This was the case for the leucaena/millet and grevil-
while Wallace (1995) proposed further modifications lea/maize systems examined at ICRISAT and ICRAF,
based on fractional interception by the components with the result that the exploitation of deep water
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 35
reserves by the trees could not be fully quantified, for calibration. The technique is particularly valuable
causing community water use to be underestimated for measuring sap flow in wood of differing age within
to an unknown extent. However, an excellent exam- the trunk, and integration over the entire sapwood area
ple of the successful application of the water balance can be achieved using multipoint probes. Cermak et al.
approach was reported by Eastham et al. (1988), who (1973) developed a method known as the trunk heat
installed access tubes to a depth of 5.6 m at various balance, which is suitable for large trunks, its main
distances from Eucalyptus grandis trees in a silvopas- advantage being that sap flow is calculated from the
toral system in Queensland, Australia. Measurements energy balance of a sector of functional xylem. The
of soil water content and concurrent information on heat balance approach initially devised by Viewig and
rainfall and deep drainage allowed community water Ziegler (1960) and later refined by Sakuratani (1981)
use to be calculated on both a short term and seasonal and Baker and van Bavel (1987) uses an external jacket
basis. Relationships between transpiration from the to supply heat at a known rate; the dissipation of this
pasture, open-pan evaporation and soil water content energy, or heat balance, is then resolved to estimate
established using lysimeters enabled transpiration by the convective flux resulting from heat transfer in the
the pasture to be calculated routinely, and the values xylem sap. Heat balance gauges may be used reli-
obtained were subtracted from community water use ably without direct calibration provided the effective
to estimate tree transpiration. This approach was used thermal conductivity of the system can be established
to show that transpiration was dominated by the pas- from measurements made under zero flux conditions,
ture component when the trees were widely spaced but but calibration against absolute, usually gravimetric,
the converse applied at higher densities, when the trees measurements of transpiration has often been adopted
formed a closed canopy. Water abstraction at depth for increased rigour.
increased with tree density and cumulative seasonal Heat balance and heat pulse systems have been used
transpiration greatly exceeded rainfall at the highest routinely for over a decade at ICRISAT and ICRAF
tree density in both years examined, suggesting that in various agroforestry systems (Marshall et al., 1994;
the system was not sustainable (Eastham and Rose, Howard et al., 1995, 1997; Lott et al., 1996). The sap
1990a,b). flow approach has the major benefit of allowing tran-
The development of sap flow techniques which pro- spiration by each component of agroforestry systems
vide direct, non-destructive measurements of transpi- to be followed continuously and reliably. For example,
ration by intact plants over extended periods has been Fig. 3 shows typical diurnal time courses for transpira-
an important factor in improving our understanding of tion from the line-planted and dispersed arrangements
resource partitioning in mixed communities. A range of perennial pigeonpea referred to previously. No dif-
of relatively straightforward and inexpensive sap flow ference between treatments was apparent during the
techniques now exist which permit continuous mea- dry season (Fig. 3a), shortly after the trees had been
surements of transpiration in species varying in size cut for fodder and their leaf area was small. However,
from small herbaceous plants such as rice (Sakuratani, during the rainy season (Fig. 3b), transpiration was
1990) to large forest trees (Granier et al., 1996), of- five times greater at midday in the dispersed pigeon-
ten with a claimed accuracy of ±5–10%. A detailed pea which, because of its wider spacing and larger
review of the principles and application of sap flow canopy, was able to exploit the soil profile more fully.
techniques is given by Smith and Allen (1996), while Transpiration tracked the diurnal time course for at-
Granier et al. (1996) evaluated the merits of several mospheric saturation deficit, although irradiance, soil
heat-based approaches. The heat pulse technique de- water content and leaf area were also important in de-
veloped by Hüber and Schmidt (1937) is the oldest termining water use. The daily values obtained in this
sap flow method, but is still widely used for both small way were used to calculate cumulative transpiration
plants and large trees, and has recently been adapted over extended periods (Fig. 3c), again emphasising the
for use with tree roots (Khan and Ong, 1996). Swanson consistently greater water use of the dispersed pigeon-
and Whitfield (1981) developed a theoretical correc- pea, which accounted for 60% of the annual rainfall as
tion for the disturbance caused by the presence of the compared to 30% in the line-planted treatment; these
sensor probes within the stem which avoids the need values exclude water use by the understorey ground-
36 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
Fig. 3. Typical diurnal time courses of transpiration and leaf to air vapour pressure difference for perennial pigeonpea during (a) the dry
and (b) the rainy seasons, and (c) the seasonal time courses for cumulative transpiration (modified from Marshall et al., 1994).
nut. Although cumulative transpiration was two-fold trees in an approach which has proved invaluable in
greater in the dispersed system, the water use ratio establishing the quantity of water extracted from the
(ew ) differed by <10%, with the result that the greater crop rooting zone, and hence the below-ground com-
productivity of the dispersed pigeonpea was directly petitive impact of trees on crop performance. The time
linked to its greater water use (Ong et al., 1996). courses for sap flow through the lateral roots of gre-
In recent studies at ICRAF, the heat balance and villea presented in Fig. 4 (Lott et al., 1996) show con-
heat pulse techniques were modified to measure water siderable diurnal variation and indicate that sap flow
movement through the tap or major lateral roots of decreased with increasing distance from the trunk; the
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 37
Fig. 4. Diurnal time courses of sap flow in lateral roots at distances of 50 (䉬), 115 (䊏) and 190 cm (䉱) from the trunks of Grevillea
robusta at Machakos, Kenya (modified from Lott et al., 1996).
trees nevertheless extracted significant quantities of ing curve to account for the influence of drought on
water from the crop rooting zone up to 190 cm from Et Although the Penman model proved suitable for
the trunk. Experiments in which the lateral roots were determining Eo , estimates of Et were less satisfac-
severed indicated that these 3 year old trees were tory, particularly during drought. A major advance
capable of extracting up to 80% of their water require- in the Penman–Monteith model was the inclusion of
ments from beneath the crop rooting zone, in agree- boundary layer and canopy resistance terms (ra and
ment with studies of the same species in which the soil rc , respectively) to take account of the environmental
around the trees was excavated to a depth of 60 cm and physiological factors which influence transpi-
(Howard et al., 1997). Studies of this type to establish ration, thereby removing the principal limitation to
the influence of soil and microclimatic factors on wa- the effectiveness of the earlier Penman model. Infor-
ter use contribute significantly to our understanding mation on net radiation (Rn ), atmospheric saturation
of the success or failure of specific agroforestry sys- deficit (D), air temperature (Ta ) and windspeed (U)
tems, and also provide valuable information for the are also required. The Penman–Monteith equation has
development and validation of improved resource subsequently been widely adopted to estimate evapo-
capture and tree and crop growth models. transpiration from open water surfaces and uniform
The development of the Penman–Monteith equa- full-cover canopies in a range of agricultural and nat-
tion (Monteith, 1963) marked a major landmark in ural systems. However, the equation is less useful at
our ability to model evapotranspiration from vege- a regional scale where surfaces are characterised by
tated surfaces (Campbell, 2000). In 1948, Penman a patchy combination of vegetation and soil, partic-
(Penman, 1948) proposed a model for determining ularly in arid and semi-arid regions. Difficulty may
potential evaporation (Eo ) from open water surfaces also be experienced when applying the approach to
based on physical variables which were either readily mixed or discontinuous vegetation due to the difficulty
measurable or available from meteorological records. of obtaining suitable values for ra and rc . However,
He later modified the model to estimate evapotrans- Leuning and Foster (1990) successfully estimated
piration (Et ) by including a term describing the ratio transpiration from single trees using a ventilated cham-
of Et from well-watered turf to evaporation from an ber, leaf energy budgets and the Penman–Monteith
open water surface (Et /Eo ), and establishing a dry- equation, but suggested that multiple net radiometers
38 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
and stratified sampling of the boundary layer and Penman–Monteith equation, remotely sensed mea-
stomatal resistances should be adopted. surements of surface temperature and the soil-adjusted
The Penman–Monteith model was first linked to re- vegetation index, and estimates of surface roughness.
motely sensed measurements by Jackson et al. (1981), Alternative approaches have been used success-
who combined the model with a one-dimensional fully to model the impact of seasonal water shortage
energy balance equation to obtain estimates of maxi- on water use and photosynthesis. For example, Sala
mum and minimum canopy temperatures; by compar- and Tenhunen (1996) used a mechanistically-based
ing these with measured foliage temperatures, they C3 photosynthesis model coupled with an empirical
were able to estimate the ratio of actual to potential stomatal model and a canopy model of light intercep-
evapotranspiration and infer the severity of water tion and microclimate to simulate transpiration and
stress. This approach was used to develop a crop wa- net photosynthesis by an oak forest in northern Spain.
ter stress index (CWSI) which has subsequently been Effective predictions of the diurnal and seasonal time
adopted for practical applications such as irrigation courses were achieved by altering a single variable
scheduling and yield prediction. within the model, a dimensionless factor describing
Moran et al. (1994) suggested that CWSI could the relationship between stomatal conductance and as-
be modified for partly vegetated surfaces by includ- similation rate, relative humidity and CO2 partial pres-
ing measurements of surface reflectance as well as sure at the leaf surface. This factor was linearly related
surface temperature. Based on the Penman–Monteith to predawn xylem water potential, a feature which
equation, they calculated the maximum and minimum proved useful in allowing canopy-level assessments of
soil temperatures associated with maximum and min- seasonal water use and CO2 fluxes. The model predic-
imum evaporation rates and coupled these values to tions of water use agreed to within 10% with exper-
a spectral vegetation index which was linearly corre- imentally determined values, even though the former
lated with ground cover to derive a water deficit index excluded transpiration by the understorey vegetation.
(WDI) covering all possible combinations of water The modelling approach therefore provided a realistic
availability and ground cover. They also demonstrated description of the diurnal and seasonal patterns of leaf
the value of the WDI approach for agricultural ap- and canopy responses as affected by water availability.
plications at the field scale, with an emphasis on Tournebize et al. (1996) also used a mod-
irrigation scheduling. elling approach to estimate water losses from a
Moran et al. (1996) extended this approach by using hedgerow/pasture system containing Giricidium
remotely sensed measurements of surface reflectance sepium and the C4 grass, Dichantium aristatum. Wa-
and temperature to enable the Penman–Monteith ter losses were estimated by determining the energy
theory to be applied to partly vegetated semi-arid balance of each component using a detailed model of
grassland at the local or regional scale without a pri- radiation transfer capable of estimating partitioning
ori knowledge of percentage groundcover or canopy between species (Sinoquet and Bonhomme, 1992)
resistance. They linked the Penman–Monteith equa- and a simplified model of heat and mass transfer. The
tion to the energy balance equation to estimate the energy balances were calculated from micrometeo-
surface temperatures associated with four states: bare rological data recorded within the canopy, while the
soil or full cover vegetation with evaporation rates stomatal conductances required to calculate transpi-
set at zero or potential. Linear extrapolations between ration were modelled as a function of photosynthet-
the soil temperatures for full ground-cover and bare ically active radiation. This approach allowed water
soil conditions were used to provide information for losses to be estimated separately for the soil, grass
the intermediate states based on measurements of and trees; good agreement was obtained between the
surface reflectance and temperature. They concluded modelled and measured time courses of transpiration
that this approach has potential for mapping evapo- from the monocrop grass and Giricidium irrespective
ration rates from heterogeneous landscapes and sug- of climatic conditions. Soil evaporation contributed
gested that, provided the vegetation type is known, up to 76% of evapotranspiration, but was lower in
the required inputs are spatially distributed measure- the agroforestry system because the trees reduced the
ments of the meteorological variables required for the quantity of radiation reaching the soil.
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 39
The importance for crop production in water-limited of the C4 species, sorghum and maize (Rees, 1986).
environments of maximising both the quantity of water As for e, direct comparisons between studies may
available and the proportion used for transpiration has be complicated by variation in factors such as atmo-
already been alluded to, but the ‘efficiency’ with which spheric saturation deficit (D). For example, Squire
absorbed water is used for dry matter production is (1990) reported that the seasonal mean ew value for
also critical. As for the radiation conversion coefficient groundnut decreased from 5.2 to 1.5 g kg−1 as mean
(e), the water use ratio (ew ) may be calculated over daytime D increased from 1 to 2 kPa. Similarly, ew
time scales ranging from instantaneous measurements for millet decreased from 6.4 to 2.1 g kg−1 as D in-
of the ratio of net CO2 and water vapour fluxes to sea- creased from 1.4 to 4.0 kPa (Azam-Ali et al., 1984;
sonal estimates based on dry matter accumulation and Squire et al., 1984). Squire (1990) concluded that D
water use. Long-term estimates are invariably lower is one of the most important factors limiting produc-
than short-term values determined under favourable tivity in dryland areas since dry matter production
conditions because of respiration, which may consume decreased at least twofold as D increased from 1 to
up to 50% of the photosynthate produced (Ong et al., 4 kPa. Monteith (1986) suggested that the product of
1996), and the impact of adverse environmental (e.g. ew and D (ew D) is often conservative, while Loomis
drought, nutrient availability) and biological factors and Connor (1992) used an analogous approach to
(e.g. pests, diseases). Below-ground biomass is rarely adjust ew for seasonal differences in relative humidity
included in calculations of ew , while above-ground to show that the adjusted values for C4 cereals were
biomass often declines as crops approach maturity due at least double those for C3 species; the adjusted val-
to senescence, leading to underestimation of ew . As ues showed little benefit of intensive 50-year breeding
for e, ew may be expressed as the ratio of biomass, programmes when calculated on the basis of total
yield or energy equivalents to the quantity of water biomass, but did when computed against economic
consumed; the first two options are often preferred yield because of increased partitioning to the grain.
because of their relative simplicity, but the latter is Thus ew values depend on the time scale over which
important when comparing species with contrasting they are calculated, whether they are corrected for
chemical composition, such as grain and oil crops. A seasonal or site-specific differences in D, and whether
further factor is that ew should be calculated on the ba- they are based on above-ground biomass, final yield or
sis of transpired water rather than evapotranspiration, energy content; the wide range of reported values em-
since water evaporated from the soil has no direct role phasises the need for consistency and care in the cal-
in dry matter production although it may influence the culation and interpretation of data for ew . For instance,
atmospheric saturation deficit experienced by the crop Vijaya Kumar et al. (1996) showed that ew decreased
and hence ew . Soil evaporation dominates evapotran- with lateness of planting and exhibited substantial
spiration during the early part of the growing season inter-annual variation when castor beans were grown
for annual crops and may comprise 30–60% of sea- under semi-arid conditions in India; the values ranged
sonal water use depending on the rate of canopy de- from 0.88–1.31 g kg−1 and were negatively correlated
velopment and maximum leaf area (Wallace, 1991). with D and air temperature. They concluded that accu-
The ew values for tropical C4 cereals are often a rate prediction of crop growth would require the effects
little more than double those for C3 species under of weather to be incorporated into simulation models.
equivalent conditions. For example, experiments in In- As for light, intercropping and agroforestry offer
dia under similar mean atmospheric saturation deficits substantial scope for spatial and temporal complemen-
(2.0–2.5 kPa) provided season-long values of 3.9 and tarity of water use resulting from improved exploita-
4.6 g kg−1 for millet (Squire et al., 1984), compared tion of available supplies. However, the opportunity
to 1.5–2.0 g kg−1 for groundnut (Ong et al., 1987; for significant complementarity is likely to be limited
Matthews et al., 1988; Azam-Ali et al., 1989). How- unless the species involved differ appreciably in their
ever, ew is not invariably higher in C4 species, since rooting patterns or duration; the former permits at least
similar values have been reported for drought tolerant partial exploitation of different soil volumes, while
C3 species such as cowpea and cotton (Gossypium the latter allows continued abstraction of moisture
barbadense) and relatively drought-sensitive cultivars following harvest of the shorter duration component.
40 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
Total water use by intercrops is frequently little differ- saturation point; in such instances, partial shade may
ent from monocrops, particularly when water losses have little effect on assimilation, with the result that
from land left bare after harvesting the shorter duration ew is improved by the concurrent reductions in tran-
component are taken into account (Morris and Garrity, spiration. These comments are less applicable to agro-
1993). These workers reviewed several experiments in forestry, in which the overstorey trees are invariably
which seasonal rainfall varied between 84 and 575 mm C3 species and the understorey species may be C4
and concluded that water use by intercrops was gen- cereals which do not respond favourably to shading.
erally within ±7% of equivalent monocrops, although
larger benefits were occasionally observed. For exam-
ple, Natarajan and Willey (1980a,b) found no differ- 4. Thermal time
ence in water use between monocrops and intercrops
of pigeonpea and sorghum up to the point when the years ago, at an international conference on soil
shorter duration sorghum was harvested. However, the physical properties and crop production in the tropics
longer duration pigeonpea extracted a further 170 mm held at Ibadan, Nigeria, JLM remarked that the litera-
prior to harvest 10 weeks later, utilising residual water ture on soil temperature published during the previous
and late-season rains that would otherwise have been 20 years had extended existing evidence that crops
lost. Nevertheless, although total water use by the in- respond to soil temperature without adding much
tercrop greatly exceeded that of monocrop sorghum, to our understanding of the underlying mechanisms
there was no advantage over monocrop pigeonpea. In (Monteith, 1979). At an earlier symposium on the
a similar study, Reddy and Willey (1981) showed that ecophysiology of tropical crops in Manaus, Brazil, he
water use by millet/groundnut intercrops exceeded that proposed that plant responses to temperature should
of the monocrops, primarily because a larger leaf area be analysed in terms of the reciprocal of the duration
index was maintained for longer. Substantial improve- of specific stages of development, rather than sim-
ments in water use may also occur when species with ply examining the rates of growth and development
complementary root distributions are used, as in the (Monteith, 1977b). He postulated a linear relationship
rice/pigeonpea system described by Jena and Misra between a base temperature, Tb , at which specific
(1988) in which the deep rooting pigeonpea extracted processes such as germination or primordial initia-
substantial quantities of water from below the rooting tion begin, and an optimum, To at which the process
zone of rice (Oriza sativa). proceeds at its maximum rate, and another linear but
The beneficial effect of intercropping generally declining relationship between To and a maximum
originates from improvements in ew rather than sea- temperature, Tm , beyond which development ceases.
sonal water use, to which several factors may con- These relationships have been confirmed by several
tribute. Intercropping may increase the proportion of studies of germination in both tropical and temperate
available water that is used for transpiration because species (Garcia-Huidobro et al., 1982a for pearl mil-
the presence of a C4 cereal results in more rapid let; Mohamed et al., 1988 for millet and groundnut).
canopy development and reduced soil evaporation; There is now general agreement that values of Tb and
however, this potential benefit must be offset against To lie within the range 9–13 and 27–32◦ C, respec-
the risk of premature depletion of available water prior tively, for a range of mainly tropical species (Angus et
to maturity. Fast growing C4 species with inherently al., 1981), but the precise range for Tm (41–47◦ C) has
high ew values may also capture more of the avail- been defined for only a few processes. The concept of
able water, thereby increasing the overall yield and thermal time which arose from these relationships, in
ew value for the system. The dominant and subsidiary contrast to earlier suggestions that plants may ‘accu-
components of many intercrops are, respectively, C4 mulate temperature’, has since been used to describe
and C3 species with relatively high and low ew val- the phenological responses to temperature of many
ues. Finally, intercropping may confer microclimatic crops, with particular success in terms of defining the
benefits including partial shade and reduced evapora- rate of progression towards key events such as flow-
tive demand on the shorter component, frequently a ering and maturity (Ong and Monteith, 1985; Roberts
C3 species with a relatively low photosynthetic light and Summerfield, 1987).
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 41
Fig. 6. Weekly daily soil temperatures under Parkia biglobosa (䊏) and Butyrospermum parkii (䊐) or in the open (䊊) during the cropping
season at Sapone, Burkina Faso (modified from Jonsson, 1995).
grown under shade rarely experienced temperatures ‘conservativeness’ as JLM would call it, rather than
above 40◦ C. The influence of such temperature effects being preoccupied by the myriad of differences be-
on crop growth and yield needs to be separated from tween species? It is commonly assumed that the di-
that of drought, although there is strong evidence from versity of life forms in tropical forests contributes to
field experiments in northern Nigeria that high soil their high resource exploitation, and research is cur-
temperatures reduce both leaf growth and the radia- rently under way to assess the links between comple-
tion conversion coefficient for pearl millet (McIntyre mentarity of resource use and the role of biodiversity
et al., 1993). in sustaining ecosystem processes (Ewel and Haggar,
1997). Their early results for three fast-growing tree
species and two understorey perennial monocrops
5. Progress and challenges ahead have confirmed previous findings in intercropping and
agroforestry research that the potential for increased
There is no doubt that JLM’s concepts of resource ecosystem productivity and resource capture results
capture and thermal time have had a profound influ- from of the inability of the dominant species to utilise
ence on the understanding of and research approaches available resources completely. This conclusion is
to tropical agriculture. Progress in tropical agro- consistent with the central hypothesis of agroforestry
forestry is more advanced, and perhaps more urgent advanced by Cannell et al. (1996) that ‘trees must
at present, than in temperate agroforestry. However, acquire resources that the crop would not otherwise
many challenges still lie ahead in tropical agricultural acquire’.
research. So far, most studies have considered only Another major challenge facing tropical biologists
two species growing simultaneously in intercropping is how to look beyond the plot and farm in order to
or agroforestry systems, but the question remains of deal with the ‘interactions’ between the mosaic of land
how to deal with the multi-species ecosystems which uses at the landscape, regional and global scale. The
typify the vast majority of tropical agriculture. Per- conventional approach is to stratify each area and sum
haps we could make more rapid progress by focusing the zones in maps as in the standard GIS methodol-
on the similarity of plant responses or attributes, ogy, which implies that the factors involved in scaling
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 43
up are proportional to the area occupied by each zone. Azam-Ali, S.N., Gregory, P.J., Monteith, J.L., 1984. Effects of
For example, agroforestry practices such as boundary planting density on water use and productivity of pearl millet
planting and the use of widely scattered trees may cre- (Pennisetum typhoides) grown on stored water. 2. Water use,
light interception and dry matter production. Exp. Agric. 20,
ate extensive interactions both above-ground in terms 215–224.
of microclimatic modifications and below-ground in Azam-Ali, S.N., Simmonds, L.P., Nageswara Rao, R.C., Williams,
terms of resource capture since tree roots may extend T.H., 1989. Population, growth and water use of groundnut
20–60 m from the trunks. Here, more progress may maintained on stored water 3. Dry matter, water use and light
interception. Exp. Agric. 25, 77–86.
be made by defining relevant concepts and establish-
Azam-Ali, S.N., Matthews, R.B., Williams, T.H., Peacock, J.M.,
ing general principles, as JLM would argue, than by 1990. Light use, water uptake and performance of individual
collecting yet more empirical information. The con- components of a sorghum/groundnut intercrop. Exp. Agric. 26,
cept that the fractal dimensions of roots and shoots 413–427.
may be used to explore such interactions (van No- Baker, J.M., van Bavel, C.H.M., 1987. Measurement of mass flow
in the stems of herbaceous plants. Plant Cell Environ. 10, 777–
ordwijk and Ong, 1997) provides an excellent exam- 782.
ple of how such principles may be applied in future Bastiaans, L., Kropff, M.J., 1993. Effects of leaf blast on
research. photosynthesis of rice. 2. Canopy photosynthesis. Neth. J. Plant
Path. 99, 205–217.
Batchelor, C.H., Roberts, J.M., 1983. Evaporation from irrigation
water, foliage and panicles of paddy rice in north-east Sri
Acknowledgements Lanka. Agric. Meteorol. 29, 11–16.
Begue, A., Desprat, J.F., Imbernon, J., Baret, F., 1991. Radiation
Sections of this review were prepared during a use efficiency of pearl millet in the Sahelian zone. Agric. For.
visit to ICRAF by CRB with partial support from Meteorol. 56, 93–110.
Bell, M.J., Wright, G.C., 1998. Peanut growth and development
the UK Overseas Department For International De- in contrasting environments. 2. Heat unit accumulation and
velopment (DFID) Grant R5810 “Resource utilisa- photo-thermal effects on harvest index. Exp. Agric. 34, 113–
tion by trees and crops in agroforestry systems.” 124.
Thanks also to Jackie Johnson and Jackie Humphreys Black, C.R., Tang, D.-Y., Ong, C.K., Solon, A., Simmonds, L.P.,
1985. Effects of soil moisture stress on the water relations and
for excellent secretarial support in preparing this
water use of groundnut stands. New Phytol. 100, 312–318.
manuscript. Campbell, G.S., 2000. Monteith symposium: topics in
environmental physics. Agric. For. Meteorol. 104, 1–4.
Cannell, M.G.R., Milne, L., Shepherd, J., Unsworth, M.H., 1987.
References Radiation interception and productivity of willow. J. Appl. Ecol.
24, 261–278.
Cannell, M.G.R., Van Noordwijk, M., Ong, C.K., 1996. The central
Amir, J., Sinclair, T., 1991a. A model of the temperature and
agroforestry hypothesis: the trees must acquire resources that
solar-radiation effects on spring wheat growth and yield. Field
the crop would not otherwise acquire. Agrofor. Syst. 34, 27–31.
Crops Res. 28, 47–58. Cermak, J., Deml, M., Penka, M., 1973. A new method of sap flow
Amir, J., Sinclair, T.R., 1991b. A model of water limitation on rate determination in trees. Biol. Plant. (Praha) 15, 171–178.
spring wheat growth and yield. Field Crops Res. 28, 59–69. Christensen, S., Goudriaan, J., 1993. Deriving light interception
Angus, J.F., Cunningham, R.B., Moncur, M.W., Mackenzie, D.H., and biomass from spectral reflectance ratio. Rem. Sen. Environ.
1981. Phasic development in field crops. 1. Thermal response 43, 87–95.
in the seedling phase. Field Crops Res. 3, 365–378. Connor, D.J., Jones, T.J., Palta, J.A., 1985. Response of sunflower
Aresta, R.B., Fukai, S., 1984. Effects of solar radiation on the to strategies of irrigation. II. Growth, yield and efficiency of
growth of cassava (Manihot esculenta Crantz). 2. Fibrous root water use. Field Crops Res. 10, 15–26.
length. Field Crops Res. 9, 361–371. Cooper, P.J.M., Gregory, P.J., Tully, D., Harris, H.C., 1987.
Arkebauer, T.J., Weiss, A., Sinclair, T.R., Blum, A., 1994. Improving water use efficiency of annual crops in rainfed
In defence of radiation use efficiency: a response to farming systems of West Asia and North Africa. Exp. Agric.
Demetriades-Shah et al. (1992). Agric. For. Meteorol. 68, 221– 23, 113–158.
227. Corlett, J.E., Ong, C.K., Black, C.R., Monteith, J.L., 1992a.
Azam-Ali, S.N., 1983. Seasonal estimates of transpiration from a Above and below-ground interactions in a leucaena/millet alley
millet crop using a porometer. Agric. For. Meteorol. 30, 13–24. cropping system. 1. Experimental design, instrumentation and
Azam-Ali, S.N., 1995. Assessing the efficiency of radiation use by diurnal trends. Agric. For. Meteorol. 60, 53–72.
intercrops. In: Sinoquet, H., Cruz, P. (Eds.), The Ecophysiology Corlett, J.E., Black, C.R., Ong, C.K., Monteith, J.L., 1992b.
of Tropical Intercropping, INRA Editions, Paris, pp. 305–318. Above and below-ground interactions in a leucaena/millet
44 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
alley cropping system. 2. Light interception and dry matter Granier, A., Biron, P., Bréda, N., Pontailler, J.-Y., Saugier, B., 1996.
production. Agric. For. Meteorol. 60, 73–91. Transpiration of trees and forest stands: short and long-term
Daughtry, C.S.T., Gallo, K.P., Goward, S.N., Prince, S.D., Kustas, monitoring using sapflow methods. Global Change Biol. 2,
W.R., 1992. Spectral estimates of absorbed radiation and 265–274.
phytomass production in corn and soybean canopies. Rem. Green, C.F., 1987. Nitrogen nutrition and wheat growth in relation
Sens. Environ. 39, 141–152. to absorbed solar radiation. Agric. For. Meteorol. 41, 207–248.
de Wit, C.T., 1958. Transpiration and Crop Yields. Versl. Green, C.F., Deuchar, C.N., 1985. On improved solarimeter
Landbouwk. Onderz. 64.6, Wageningen. construction. J. Exp. Bot. 36, 690–693.
Demetriades-Shah, T.H., Fuchs, M., Kanemasu, E.T., Flitcroft, I.D., Gregory, P.J., Reddy, M.S., 1982. Root growth in an intercrop of
1992. A note of caution concerning the relationship between pearl millet/groundnut. Field Crops Res. 5, 241–252.
cumulated intercepted solar radiation and crop growth. Agric. Hammer, G.L., Vanderlip, R.L., 1989. Genotype-by-environment
For. Meteorol. 58, 193–207. interaction in grain sorghum. I. Effects of temperature on
Demetriades-Shah, T.H., Fuchs, M., Kanemasu, E.T., Flitcroft, radiation use efficiency. Crop. Sci. 29, 370–376.
I.D., 1994. Further discussions on the relationship between Hauser, S., 1993. Root distribution of Dactyladenia (Acioa) barteri
cumulated intercepted solar radiation and crop growth. Agric. and Senna (Cassia) siamea in alley cropping on Ultisol. 1.
For. Meteorol. 68, 231–242. Implication for field experimentation. Agrofor. Syst. 24, 111–
Eastham, J., Rose, C.W., 1990a. Tree/pasture interactions at a 121.
range of tree densities in an agroforestry experiment. I. Rooting Hawkins, R.C., Cooper, P.J.M., 1981. Growth, development and
patterns. Aust. J. Agric. Res. 41, 683–695. grain yield of maize. Exp. Agric. 17, 203–208.
Eastham, J., Rose, C.W., 1990b. Tree/pasture interactions at a Howard, S.B., Ong, C.K., Rao, M.R., Mathuva, M., Black,
range of tree densities in an agroforestry experiment. II. Water C.R., 1995. Partitioning of light and water in Leucaena/maize
uptake in relation to rooting patterns. Aust. J. Agric. Res. 41, agroforestry systems. In: Sinoquet, H., Cruz, P. (Eds.), The
697–707. Ecophysiology of Tropical Intercropping. INRA Editions, Paris,
Eastham, J., Rose, C.W., Cameron, D.M., Rance, S.J., 1988. The pp. 175–187.
effect of tree spacing on evaporation from an agroforestry Howard, S.B., Ong, C.K., Black, C.R., Khan, A.A.H., 1997. Using
experiment. Agric. For. Meteorol. 42, 355–368. sap flow gauges to quantify water uptake by tree roots from
Ewel, J.J., Haggar, J.P., 1997. Primary productivity and resource beneath the crop rooting zone in agroforestry systems. Agrofor.
partitioning in model tropical ecosystems. Ecology 78, 1211– Syst. 35, 15–29.
1221. Hüber, B., Schmidt, E., 1937. Eine Kompensations methode zur
Fukai, S., Trenbath, B.R., 1993. Processes involving intercrop thermoelektrischen Messung langsamer Stafströme. Bericht der
productivity and yields of component crops. Field Crops Res. Deutschen Botanischen Gesellschaft 55, 514–529.
34, 247–271. Jackson, R.D., Idso, D.B., Reginato, R.J., Pinter Jr., P.J., 1981.
Garcia, R., Kanemasu, E.T., Blad, B.L., Bauer, A., Hatfield, J.L., Canopy temperature as a crop water stress indicator. Water
Major, D., Reginato, R.J., Hubbard, K.G., 1988. Interception Resour. Res. 17, 1133–1138.
and use efficiency of light in winter wheat under different Jena, D., Misra, C., 1988. Effect of crop geometry (row
nitrogen regimes. Agric. For. Meteorol. 44, 175–186. proportions) on the water balance of the root zone of a
Garcia-Huidobro, J., Monteith, J.L., Squire, G.R., 1982a. Time, pigeonpea and rice intercropping system. Exp. Agric. 24, 385–
temperature and germination of pearl millet (Pennisetum 391.
thyphoides S&H). 1. Constant temperature. J. Exp. Bot. 33, Jones, C.A., Kiniry, J.R., 1986. CERES-Maize: A simulation of
288–296. maize growth and development. Texas A & M University Press,
Garcia-Huidobro, J., Monteith, J.L., Squire, G.R., 1982b. Time, College Station, TX, 194 pp.
temperature and germination of pearl millet (Pennisetum Jonsson, K., 1995. Agroforestry in dry savanna areas in Africa:
typhoides S&H.). 2. Alternating temperature. J. Exp. Bot. 33, interactions between trees, soils and crops. Ph.D. thesis,
297–302. Swedish University of Agricultural Sciences, Umea, Sweden.
Garcia-Huidobro, J., Monteith, J.L., Squire, G.R., 1985. Time, Kanemasu, E.T., Demetriades-Shah, T.H., Su, H., Lang, A.R.G.,
temperature and germination of pearl millet (Pennisetum 1990. Estimating grassland biomass using remotely sensed data.
typhoides S & H). 3. Inhibition of germination by short exposure In: Steven, M.D., Clark, J.A. (Eds.), Applications of Remote
to high temperature. J. Exp. Bot. 36, 338–343. Sensing in Agriculture, Butterworths, London, pp. 185–200.
Giauffret, C., Bonhomme, R., Dorvillez, D., Derieux, M., 1991. Keating, B.A., Carberry, P.S., 1993. Resource capture and use in
Conversion of intercepted radiation into aerial biomass for three intercropping: solar radiation. Field Crops Res. 34, 273–302.
maize genotypes: influence of plant density. Maydica 36, 25– Khalifa, F.M., Ong, C.K., 1990. Effect of supra-optimal
27. temperatures on germination of pearl millet (Pennisetum
Goetz, S.J., Prince, S.D., 1996. Remote sensing of net primary glaucum (L) R.BR) hybrids. Ann. Arid Zone Res. 29, 279–288.
production in boreal forest stands. Agric. For. Meteorol. 78, Khan, A.H.H., Ong, C.K., 1996. A low cost heat pulse method
149–179. for measuring tree root water uptake. Agrofor. For. 7, 19–22.
Goward, S.N., Tucker, C.J., Dye, D.G., 1985. North American Kiniry, J.R., 1994. A note of caution concerning the paper by
vegetation patterns observed with the NOAA-7 advanced very Demetriades-Shah et al. (1992). Agric. For. Meteorol. 68, 229–
high resolution radiometer. Vegetatio 64, 3–14. 230.
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 45
Kiniry, J.R., Jones, C.A., O’Toole, J.C., Blanchet, R., Cabelguenne, Monteith, J.L., 1990. Conservative behaviour in the response of
M., Spanel, D.A., 1989. Radiation-use efficiency in biomass crops to water and light. In: Rabbinge, R., Goudriaan, J., van
accumulation prior to grain filling for five grain crop species. Keulen, H., Penning de Vries, F.W.T., van Laar, H.H. (Eds.),
Field Crops Res. 20, 51–64. Theoretical Production Ecology: Reflections and Prospects.
Landsberg, J.J., 1986. Physiological Ecology of Forest Production. Pudoc, Wageningen, pp. 3–16.
Academic Press, London, UK, 198 pp. Monteith, J.L., 1994. Validity of the correlation between
Leuning, R., Foster, I.J., 1990. Estimation of transpiration by single intercepted radiation and biomass. Agric. For. Meteorol. 68,
trees: comparison of a ventilated chamber, leaf energy budgets 213–220.
and a combination equation. Agric. For. Meteorol. 51, 63–86. Monteith, J.L., Elston, J., 1983. Performance and productivity of
Loomis, R.S., Connor, D.J., 1992. Crop Ecology: Productivity and foliage in the field. In: Dale, J.E., Milthorpe, L. (Eds.), The
Management in Agricultural Systems. Cambridge University Growth and Function of Leaves. Cambridge University Press,
Press, UK, 600 pp. Cambridge, UK, pp. 499–518.
Lott, J.E., 1998. Resource capture and use in Semi- and overstorey Monteith, J.L., Marshall, B., Saffell, R.A., Clarke, D., Gallagher,
agroforestry systems. Ph.D. thesis, University of Nottingham, J.N., Gregory, P.J., Ong, C.K., Squire, G.R., Terry, A., 1983.
U.K. Experimental control of a glasshouse suite for crop physiology.
Lott, J.E., Khan, A.A.H., Ong, C.K., Black, C.R., 1996. Sap flow J. Exp. Bot. 34, 309–321.
measurements of lateral roots in agroforestry systems. Tree Monteith, J.L., Ong, C.K., Corlett, J.E., 1991. Microclimatic
Physiol. 16, 995–1001. interactions in agroforestry systems. For. Ecol. Manage. 45,
Marshall, B., Willey, R.W., 1983. Radiation and growth in an 31–44.
intercrop of pearl millet/groundnut. Field Crops Res. 7, 141– Monteith, J.L., Scott, R.K., Unsworth, M.U. (Eds.), 1994. Resource
160. Capture by Crops. Nottingham University Press, Loughborough,
Marshall, F.M., 1995. Resource partitioning and productivity of UK, 469 pp.
perennial pigeonpea/groundnut agroforestry systems in India. Moran, M.S., Clarke, T.R., Inoue, Y., Vidal, A., 1994. Estimating
PhD thesis, University of Nottingham, UK. crop water deficit using the relation between surface-air
Marshall, F.M., Black, C.R., Ong, C.K., 1994. Heat balance
temperature and spectral vegetation index. Rem. Sens. Environ.
measurements of transpiration in perennial pigeonpea–
41, 169–184.
groundnut agroforestry systems. In: Monteith, J.L., Scott,
Moran, M.S., Rahman, A.F., Washburne, J.C., Goodrich,
R.K., Unsworth, M.H. (Eds.), Resource Capture by Crops.
D.C., Weltz, M.A., Kustas, W.P., 1996. Combining the
Nottingham University Press, Loughborough, UK, pp. 426–429.
Penman–Monteith equation with measurements of surface
Matthews, R.B., Harris, D., Nageswara Rao, R.C., Williams,
temperature and reflectance to estimate evaporation rates of
J.H., Wadia, K.D.R., 1988. The physiological basis for yield
semi-arid grassland. Agric. For. Meteorol. 80, 87–109.
differences between four genotypes of groundnut (Arachis
Morris, R.A., Garrity, D.R., 1993. Resource capture and utilization
hypogaea) in response to drought. I. Dry matter production and
in intercropping: water. Field Crops Res. 34, 303–317.
water use. Exp. Agric. 24, 191–202.
Muchow, R.C., Sinclair, T.R., 1991. Water deficit effects on maize
McIntyre, B.D., Flower, D.J., Riha, S.J., 1993. Temperature and
yields modelled under current and greenhouse climates. Agron.
soil water effects on radiation use and growth of pearl millet
J. 83, 1052–1059.
in a semi-arid environment. Agric. For. Meteorol. 66, 211–227.
Mohamed, H.A., Clark, J.A., Ong, C.K., 1988. Genotypic Natarajan, M., Willey, R.W., 1980a. Sorghum-pigeonpea
differences in the temperature responses of tropical crops. J. intercropping and the effects of plant population density. I.
Exp. Bot. 39, 1121–1128. Growth and yield. J. Agric. Sci., Cambr. 95, 51–58.
Monteith, J.L., 1963. Gas exchange in plant communities. In: Natarajan, M., Willey, R.W., 1980b. Sorghum-pigeonpea
Evans, L.T. (Ed.), Environmental Control of Plant Growth. intercropping and the effects of plant population density. II.
Academic Press, New York, pp. 95–112. Resource use. J. Agric. Sci. Cambr. 95, 59–65.
Monteith, J.L., 1977a. Climate and the efficiency of crop Ong, C.K., Squire, G.R., 1984. Response to temperature in a stand
production in Britain. Phil. Trans. R. Soc. London. B. 281, of pearl millet. 7. Final number of spikelets and grains. J.
277–294. Exp. Bot. 35, 1233–1240.
Monteith, J.L., 1977b. Climate. In: Alvim, P. de T., Kozlowski, Ong, C.K., Monteith, J.L., 1985. Response of pearl millet to light
T.T. (Eds.), Ecophysiology of Tropical Crops. Academic Press, and temperature. Field Crops Res. 11, 141–160.
London, pp. 1–27. Ong, C.K., Black, C.R., 1994. Complementarity in resource use
Monteith, J.L., 1979. Soil temperature and crop growth in the in intercropping and agroforestry systems. In: Monteith, J.L.,
tropics. In: Lal, R., Greenland, D.J. (Eds.), Soil Physical Scott, R.K., Unsworth, M.H. (Eds.), Resource Capture by Crops.
Properties and Crop Production in the Tropics. Wiley, Bath, Nottingham University Press, Loughborough, UK, pp. 255–278.
UK, pp. 249–262. Ong, C.K., Simmonds, L.P., Matthews, R.B., 1987. Responses to
Monteith, J.L., 1981. Evaporation and surface temperature. Quart. saturation deficit in a stand of groundnut (Arachis hypogaea
J. Roy. Meteorol. Soc. 107, 1–27. L.). 2 Growth and development. Ann. Bot. 59, 121–128.
Monteith, J.L., 1986. Significance of coupling between saturation Ong, C.K., Corlett, J.E., Singh, R.P., Black, C.R., 1991. Above and
vapour pressure deficit and rainfall in monsoon climates. Exp. below-ground interactions in agroforestry systems. For. Ecol.
Agric. 17, 113–126. Manage. 45, 45–57.
46 C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47
Ong, C.K., Black, C.R., Marshall, F.M., Corlett, J.E., 1996. Singh, R.P., Van den Beldt, R.J., Hocking, D., Korwar, G.R.,
Principles of resource capture and utilisation of light and water. 1989. Alley cropping in the semi-arid areas of India. In:
In: Ong, C.K., Huxley, P.A. (Eds.), Tree-Crop Interactions: A Proceedings of a Workshop on Alley Cropping. 10–14 March
Physiological Approach. CAB International, Wallingford, UK, 1988, International Institute for Tropical Agriculture, Ibadan,
pp. 73–158. Nigeria, pp. 108–122.
Penman, H.L., 1948. Natural evaporation from open water, bare Sinoquet, H., Bonhomme, R., 1992. Modelling radiative transfer
soil and grass. Proc. Roy. Soc. Lond. A 194, 120–145. in mixed and row intercropping systems. Agric. For. Meteorol.
Potter, C.S., Randerson, J.T., Field, C.B., Matson, P.A., Vitousek, 62, 219–240.
P.M., Mooney, H.A., Klooster, S.A., 1993. Terrestrial ecosystem Smith, D.M., Allen, S.J., 1996. Measurement of sap flow in plant
production: a process model based on global satellite and stems. J. Exp. Bot. 47, 1833–1844.
surface data. Global Biogeochem. Cycl. 7, 811–841. Spaeth, S.C., Sinclair, T.R., Ohnuma, T., Konno, S., 1987.
Prince, S.D., Tucker, C.J., 1986. Satellite remote sensing Temperature, radiation and duration dependence of high
of rangelands in Botswana. Part II. NOAA AVHRR and soybean yields: measurements and simulation. Field Crops Res.
herbaceous vegetation. Int. J. Rem. Sens. 7, 1555–1570. 16, 297–307.
Rau, M.R., Muraya, P., Huxley, P.A., 1993. Observations of some Spitters, C.J.T., 1990. Crop growth models: their usefulness and
tree root systems in agroforestry situations and their graphical limitations. Acta Hortic. 267, 349–368.
representations. Exp. Agric. 29, 183–194. Squire, G.R., 1990. The Physiology of Tropical Crop Production.
Reddy, M.S., Willey, R.W., 1981. Growth and resource use studies CAB International, Wallingford, UK, 236 pp.
in an intercrop of pearl millet/groundnut. Field Crops Res. 4, Squire, G.R., Gregory, R.J., Monteith, J.L., Russell, M.B., Singh,
13–24. Piara, 1984. Control of water use by pearl millet (Pennisetum
Rees, D.J., 1986. The effects of population density and typhoides S&H). Exp. Agric. 20, 135–139.
intercropping with cowpea on the water use and growth of Steinmetz, S., Guerif, M., Delecolle, R., Baret, F., 1990. Spectral
sorghum in semi-arid conditions in Botswana. Agric. For. estimates of the absorbed photosynthetically active radiation
Meteorol. 37, 293–308. and light-use efficiency of a winter wheat crop subjected to
Ritchie, J.T., Otter, S., 1985. Description and performance of nitrogen and water deficiencies. Rem. Sens. Environ. 11, 1797–
CERES-Wheat: A user-oriented wheat yield model. In: Willis, 1808.
B. (Ed.), ARS Wheat Yield Project, ARS-38. US Department Stirling, C.M., Williams, J.H., Black, C.R., Ong, C.K., 1990. The
of Agriculture, Agricultural Research Service, Washington, DC, effect of timing of shade on development, dry matter production
pp. 159–175. and light use efficiency in groundnut (Arachis hypogaea L.)
Roberts, E.H., Summerfield, R.J., 1987. Measurement and under field conditions. Aust. J. Agric. Res. 41, 633–644.
prediction of flowering in annual crops. In: Atherton, J.G. Stockle, C.O., Kiniry, J.R., 1990. Variability in crop radiation-use
(Ed.), Manipulation of Flowering. Butterworths, London, UK, efficiency associated with vapour pressure deficit. Field Crops
pp. 17–50. Res. 25, 171–181.
Ruimy, A., Dedieu, G., Saugier, B., 1994. Methodology for the Swanson, R.H., Whitfield, W.A., 1981. A numerical analysis of
estimation of terrestrial net primary production from remotely heat pulse velocity theory and practice. J. Exp. Bot. 32, 221–
sensed data. J. Geophys. Res. 99, 5263–5283. 239.
Russell, G., Jarvis, P.G., Monteith, J.L., 1988. Absorption of Tournebize, R., Sinoquet, H., Bussiére, F., 1996. Modelling
radiation by canopies and stand growth. In: Russell, G., evapotranspiration partitioning in a shrub/grass alley crop.
Marshall, B., Jarvis, P.G. (Eds.), Plant Canopies: Their Growth, Agric. For. Meteorol. 81, 255–272.
Form and Function. Cambridge University Press, Cambridge, van Noordwijk, M., Ong, C.K., 1997. The landscape and beyond
UK, pp. 21–40. — scaling up complex agroecosystems: emerging new concepts,
Sakuratani, T., 1981. A heat balance for measuring water flux in objectives and methods. Agrofor. For. 7, 12–18.
the stem of intact plants. J. Agric. Meteorol. 37, 9–17. Viewig, G.I.I., Ziegler, I.I., 1960. Thermoelektrische Registrierung
Sakuratani, T., 1990. Measurement of the sap flow rate in stem of der Gescwindigkeit des Transpirationsstromes. Berichte
rice plant. J. Agric. Meteorol. 45, 277–280. Deutsche Botanische Gesellschaft 73, 221–226.
Sala, A., Tenhunen, J.D., 1996. Simulations of canopy net Vijaya Kumar, P., Srivastava, N.N., Victor, U.S., Gangadhar Rao,
photosynthesis and transpiration in Quercus ilex L. under the D., Subba Rao, A.V.M., Ramakrishna, Y.S., Ramana Rao, B.V.,
influence of seasonal drought. Agric. For. Meteorol. 78, 203– 1996. Radiation and water use efficiencies of rainfed castor
222. beans (Ricinus communis L.) in relation to different weather
Shuttleworth, W.J., Wallace, J.S., 1985. Evaporation from sparse parameters. Agric. For. Meteorol. 81, 241–253.
crops — an energy combination theory. Quart. J. Roy. Meteorol. Wallace, J.S., 1991. The measurement and modelling of
Soc. 111, 839–855. evaporation from semi-arid land. In: Sivakumar, M.V.K.,
Sinclair, T.R., Horie, T., 1989. Leaf nitrogen, photosynthesis, and Wallace, J.S., Renard, C., Giroux, C. (Eds.), Soil Water Balance
radiation use efficiency: a review. Crop Sci. 29, 90–98. in the Sudano-Sahelian Zone. Proc. Int. Workshop, Niamey,
Sinclair, T.R., Salado-Navarro, L., Morandu, E.N., Bodrero, N.C., Niger, February 1991. IAHS Publ. No. 199. IAHS Press,
Martignone, R.A., 1992. Soybean yield in Argentina in response Institute of Hydrology, Wallingford, UK, pp. 131–148.
to weather variation among cropping seasons. Field Crops Res. Wallace, J.S., 1995. Towards a coupled light partitioning and
30, 1–11. transpiration model for use in intercrops and agroforestry. In:
C. Black, C. Ong / Agricultural and Forest Meteorology 104 (2000) 25–47 47
Sinoquet, H., Cruz, P. (Eds.), The Ecophysiology of Tropical Williams, J.H., 2000. The implications and applications of resource
Intercropping. INRA Editions, Paris, pp. 153–162. capture concepts to crop improvement by plant breeding. Agric.
Wallace, J.S., 1996. The water balance of mixed tree-crop systems. For. Meteorol. 104, 49–58.
In: Ong, C.K., Huxley, P. (Eds.), Tree-Crop Interactions: A Yoshida, S., Hasegawa, S., 1982. Drought Resistance in Crops
Physiological Approach. CAB International, Wallingford, UK, with Emphasis on Rice. International Rice Research Institute,
pp. 189–234. Los Banos, Philippines.