Complementary use of resources is considered a strong driver of enhanced performance in mixed-species assemblages. In income-producing agroforestry systems, economically valuable species will ideally benefit from resource partitioning. In agroforests in southern India, we assessed soil water uptake depths of coffee and different shade tree species at the end of the dry season, when water availability is regarded critical. Water isotopes ratios (δD and δ18O) from soil and plant samples were analyzed and an isotope mixing model was applied. Results suggest that coffee plants drew water mainly from the top soil (56% from 0 to 20 cm), and to a lower extent from the subsoil. Jackfruit trees had very similar uptake patterns, which suggest competition. Mango trees showed reverse patterns of water uptake with soil depth, i.e. drew more than 75% from the subsoil, which suggests spatial complementarity to coffee. Within and across shade tree species, soil water uptake from the top soil increased with increasing diameter. We suggest that shade tree species choice and diameter regulation are viable options in management for soil water use complementarity.

We investigated soluble carbohydrate transport in trees that differed in their phloem loading strategies in order to better understand the transport of photosynthetic products into the roots and the rhizosphere as this knowledge is needed to better understand the respiratory processes in the rhizosphere. We compared beech, which is suggested to use mainly passive loading of transport sugars along a concentration gradient into the phloem, with ash that uses active loading and polymer trapping of raffinose family oligosaccharides (RFOs). We pulse-labeled 20 four-year old European beech and 20 four-year old ash trees with 13CO2 and tracked the fate of the label within different plant compartments. We extracted soluble carbohydrates from leaves, bark of stems and branches, and fine roots, measured their amount and isotopic content and calculated their turnover times. In beech one part of the sucrose was rapidly transported into sink tissues without major exchange with storage pools whereas another part of sucrose was strongly exchanged with unlabeled possibly stored sucrose. In contrast the storage and allocation patterns in ash depended on the identity of the transported sugars. RFO were the most important transport sugars that had highest turnover in all shoot compartments. However, the turnover of RFOs in the roots was uncoupled from the shoot. The only significant relation between sugars in the stem base and in the roots of ash was found for the amount (r2 = 0.50; p = 0.001) and isotopic content (r2 = 0.47; p = 0.01) of sucrose. The negative relation of the amounts suggested an active transport of sucrose into the roots of ash. Sucrose concentration in the root also best explained the concentration of RFOs in the roots suggesting that RFO in the roots of ash may be resynthesized from sucrose. Our results interestingly suggest that in both tree species only sucrose directly entered the fine root system and that in ash RFOs are transported indirectly into the fine roots only. The direct transport of sucrose might be passive in beech but active in ash (sustained active up- and unloading to co-cells), which would correspond to the phloem loading strategies. Our results give first hints that the transport of carbohydrates between shoot and root is not necessarily continuous and involves passive (beech) and active (ash) transport processes, which may be controlled by the phloem unloading.

Oil palm (Elaeis guineensis Jacq.) water use was assessed by sap flux density measurements with the aim to establish the method and derive water-use characteristics. Thermal dissipation probes were inserted into leaf petioles of mature oil palms. In the laboratory, we tested our set-up against gravimetric measurements and derived new parameters for the original calibration equation that are specific to oil palm petioles. In the lowlands of Jambi, Indonesia, in a 12-year-old monoculture plantation, 56 leaves on 10 palms were equipped with one sensor per leaf. A 10-fold variation in individual leaf water use among leaves was observed, but we did not find significant correlations to the variables trunk height and diameter, leaf azimuthal orientation, leaf inclination or estimated horizontal leaf shading. We thus took an un-stratified approach to determine an appropriate sampling design to estimate stand transpiration (Es, mm day(-1)) rates of oil palm. We used the relative standard error of the mean (SEn, %) as a measure for the potential estimation error of Es associated with sample size. It was 14% for a sample size of 13 leaves to determine the average leaf water use and four palms to determine the average number of leaves per palm. Increasing these sample sizes only led to minor further decreases of the SEn of Es. The observed 90-day average of Es was 1.1 mm day(-1) (error margin ± 0.2 mm day(-1)), which seems relatively low, but does not contradict Penman-Monteith-derived estimates of evapotranspiration. Examining the environmental drivers of Es on an intra-daily scale indicates an early, pre-noon maximum of Es rates (11 am) due to a very sensitive reaction of Es to increasing vapor pressure deficit in the morning. This early peak is followed by a steady decline of Es rates for the rest of the day, despite further rising levels of vapor pressure deficit and radiation; this results in pronounced hysteresis, particularly between Es and vapor pressure deficit.

We asked how shade tree admixture affects cacao water use in agroforests. In Central Sulawesi, Indonesia, cacao and shade tree sap flux was monitored in a monoculture, in a stand with admixed Gliricidia trees and in a mixture with a multi-species tree assemblage, with both mixtures having similar canopy openness. A Jarvis type sap flux model suggested a distinct difference in sap flux response to changes in vapor pressure deficit and radiation among cacao trees in the individual cultivation systems. We argue that differences originate from stomatal control of transpiration in the monoculture and altered radiation conditions and a different degree of uncoupling of the VPD from the bulk atmosphere inside shaded stands. Probably due to high sap flux variability among trees, these differences however did not result in significantly altered average daily cacao water use rates which were 16 L day−1 in the multi-species assemblage and 22 L day−1 in the other plots. In shaded stands, water use of single cacao trees increased with decreasing canopy gap fraction in the overstory since shading enhanced vegetative growth of cacao fostering transpiration per unit ground area. Estimated transpiration rates of the cacao tree layer were further controlled by stem density and amounted to 1.2 mm day−1 in the monoculture, 2.2 mm day−1 for cacao in the cacao/Gliricidia stand, and 1.1 mm day−1 in the cacao/multi-species stand. The additional transpiration by the shade trees is estimated at 0.5 mm day−1 for the Gliricidia and 1 mm day−1 for the mixed-species cultivation system.

Stable isotopes of oxygen and hydrogen are often used to determine plant water uptake depths. We investigated whether and to what extend soil moisture, clay content, and soil calcium carbonate influences the water isotopic composition.

Compared to monocultures, diverse ecosystems are often expected to show more comprehensive resource use. However, with respect to diversity–soil-water-use relationships in forests, very little information is available. We analysed soil water uptake in 100 tree clusters differing in tree species diversity and species composition in the Hainich forest in central Germany. The clusters contained all possible combinations of five broadleaved tree species in one-, two- and three-species clusters (three diversity levels), replicated fourfold (20 one-species, 40 two-species and 40 three-species clusters). We estimated soil water uptake during a summer dry period in 0–0.3 m soil depth, based on throughfall and soil moisture measurements with a simple budgeting approach. Throughout the whole vegetation period in 2009, soil water uptake was additionally determined at a higher temporal resolution and also for a greater part of the soil profile (0–0.7 m) on a subset of 16 intensive clusters. During the dry spell, mean soil water uptake was 1.9 ± 0.1 mm day−1 in 0–0.3 m (100 clusters) and 3.0 ± 0.5 mm day−1 in 0–0.7 m soil depth (16 clusters), respectively. Besides a slightly higher water use of Fraxinus clusters, we could not detect any effects of species identity or diversity on cluster water use. We discuss that water use may indeed be a conservative process, that differences in tree-species-specific traits may be compensated for by other factors such as herb layer coverage and tree spatial arrangement, and that diversity-driven differences in water use may arise only at a larger scale. We further conclude that with respect to stand water use "tree diversity'' alone is not an appropriate simplification of the complex network of interactions between species traits, stand properties and environmental conditions that have varying influence on stand water use, both in space and time.

Reliable estimates of water use by trees and other woody plants are crucial for an improved understanding of plant physiology and for water resource management. Since the 1980s, the thermal dissipation probe (TDP) method has been widely applied in trees, proved to be fairly accurate but is challenging in remote areas. Also in the 1980s, the deuterium (D(2)O or deuterium oxide) tracing method was proposed, which so far has less often been applied. However, deuterium tracing requires less sophisticated equipment in the field and new analytical methods reduce costs and increase sample throughput. The objectives of this study were (i) to compare plant water use estimates of the TDP and D(2)O method and (ii) to determine whether the D(2)O method is appropriate for assessing absolute magnitudes of plant water use. The two methods were employed on five tropical tree species and a bamboo species growing in a reforestation stand in the Philippines and an agroforestry system in Indonesia. For bamboo, an increase in D(2)O values in neighbouring, non-labelled culms suggests that injected D(2)O was partly redistributed among culms, which would seriously limit the accurate estimation of water use for the target culm. For trees, water use estimates resulting from the D(2)O tracing method were proportional to the TDP results (r(2) = 0.85, P < 0.001), but absolute values were, on average, about seven times higher. This overestimation may be due to the assumptions underlying the D(2)O tracing method, such as the conservation of tracer mass, not being met. Further, it cannot be excluded that underestimation of water use by the TDP method contributed partly to the observed difference. However, when considering known sources of error, a large part of the observed difference remains unexplained. Based on our results, the use of the D(2)O tracing method cannot be recommended without further experimental testing if absolute values of whole-plant water use are a major goal. However, the D(2)O tracing method appears suitable for answering other questions, such as relative differences in water use among trees, water redistribution among neighbours and internal water transport and storage processes in plants.

Agroforestry systems may play a critical role in reducing the vulnerability of farmers' livelihood to droughts as tree-based systems provide several mechanisms that can mitigate the impacts from extreme weather events. Here, we use a replicated throughfall reduction experiment to study the drought response of a cacao/Gliricidia stand over a 13-month period. Soil water content was successfully reduced down to a soil depth of at least 2.5 m. Contrary to our expectations we measured only relatively small nonsignificant changes in cacao (-11%) and Gliricidia (-12%) sap flux densities, cacao leaf litterfall (+8%), Gliricidia leaf litterfall (-2%), soil carbon dioxide efflux (-14%), and cacao yield (-10%) during roof closure. However, cacao bean yield in roof plots was substantially lower (-45%) compared with control plots during the main harvest following the period when soil water content was lowest. This indicates that cacao bean yield was more sensitive to drought than other ecosystem functions. We found evidence in this agroforest that there is complementary use of soil water resources through vertical partitioning of water uptake between cacao and Gliricidia. This, in combination with acclimation may have helped cacao trees to cope with the induced drought. Cacao agroforests may thus play an important role as a drought-tolerant land use in those (sub-) tropical regions where the frequency and severity of droughts is projected to increase.

In the tropics, reforestations and agroforestry become increasingly important and may help mitigate climate change. However, high water use by trees may deplete water resources for associated crops or other purposes. Choice of tree species might reduce water use rates to acceptable levels, but available information on species-specific water use characteristics is scarce. We addressed the following questions: 1) do species differ in xylem sap flux response to fluctuating environmental conditions, 2) are there species-specific differences in quantities of water used, and specifically 3) do universal rules relating tree size to water use apply? This chapter combines data on tree sap flux and water use gathered in Indonesia, Panama and the Philippines. These studies applied the same methods and were conducted in recently established stands (5-12 years old when studied) characterised by small diameter trees and relatively simple stand structure. We analyse data from more than 100 trees belonging to 17 species using a simple sap flux model. Model application suggested species-specific differences in parameters such as maximal sap flux velocity and responses to radiation and vapour pressure deficit. With respect to the quantity of water used per tree, we observed a strong correlation between tree diameter and tree water use, which confirms earlier publications. However, e.g. in the stands in the Philippines where tree diameter explained 65% of observed variation; some species clearly followed distinct trajectories. For a given diameter, up to twofold differences in tree water use among species were observed. Our findings thus support the idea that species selection can be used to control tree water use of future reforestations and within agroforestry systems. This will be especially relevant in areas where water resources are limited already or where climate scenarios predict decreasing precipitation.

In South-east Asia, ENSO-related droughts represent irregularly occurring hazards for agroforestry systems containing cocoa which are predicted to increase in severity with expected climate warming. To characterize the drought response of mature cocoa trees, we conducted the Sulawesi Throughfall Displacement Experiment in a shaded (Gliricidia sepium) cocoa agroforestry system in Central Sulawesi, Indonesia. Three large sub-canopy roofs were installed to reduce throughfall by about 80% over a 13-month period to test the hypotheses that (i) cocoa trees are sensitive to drought due to their shallow fine root system, and (ii) bean yield is more sensitive to drought than leaf or stem growth. As 83% of fine root (diameter <2 mm) and 86% of coarse root biomass (>2 mm) was located in the upper 40 cm of the soil, the cocoa trees examined had a very shallow root system. Cocoa and Gliricidia differed in their vertical rooting patterns, thereby reducing competition for water. Despite being exposed for several months to soil water contents close to the conventional wilting point, cocoa trees showed no significant decreases in leaf biomass, stem and branch wood production or fine root biomass. Possible causes are active osmotic adjustment in roots, mitigation of drought stress by shading from Gliricidia or other factors. By contrast, production of cocoa beans was significantly reduced in the roof plots, supporting reports of substantial reductions in bean yields during ENSO-related drought events in the region. We conclude that cocoa possesses traits related to drought tolerance which enable it to maintain biomass production during extended dry periods, whereas bean yield appears to be particularly drought sensitive.

In a cacao agroforest with Gliricidia shade trees we studied the effects of experimentally reduced throughfall on tree sap flux densities and soil water budget. The objectives of our study were: (1) to identify the response of cacao and Gliricidia sap flux densities to reduced soil water availability, and (2) to analyze the changes in stand level soil water partitioning, measured as changes in transpiration, drainage and soil moisture storage under reduced throughfall. The study was conducted in Sulawesi, Indonesia at a site where ambient precipitation was 2841mmyr−1. The research site was subdivided into three control plots and three plastic roof plots in which net precipitation was reduced by 71% through the use of a sub-canopy roof for 13 months, to create rainfall conditions similar to an extended El Niño event. The two species differed in their sap flux response to declining soil water content. Daily Gliricidia sap flux declined at higher soil water contents than cacao, while Gliricidia sap flux declined in response to all levels of vapor pressure deficit (VPD) and cacao only responded to high levels of VPD. Average monthly sap flux densities of cacao and Gliricidia in the roof plots decreased linearly with decreasing soil water content reaching a maximum reduction of 21% (cacao) and 29% (Gliricidia) as compared to control plots. Cacao withdrew water mainly from the topsoil where small amounts of throughfall entering gaps in the roof frequently rewetted the soil. Therefore the competition between the two species for water resources during long periods with only little precipitation was low or in favor of cacao. Average daily stand transpiration was 1.3mm in roof plots and 1.5mm in control plots. Measured soil water contents were simulated satisfactorily by the application of a one-dimensional soil water model based on Richards unsaturated flow equation. The model suggested that drainage amounted to 1554mmyr−1 in control plots. Drainage in roof plots was greatly reduced, totaling 299mmyr−1. Soil moisture storage in the control underwent minor variations while it was heavily reduced in roof plots. Drainage in roof plots was a large driver of changes in soil moisture storage especially in deeper soil layers while precipitation and root water uptake were the main influencing factors for the topsoil. Deep root water uptake seemed to play a minor role in this experiment. Despite the strong reduction in net precipitation, declines in stand transpiration were moderate, suggesting that the stand did not react sensitively to the reduced throughfall scenario.

Abstract. Climate change induced droughts pose a serious threat to ecosystems across the tropics and sub-tropics, particularly to those areas not adapted to natural dry periods. In order to study the vulnerability of cacao (Theobroma cacao) – Gliricidia sepium agroforestry plantations to droughts a large scale throughfall displacement roof was built in Central Sulawesi, Indonesia. In this 19-month replicated experiment, we measured soil surface CO2 efflux (soil respiration) in three simulated drought plots compared with three adjacent control plots. Soil respiration rates peaked at intermediate soil moisture and decreased under increasingly dry conditions (drought induced), but also decreased when soils became water saturated, as evidenced in control plots. The simulated drought plots exhibited a slight decrease in soil respiration compared to the control plots (average 13% decrease). The strength of the drought effect was spatially variable – while some measurement chamber sites reacted strongly ("responsive") to the decrease in soil water content (up to R2=0.70) (n=11), others did not react at all ("non-responsive") (n=7). The degree of soil CO2 respiration drought response was highest around cacao tree stems and decreased with distance from the stem (R2=0.22). A significant correlation was measured between "responsive" soil respiration chamber sites and sap flux density ratios of cacao (R=0.61) and Gliricidia (R=0.65). Leaf litter CO2 respiration decreased as conditions became drier. During dry periods the litter layer contributed approximately 3–4% of the total CO2 efflux and up to 40% during wet periods. A CO2 flush was recorded during the rewetting phase that lasted for approximately two weeks, during which time accumulated labile carbon stocks mineralized. The net effect on soil CO2 emissions over the duration of the experiment was neutral, control plots respired 11.1±0.5 Mg C ha−1 yr−1, while roof plots respired 10.5±0.5 Mg C ha−1 yr−1.

Water use characteristics of cacao ( Theobroma cacao ) and Gliricidia sepium shade trees were studied in an agroforest on Sulawesi, Indonesia. The objectives were: (1) to identify environmental and tree structural factors controlling water use, (2) to analyse the effect of shade tree cover on cacao water use and (3) to estimate stand level transpiration. Sap flux density was measured in up to 18 trees per species and described with a Jarvis‐type model. Model parameters suggested a 49% higher maximum sap flux density in cacao than in Gliricidia and species differences in the response to vapour pressure deficit and radiation. Tree water use was positively related to tree diameter in both species, but this relationship tended to differ between species. In cacao trees maximal tree water use increased with decreasing canopy gap fraction above the trees ( R 2 adj = 0·39, p = 0·04). This was paralleled by an increase of cacao stem diameter and leaf area with decreasing gap fraction. Maximum water use rate per unit crown area of cacao was 13% higher than that of Gliricidia . At the stand level the average transpiration rate was estimated at 1·5 mm day −1 per unit ground area, 70% of which was contributed to by cacao. We conclude that, in the given stand, species differed substantially in water use characteristics, while estimated stand transpiration is in line with findings from other studies for cacao stands. Shade trees may enhance stand transpiration through own water use and additionally by increasing water use rates of cacao trees. Copyright © 2009 John Wiley & Sons, Ltd.