Transpiration characteristics of a rubber plantation in central Cambodia

Predicting Tree Sap Flux and Stomatal Conductance from Drone-Recorded Surface Temperatures in a Mixed Agroforestry System—A Machine Learning Approach

Plant transpiration is a key element in the hydrological cycle. Widely used methods for its assessment comprise sap flux techniques for whole-plant transpiration and porometry for leaf stomatal conductance. Recently emerging approaches based on surface temperatures and a wide range of machine learning techniques offer new possibilities to quantify transpiration. The focus of this study was to predict sap flux and leaf stomatal conductance based on drone-recorded and meteorological data and compare these predictions with in-situ measured transpiration. To build the prediction models, we applied classical statistical approaches and machine learning algorithms. The field work was conducted in an oil palm agroforest in lowland Sumatra. Random forest predictions yielded the highest congruence with measured sap flux (r2 = 0.87 for trees and r2 = 0.58 for palms) and confidence intervals for intercept and slope of a Passing-Bablok regression suggest interchangeability of the methods. Differences in model performance are indicated when predicting different tree species. Predictions for stomatal conductance were less congruent for all prediction methods, likely due to spatial and temporal offsets of the measurements. Overall, the applied drone and modelling scheme predicts whole-plant transpiration with high accuracy. We conclude that there is large potential in machine learning approaches for ecological applications such as predicting transpiration.

Expanding Rubber Plantations in Southern China: Evidence for Hydrological Impacts

While there is increasing evidence concerning the detrimental effects of expanding rubber plantations on biodiversity and local water balances, their implications on regional hydrology remain uncertain. We studied a mesoscale watershed (100 km2) in the Xishuangbanna prefecture, Yunnan Province, China. The influence of land-cover change on streamflow recorded since 1992 was isolated from that of rainfall variability using cross-simulation matrices produced with the monthly lumped conceptual water balance model GR2M. Our results indicate a statistically significant reduction in wet and dry season streamflow from 1992 to 2002, followed by an insignificant increase until 2006. Analysis of satellite images from 1992, 2002, 2007, and 2010 shows a gradual increase in the areal percentage of rubber tree plantations at the watershed scale. However, there were marked heterogeneities in land conversions (between forest, farmland, grassland, and rubber tree plantations), and in their distribution across elevations and slopes, among the studied periods. Possible effects of this heterogeneity on hydrological processes, controlled mainly by infiltration and evapotranspiration, are discussed in light of the hydrological changes observed over the study period. We suggest pathways to improve the eco-hydrological functionalities of rubber tree plantations, particularly those enhancing dry-season base flow, and recommend how to monitor them.

Oil Palm and Rubber Tree Water Use Patterns: Effects of Topography and Flooding

Oil palm and rubber plantations extend over large areas and encompass heterogeneous site conditions. In periods of high rainfall, plants in valleys and at riparian sites are more prone to flooding than plants at elevated topographic positions. We asked to what extent topographic position and flooding affect oil palm and rubber tree water use patterns and thereby influence spatial and temporal heterogeneity of transpiration. In an undulating terrain in the lowlands of Jambi, Indonesia, plantations of the two species were studied in plot pairs consisting of upland and adjacent valley plots. All upland plots were non-flooded, whereas the corresponding valley plots included non-flooded, long-term flooded, and short-term flooded conditions. Within each plot pair, sap flux densities in palms or trees were monitored simultaneously with thermal dissipation probes. In plot pairs with non-flooded valleys, sap flux densities of oil palms were only slightly different between the topographic positions, whereas sap flux densities of rubber trees were higher in the valley than at the according upland site. In pairs with long-term flooded valleys, sap flux densities in valleys were lower than at upland plots for both species, but the reduction was far less pronounced in oil palms than in rubber trees (-22 and -45% in maximum sap flux density, respectively). At these long-term flooded valley plots palm and tree water use also responded less sensitively to fluctuations in micrometeorological variables than at upland plots. In short-term flooded valley plots, sap flux densities of oil palm were hardly affected by flooding, but sap flux densities of rubber trees were reduced considerably. Topographic position and flooding thus affected water use patterns in both oil palms and rubber trees, but the changes in rubber trees were much more pronounced: compared to non-flooded upland sites, the different flooding conditions at valley sites amplified the observed heterogeneity of plot mean water use by a factor of 2.4 in oil palm and by a factor of 4.2 in rubber plantations. Such strong differences between species as well as the pronounced heterogeneity of water use across space and time may be of relevance for eco-hydrological assessments of tropical plantation landscapes.

Rubber yield prediction by meteorological conditions using mixed models and multi-model inference techniques

Linear mixed models were developed and used to predict rubber (Hevea brasiliensis) yield based on meteorological conditions to which rubber trees had been exposed for periods ranging from 1 day to 2 months prior to tapping events. Predictors included a range of moving averages of meteorological covariates spanning different windows of time before the date of the tapping events. Serial autocorrelation in the latex yield measurements was accounted for using random effects and a spatial generalization of the autoregressive error covariance structure suited to data sampled at irregular time intervals. Information theoretics, specifically the Akaike information criterion (AIC), AIC corrected for small sample size (AICc), and Akaike weights, was used to select models with the greatest strength of support in the data from a set of competing candidate models. The predictive performance of the selected best model was evaluated using both leave-one-out cross-validation (LOOCV) and an independent test set. Moving averages of precipitation, minimum and maximum temperature, and maximum relative humidity with a 30-day lead period were identified as the best yield predictors. Prediction accuracy expressed in terms of the percentage of predictions within a measurement error of 5 g for cross-validation and also for the test dataset was above 99 %.

A simple framework to analyze water constraints on seasonal transpiration in rubber tree (Hevea brasiliensis) plantations

Climate change and fast extension in climatically suboptimal areas threaten the sustainability of rubber tree cultivation. A simple framework based on reduction factors of potential transpiration was tested to evaluate the water constraints on seasonal transpiration in tropical sub-humid climates, according pedoclimatic conditions. We selected a representative, mature stand in a drought-prone area. Tree transpiration, evaporative demand and soil water availability were measured every day over 15 months. The results showed that basic relationships with evaporative demand, leaf area index and soil water availability were globally supported. However, the implementation of a regulation of transpiration at high evaporative demand whatever soil water availability was necessary to avoid large overestimates of transpiration. The details of regulation were confirmed by the analysis of canopy conductance response to vapor pressure deficit. The final objective of providing hierarchy between the main regulation factors of seasonal and annual transpiration was achieved. In the tested environmental conditions, the impact of atmospheric drought appeared larger importance than soil drought contrary to expectations. Our results support the interest in simple models to provide a first diagnosis of water constraints on transpiration with limited data, and to help decision making toward more sustainable rubber plantations.