Donald Gaydon

Uncertainties in predicting rice yield by current crop models under a wide range of climatic conditions

Predicting rice (Oryza sativa) productivity under future climates is important for global food security. Ecophysiological crop models in combination with climate model outputs are commonly used in yield prediction, but uncertainties associated with crop models remain largely unquantified. We evaluated 13 rice models against multi-year experimental yield data at four sites with diverse climatic conditions in Asia and examined whether different modeling approaches on major physiological processes attribute to the uncertainties of prediction to field measured yields and to the uncertainties of sensitivity to changes in temperature and CO2 concentration [CO2 ]. We also examined whether a use of an ensemble of crop models can reduce the uncertainties. Individual models did not consistently reproduce both experimental and regional yields well, and uncertainty was larger at the warmest and coolest sites. The variation in yield projections was larger among crop models than variation resulting from 16 global climate model-based scenarios. However, the mean of predictions of all crop models reproduced experimental data, with an uncertainty of less than 10% of measured yields. Using an ensemble of eight models calibrated only for phenology or five models calibrated in detail resulted in the uncertainty equivalent to that of the measured yield in well-controlled agronomic field experiments. Sensitivity analysis indicates the necessity to improve the accuracy in predicting both biomass and harvest index in response to increasing [CO2 ] and temperature.

Use of modelling to explore the water balance of dryland farming systems in the Murray-Darling Basin, Australia

Abstract The Agricultural Production Systems Simulator (APSIM) modelling framework was used to explore components of the water balance for a range of farming systems in the Murray-Darling Basin (MDB) of Australia. Water leaking below the root zone of annual crops and pastures in this region is leading to development of dryland salinity and delivery of salt to waterways. Simulation modelling was used to identify the relative magnitude of transpiration, soil evaporation, runoff and drainage and to explore temporal variability in these terms for selected locations over the 1957–1998 climate record. Two transects were used to explore the impact of climate on water balance, with all other factors held constant, including the soil. An east–west transect at approximately latitude 33°S demonstrates the primary effect of annual average rainfall ranging from 300 to 850 mm. A north–south transect along approximately the 600 mm rainfall isohyet demonstrates a secondary effect of rainfall distribution, with the fraction of annual rainfall received in winter months rising from 40% in the north to 70% in the south. Water excess (i.e. runoff plus drainage) is strongly episodic, with 60% simulated to occur in 25% of years. Longer term cycles are also evident in the time series simulations, with strong below average periods from 1959–1968 and 1979–1988 interspersed with extended periods of above average water excess from 1969–1978 and 1989–1993. Water excess was highest for the annual wheat farming system and lowest for perennial lucerne pasture. Other systems that mix summer and winter annuals (opportunity cropping) or include wheat and lucerne pasture in different temporal combinations (phase farming and companion cropping) were intermediate in their simulated water excess. These differences in water balance of the farming systems simulated were associated with differences in grain and forage yields that will affect their economic viability. The predictions of annual water excess derived from the dynamic, daily time-step modelling using APSIM for a wheat based farming system were of similar magnitude as those predicted by the Zhang et al. (2001) static model for shallow rooted pasture catchments, whilst continuous lucerne was similar to predictions for deep rooted forest catchments. To capture the effect of rainfall distribution between winter and summer an additional term was added to the Zhang model. This modified function captured 88% of the variation in the APSIM predictions of annual average water excess from annual wheat systems for 78 locations in the MDB.

A taxonomy-based approach to shed light on the babel of mathematical models for rice simulation

For most biophysical domains, differences in model structures are seldom quantified. Here, we used a taxonomy-based approach to characterise thirteen rice models. Classification keys and binary attributes for each key were identified, and models were categorised into five clusters using a binary similarity measure and the unweighted pair-group method with arithmetic mean. Principal component analysis was performed on model outputs at four sites. Results indicated that (i) differences in structure often resulted in similar predictions and (ii) similar structures can lead to large differences in model outputs. User subjectivity during calibration may have hidden expected relationships between model structure and behaviour. This explanation, if confirmed, highlights the need for shared protocols to reduce the degrees of freedom during calibration, and to limit, in turn, the risk that user subjectivity influences model performance. A taxonomy-based approach was used to classify AgMIP rice simulation models.Different model structures often resulted in similar outputs.Similar structures often led to large differences in outputs.User subjectivity likely hides relationships between model structure and behaviour.Shared protocols are still needed to limit the risks during calibration.

A methodology for up-scaling irrigation losses

This paper presents a methodology for up-scaling field irrigation losses and quantifying relative losses at the irrigation area level for potential water savings. Two levels of analysis were considered: First, the field level where irrigation is applied. Second, the irrigation area level, where the field level losses are aggregated, or up-scaled, using average loss functions. In this up-scaling approach, detailed crop-soil-water modelling can capture the variability of physical parameters (such as soils, crops, water table depth, and management practices) at the field level which are then used to derive loss functions for aggregating losses at higher scales (irrigation area level). This allows potential field-level adaptations and water management changes made by individual farmers to be assessed for impact at the larger irrigation area level. The APSIM farming systems model was used for simulation of crops (wheat, rice, and soybean) and their interaction with the wider system processes at the field level. Given the climate, soil, and management information (sowing, fertilisation, irrigation, and residue management), the model simulates infiltration, the soil moisture profile, plant water uptake, soil evaporation, and deep drainage on a daily basis. Then, by placing the field level analysis in the context of the wider irrigation system or catchment, it is possible to correlate field level interventions (e.g. water savings measures) with water requirements at these higher levels. Application of this method in the Coleambally Irrigation Area in NSW, Australia, demonstrated that an exponential function can describe the relationship between deep drainage losses and the water table depth for different soil, crop, and water table depth combinations. The rate of loss increase (slope of the curve) with the water table depth is higher on lighter (higher intake rates) soils than on heavy soils and is more pronounced in areas under rice cultivation. We also demonstrate that this analysis technique can assist in identifying spatial distribution of losses in irrigation areas, considering water table depth as an additional factor, leading to targeted areas for water-saving measures.

Causes of variation among rice models in yield response to CO2 examined with Free-Air CO2 Enrichment and growth chamber experiments

The CO2 fertilization effect is a major source of uncertainty in crop models for future yield forecasts, but coordinated efforts to determine the mechanisms of this uncertainty have been lacking. Here, we studied causes of uncertainty among 16 crop models in predicting rice yield in response to elevated [CO2] (E-[CO2]) by comparison to free-air CO2 enrichment (FACE) and chamber experiments. The model ensemble reproduced the experimental results well. However, yield prediction in response to E-[CO2] varied significantly among the rice models. The variation was not random: models that overestimated at one experiment simulated greater yield enhancements at the others. The variation was not associated with model structure or magnitude of photosynthetic response to E-[CO2] but was significantly associated with the predictions of leaf area. This suggests that modelled secondary effects of E-[CO2] on morphological development, primarily leaf area, are the sources of model uncertainty. Rice morphological development is conservative to carbon acquisition. Uncertainty will be reduced by incorporating this conservative nature of the morphological response to E-[CO2] into the models. Nitrogen levels, particularly under limited situations, make the prediction more uncertain. Improving models to account for [CO2] × N interactions is necessary to better evaluate management practices under climate change.

Rats in rice: linking crop and pest models to explore management strategies

Context Rodents cause yield losses of 10–15% in irrigated lowland rice crops in Vietnam, with farmers spending a lot of time and money trying to control them. Despite this, there is little understanding about the optimal timing of rodent control and the level of reduction required to maximise rice crop yields. This is compounded by the ability of rice crops to compensate for damage, and farmers applying control at the wrong time. Aims We explored the optimal timing and intensity of rodent control required to increase yields of irrigated lowland rice crops in the Mekong Delta, Vietnam. Methods We developed a system analysis framework using the rice model APSIM-Oryza validated against a hand-clipped field experiment, linked with a rodent population model and field data on rodent damage rates in rice crops. A range of intensities of reduced feeding rates and timing were explored in simulated scenarios. The responses were examined over three rice crop seasons in An Giang province, Mekong Delta, Vietnam. Key results The rice crop model was benchmarked, validated and shown to adequately compensate for rodent damage. Highest yield losses occurred in the third rice crop (16% yield loss). A one-off rodent control action at the booting stage of the rice crop with 50% control effectiveness achieved a 5% yield increase. The community trap barrier system (CTBS) with 30% control effectiveness achieved a 5% yield increase. Conclusions The modelling demonstrated the importance of rodent management timing and that control should be applied before the onset of the rodent breeding season, which normally starts at maximum tillering or booting stages. Implications We conclude that modelling can improve pest management decisions by optimising timing and level of effectiveness to achieve yield increases.

Nitrogen dynamics in flooded soil systems: An overview on concepts and performance of models

Abstract Extensive modelling studies on nitrogen (N) dynamics in flooded soil systems have been published. Consequently, many N dynamics models are available for users to select from. With the current research trend, inclined towards multidisciplinary research, and with substantial progress in understanding of N dynamics in flooded soil systems, the objective of this paper is to provide an overview of the modelling concepts and performance of 14 models developed to simulate N dynamics in flooded soil systems. This overview provides breadth of knowledge on the models, and, therefore, is valuable as a first step in the selection of an appropriate model for a specific application.