Search

Growing crop production increases agricultural water consumption (AWC), but whether these two factors are matched in northern and southern China in the past two decades remains controversial. Here we used a remote sensing‐based water‐carbon coupling model combined with satellite observations to quantify the crop production, AWC, and water deficit changes in China over 2003–2018. We show that in northern China, crop production and AWC significantly (p < 0.05) increased, mainly due to planting area expansion, and the increasing rates were 2.6 and 6.2 times greater than those in southern China. The increasing irrigation demand in northern China greatly reduced the availability of stored terrestrial water and could not be met by the agricultural water supply in the next two decades. Our results show that the conflict between the limited agricultural water supply and the increasing irrigation demand cannot be resolved by the South‐to‐North Water Diversion Project, and better management in water resources and crop production is needed to achieve the United Nations Sustainable Development Goals. Plain Language Summary: Water is essential to agricultural production. Increasing water consumption in water supply limited areas may result in a water crisis and threaten sustainable food security, especially in China, the most populated agricultural country in the world. There are continuous arguments regarding the matches between agricultural water consumption and supply in China for the past decades and for the future. Satellite‐based observation is an advanced technique to detect the real crop growth characteristics in large scales. Here we used a satellite‐driven water‐carbon coupling model to show the conflicts between the agricultural production and water resources in northern and southern China. We found that expansion in arable land areas increased irrigation water demand in northern China during 2003–2018 and has intensively reduced the available terrestrial water resources in agricultural areas. The additional water supplied by China's largest water transfer project has limited capability to mitigate the agricultural water crisis problems in northern regions in the near future, calling for improved agricultural water management measures to ensure food‐water security in a changing environment. Key Points: The crop production‐water resource conflicts between northern and southern China intensified during 2003–2018Crop water consumption and demand increased faster in north China than in south China due to crop production expansionChina's largest water transfer project has limited capacity to ease water crisis in next two decades, calling for improved water governance [ABSTRACT FROM AUTHOR]

Within-field yield variability affects crop production and management decisions. To understand and manage this variability, different techniques have been deployed to measure and monitor the crops (and soils) at various spatial scales, including manual measurements, harvester-mounted yield monitors, proximal and remote sensing and crop simulation modelling. The value of this increasing data availability to enhance process understanding and on-ground management is unclear. This study aimed to investigate the value of the increasingly available spatial data from different sources to understand important soil-plant processes amenable to improvement in both simulation modelling and for better management decisions for dryland cropping. We collected three types of measurement data (manual sampling, sensed data from satellite and drone, and yield maps) over a 10 ha field and conducted simulations using the process-based soil-plant model APSIM at different spatial scales (varied from 1 m2 up to 10 ha). We assessed the agreement between ground measurements and yield maps, analysed the potential to use remotely sensed vegetation indices to estimate yield, and the scale at which process-based modelling could be reliable. Wheat yield extracted from yield map at 1 m2 spatial resolution only explained 30% of the variation in yield measured from 1 m2 manual sampling, with better agreement when data was aggregated to 1 ha strip-scale (R2 = 0.66, NRMSE = 9.1%). Remotely sensed vegetation indices (VI) were better correlated with the yield map when aggregating images to coarse spatial resolution (> 50 m × 50 m), while high-resolution drone VI increased the correlation at finer scales. However, the relationship and the timing of the highest correlation differed between years. APSIM simulated point-based yield measured from manual samples with NRMSE of 19.4%, but it was difficult to capture spatial variation in yield due largely to uncertainties in input data. However, APSIM simulations captured the average crop growth dynamics and yield well at 1 ha strip- and 10 ha whole field scales. The results highlight the need for caution when using yield maps and remote sensing data to quantify spatial variability and inform spatially explicit management decisions at a fine resolution (e.g., 1 m2). In our case, remote sensing data and yield maps only became consistent and process-based modelling became skilful at scales larger than a 1 ha strip. Despite an increasing amount of high-resolution spatial data, the usefulness at fine resolution needs further investigation, particularly under heterogeneous field conditions. Such data need to be analysed in conjunction with the landscape, soil and climate data to understand the drivers of spatial variability and inform process understanding and modelling. This further implies potential problems in developing spatial management practices at finer scales using such data. • Confidence in techniques to measure and model yield varies with spatial scales. • Yield extracted from yield maps agreed with destructive sampling when data aggregated to 1 ha. • NDVI and yield maps only become consistent when the spatial resolution is coarse (> 50 m). • Process-based modelling becomes meaningful/skilful at scales larger than 1 ha strip. • This study implies potential problems in spatially explicit management practices at finer scales. [ABSTRACT FROM AUTHOR]

Plant available water holding capacity (PAWC) influences the efficiency of rainfall used by crops and consequently crop productivity in dryland cropping systems. However, how soil PAWC interacts with rainfall to determine crop yield has not been quantified systematically. In this study, we used an agricultural production systems model (APSIM) and wheat as an example to quantify crop growth and yield in response to a wide range of soil PAWCs. Results show that wheat yield increased with soil PAWCs following a negative exponential curve. There is a critical PAWC above which PAWC has no impact. A summary model incorporating the impact of soil PAWC, annual rainfall and the fraction of rainfall during growing season together could explain over 95% variations in long-term averaged wheat yield. The results and the summary model provide the scientific basis to systematically explain crop yield variation in response to soil PAWC change across contrasting climates. • Soil water holding capacity (PAWC) impacts crop yield differently across climates. • A modelling approach is used to quantify relationship between yield, PAWC & climate. • Wheat yield increased with PAWC until it plateaus above a critical PAWC. • The critical PAWC is closely correlated with rainfall outside the growing season. • A negative exponential model with PAWC & rainfall explained >95% yield variations. [ABSTRACT FROM AUTHOR]

Early vigour, or faster early leaf area development, has been considered an important trait for rainfed wheat in dryland regions such as Australia. However, early vigour is a genetically complex trait, and results from field experiments have been highly variable. Whether early vigour can lead to improved water use efficiency and crop yields is strongly dependent on climate and management conditions across the entire growing season. Here, we present a modelling framework for simulating the impact of early vigour on wheat growth and yield at eight sites representing the major climate types in Australia. On a typical soil with plant available water capacity (PAWC) of 147 mm, simulated yield increase with early vigour associated with larger seed size was on average 4% higher compared with normal vigour wheat. Early vigour through selection of doubled early leaf sizes could increase yield by 16%. Increase in yield was mainly from increase in biomass and grain number, and was reduced at sites with seasonal rainfall plus initial soil water <300 mm. Opportunities exists for development of early vigour wheat varieties for wetter sites. Soil PAWC could play a significant role in delivering the benefit of early vigour and would require particular attention. [ABSTRACT FROM AUTHOR]

Wheat production in Southwest China (SWC) plays a vital role in guaranteeing local grain security, but it is threatened by increasingly frequent seasonal drought in recent years. In spite of the importance, the impact of past climate change on wheat potential productivity and drought severity has not been properly addressed. In this study, we employed a relatively simple resource use efficiency model to analyze the spatiotemporal changes of the potential productivity (PP) and rainfed productivity (RP) of wheat ( Triticum aestivum L.) in Southwest China (SWC) from 1962 to 2010. A wheat drought severity index was defined as the relative difference between PP and RP, i.e., (PP-RP)/PP, to evaluate the changing frequency and severity of drought under warming SWC. Across the entire region from 1962 to 2010, the negative impact of decreasing sunshine hours (0.06 h day per decade, p < 0.05) on PP was offset by the increase in average temperature of wheat growing season (0.22 °C per decade, p < 0.01). PP increased by 283 kg ha per decade ( p < 0.01), while RP did not show significant trend due to increased water stress. The gap between PP and RP has increased by 26 kg ha per decade ( p < 0.01). Moderate and severe drought mostly occurred in central and southern SWC. The percentage of stations experienced moderate or severe drought increased by 2.0 % ( p < 0.05) per decade, and reached 52 % in recent decade. Our results, together with the uneven distribution of rainfall, indicate great potential for irrigation development to harvest water and increase wheat yield under the warming climate in SWC. [ABSTRACT FROM AUTHOR]

High-input agriculture in China has successfully increased crop productivity in the past decades, but at a significant environmental cost. It is essential to improve management strategies to mitigate greenhouse gas (GHG) emissions and other environmental costs, while maintaining grain yields. However, there is a lack of studies to evaluate mitigation strategies under long-term climate variability. This paper combines field experimental data and soil–plant systems modeling to investigate the potential for improving water and nitrogen management of a wheat–maize double cropping system in North China Plain. The APSIM model was calibrated against the data and then applied to simulate crop yield and N 2 O emissions from soil in response to irrigation and nitrogen inputs. Our results show that the N fertilizer rate and irrigation amount under the local farmer practice could be reduced by 28% and 14% without sacrificing crop yield. This in turn led to a reduction in GHG emissions by 31%, mainly attributed to the decrease in emissions from the production and transportation of N fertilizer and direct N 2 O emissions from soil. Additionally, the results indicate that the direct N 2 O emissions from soil was positively correlated with N inputs, implying an increasing emission factor (N 2 O produced per unit of N input) with N application rates. It is concluded that potential exists to optimize N fertilizer rate and irrigation amount to reduce GHG emissions while still maintaining crop yield in the agro-ecosystems in North China Plain. [ABSTRACT FROM AUTHOR]

El Niño-Southern Oscillation (ENSO) contributes to climate anomalies, especially those related to regional rainfall, which affect crop production. Although the North China Plain (NCP) is the most important agricultural production region in China, the impact of ENSO events on local climate and crop production has received only limited attention. Therefore, the impact of different phases of ENSO on local climate and production of winter wheat and summer maize, both rain fed and irrigated, was investigated at three sites using the agricultural production systems simulator model. Data on daily temperature, precipitation, and sunshine hours for 50 years (1956-2006) were analysed to build climate scenarios for three categories of ENSO: years with El Niño events, years with La Niña events, and neutral years. The pattern of climate change was generally similar across the three sites: annual precipitation decreased slightly and annual mean sunshine hours decreased significantly, whereas annual mean minimum temperature increased significantly, leading to a significant increase in mean air temperature. Precipitation decreased and temperature and sunshine hours increased in both El Niño and La Niña years but remained stable in neutral years. Under full irrigation, the probability of exceeding distribution that crop yield would be higher was not markedly affected ( P > 0.05), although the yields in both El Niño and La Niña years differed markedly from those in neutral years, especially in maize. Under rain-fed conditions, the yield of maize was decreased greatly ( P < 0.05), the probability distribution of such reduction being the highest in La Niña years at all the sites ( P < 0.05). At the provincial level, yields from well-managed fields differed ( P > 0.05) with the ENSO category: production of maize was more vulnerable than that of wheat in El Niño and La Niña years. El Niño and La Niña had similar effects on climatic variables across the NCP: low yields in El Niño and La Niña years due to lower precipitation and high yields in neutral years due to longer sunshine hours and additional irrigation. [ABSTRACT FROM AUTHOR]

Abstract: Development of successful strategies to alleviate adverse impact of climate change on crop production relies on understanding of interactions between climate and crop physiology, and presents a new opportunity for sustainable agriculture. In this study we combine the analysis of 30 years of climate data and observed rice data with crop modeling to investigate the impact of climate change and changes in rice varieties on rice growth and grain yield from 1981 to 2009 at three sites (Nanchang, Hengyang and Gaoyao) in double rice regions in China. The results revealed that while there was a warming trend in general, significant warming mainly occurred before jointing stage of early rice and after jointing of later rice. The adoption of new rice cultivars could only partly mitigate the negative impact of warming on rice growth duration and biomass growth. However, the changes of varieties increased the grain yield of both early and late rice through increased harvest index. The major variety changes involved reduced intrinsic earliness, extended grain filling period, and improved harvest index. In the face of future climate change, a planned breeding effort is needed to maintain or increase grain yield of the double rice system. [Copyright &y& Elsevier]

In the North China Plain, the grain yield of irrigated wheat-maize cropping system has been steadily increasing in the past decades under a significant warming climate. This paper combined regional and field data with modeling to analyze the changes in the climate in the last 40 years, and to investigate the influence of changes in crop varieties and management options to crop yield. In particular, we examined the impact of a planned adaptation strategy to climate change -'Double-Delay' technology, i.e., delay both the sowing time of wheat and the harvesting time of maize, on both wheat and maize yield. The results show that improved crop varieties and management options not only compensated some negative impact of reduced crop growth period on crop yield due to the increase in temperature, they have contributed significantly to crop yield increase. The increase in temperature before over-wintering stage enabled late sowing of winter wheat and late harvesting of maize, leading to overall 4-6% increase in total grain yield of the wheat-maize system. Increased use of farming machines and minimum tillage technology also shortened the time for field preparation from harvest time of summer maize to sowing time of winter wheat, which facilitated the later harvest of summer maize. [ABSTRACT FROM AUTHOR]

Abstract: China is one of the most important rice production countries in the world, and maintaining high rice productivity in China is very important for world food security. While previous studies showed that rice production in China has been and will be negatively impacted by global warming, the confounding effects of climatic change, variety improvement and agronomic managements have not been separately investigated. In this paper we combine an analysis of climate and rice growth data with crop modeling to investigate the impact of changes in climate, rice varieties, and agronomic management on rice productivity at four sites (Wuchang, Xinyang, Zhenjiang and Hanyuan) in China. The results showed a significant increase in minimum temperature during all rice growth stages at Wuchang and Zhenjiang, and from heading to maturity at Xinyang, but little change at Hanyuan. Global warming would have led to a reduction in the length of rice growing period and a reduction in grain yield at all study sites, if no varietal changes had occurred. However, the adoption of new rice varieties stabilized growing duration, increased harvest index and grain yield at three of the four sites. In the face of future warming, a planned breeding effort may be needed to offset the negative impact of future climate change. [Copyright &y& Elsevier]

Abstract: Long-term soil carbon (C) dynamics in agro-ecosystems is controlled by interactions of climate, soil and agronomic management. A modeling approach is a useful tool to understand the interactions, especially over long climatic sequences. In this paper, we examine the performance of the Agricultural Production Systems sIMulator (APSIM) to predict the long-term soil C dynamics under various agricultural practices at four semi-arid sites across the wheat-belt of eastern Australia. We further assessed the underlying factors that regulate soil C dynamics in the top 30cm of soil through scenario analysis using the validated model. The results show that APSIM is able to predict aboveground biomass production and soil C dynamics at the study sites. Scenario analyses indicate that nitrogen (N) fertilization combined with residue retention (SR) has the potential to significantly slow or reverse the loss of C from agricultural soils. Optimal N fertilization (Nopt) and 100% SR, increased soil C by 13%, 46% and 45% at Warra, Wagga Wagga and Tarelee, respectively. Continuous lucerne pasture was the most efficient strategy to accumulate soil C, resulting in increases of 49%, 57% and 50% at Warra, Wagga Wagga and Tarlee, respectively. In contrast, soil C decreases regardless of agricultural practices as a result of cultivation of natural soils at the Brigalow site. Soil C input, proportional to the amount of retained residue, is a significant predictor of soil C change. At each site, water and nitrogen availability and their interaction, explain more than 59% of the variation in soil C. Across the four sites, mean air temperature has significant (P <0.05) effects on soil C change. There was greater soil C loss at sites with higher temperature. Our simulations suggest that detailed information on agricultural practices, land use history and local environmental conditions must be explicitly specified to be able to make plausible predictions of the soil C balance in agro-ecosystems at different agro-ecological scales. [Copyright &y& Elsevier]

Agroecosystems face double pressures of producing more food to feed growing global population and reducing greenhouse gas (GHG) emissions to mitigate climate change. The Huang-Huai-Hai (HHH) plain produces ∼1/3 wheat and maize of China with very high resource inputs, particularly synthetic nitrogen (N) fertilizers since the 1980 s. Although fertilizer input has substantially increased crop yield and enhanced biomass carbon (C) input to the soil and thus stimulating soil C sequestration, GHG emissions (e.g., nitrous oxide (N 2 O)) relating to the fertilizers have been also dramatically increased. Yet, a systematic regional assessment on the trade-offs between crop yield, soil C sequestration and N 2 O emissions as impacted by management practices and environmental conditions is lacking. Here we calibrated a farming system model to conduct comprehensive assessment on crop yield and GHG emissions (soil CO 2 and N 2 O emissions) during the period 1981–2010 across the HHH plain at the resolution of 10 km. We found that soil in HHH plain was a C sink with an annual C sequestration rate of 1.53 CO 2 -eq ha −1 yr –1 (0–30 cm soil layer) during the period under current typical agricultural practices, but this sink could only offset about 68% of global warming potential from contemporary N 2 O emissions. By reducing the annual N input rate (from current more than 300 to ∼250 kg N ha −1 yr −1 ) and enhancing crop residue retention rate (from current 30% to 100%), the HHH plain could act as a net sink of GHG without sacrificing grain yield. Apart from management, the effects of three key environmental factors, i.e., mean annual rainfall and temperature and initial soil organic carbon stock on dynamics of crop yield, soil CO 2 and N 2 O emissions were also studied. The results will have important implications for the development of management strategies to maintain yield while reducing GHG emissions. [ABSTRACT FROM AUTHOR]

Crop phenology is related to the partitioning of assimilates to different organs, crop productivity and timing of crop management. Understanding the uncertainty in simulated crop phenology can help target future direction of model improvement and assess climate change impact more accurately. However, the uncertainty in maize phenology modelling across regions and under climate scenarios has not been properly addressed. This study investigated the uncertainty in simulated maize phenology using six widely used models (SIMCOY, MAIS, Beta, WOFOST, CERES, and APSIM). The models were firstly calibrated and validated using long-term observational data across China’s Maize Belt. The validated models were then used to simulate maize phenology changes in response to climate change. The results showed that the six models could reach acceptable precision (NRMSE < 8% for all the six models) by the calibration under current climate. However, the uncertainty between models in simulated maize phenology increased with the coefficient of fluctuation from 3.2% under the baseline to 6.3% under RCP4.5 and 7.4% under RCP8.5 in 2030s and 8.9% under RCP4.5 and 14.5% under RCP8.5 in 2080s for the simulated silking date, and from 4.2% under the baseline to 7.0% under RCP4.5 and 7.7% under RCP8.5 in 2030s and 10.2% under RCP4.5 and 16.7% under RCP8.5 in 2080s for the simulated maturity date in North China Plain. This highlights a significant knowledge gap in understanding how the key physiological processes of maize respond to changing temperature, particularly temperatures beyond the optimum. The uncertainty in predicted phenology is largest for summer maize in North China Plain, smaller for spring maize in northeast and southwest China. The increased uncertainty in North China Plain was due to more frequent supra-optimal temperatures, where different models disagree most in terms of phenology response to temperature, highlighting a key area for future model improvement. This implies that there could be a large uncertainty in simulated maize yield under future climate change in previous modelling studies conducted with a single crop growth model due to the uncertainty in simulated maize phenology. [ABSTRACT FROM AUTHOR]

The North China Plain (NCP) is one of the main grain production regions in China. However, the annual double cropping system of winter wheat and maize consumed a large amount of groundwater, has led to decline in groundwater table. For conservation of groundwater resource, deficit irrigation is promoted to reduce irrigation water use in recent years. This study combined process-based modeling and experimental data together to evaluate the effects of different irrigation strategies on crop production and groundwater table change for the past three decades in NCP. Data from a six year field trial (2006–2012) under four irrigation schedules and 28 year (1984–2012) field experiment under full irrigation was used to test the agricultural production systems simulator (APSIM) to simulate the responses of winter wheat and maize to different irrigation management. The results showed that APSIM model could well simulate the grain yield and water consumption of the double cropping system under the changing climate and management practices. Simulation results with four irrigation scenarios (critical stage irrigation (CI), minimum irrigation (MI), rainfed (RF) and full irrigation (FI)) from 1984 to 2012 showed that irrigation water use efficiency (IWUE) was highest under CI, although changing from FI would lead to reduction in average annual grain yield by 19.5%, 33.7%, and 58.8% under CI, MI and RF, respectively. Even the minimum irrigation strategy (MI—one irrigation for each crop) will result in continuous decline in the groundwater table, implying an inevitable future shift to alternative cropping systems like dryland wheat-maize system, single wheat or maize. Such change will likely result in lower crop productivity and increased inter-annual variability in crop yields, which will demand improved risk management strategies to minimize loss in bad year while maximize return in good years. [ABSTRACT FROM AUTHOR]

• Soil P dynamics in Vertisols are not well understood. • Crop productivity in this sub-tropical region is mostly driven by soil water availability. • APSIM Soil-P parameters attribution lacks guidance. • Labile-P simulation relies on choice of the "rate loss available" coefficient. • More mechanistic understanding required to better understand soil P dynamics. Crop production in the tropics and sub-tropics increasingly relies on phosphorus (P) fertilisers to maintain crop yields. As a result, there is a need to better understand the effectiveness of P management strategies, especially in Vertisols that represent a large proportion of the cropped area. The complexity of soil P dynamics and the interaction with environmental factors and farming practices make this challenging but the use of agricultural systems models such as the Agricultural Production Systems sIMulator (APSIM) can potentially improve our understanding of these interactions. This study (1) gives the state of knowledge of P behaviour in Vertisols and identifies which soil P processes are simulated in the APSIM Soil-P module, (2) tests the ability of APSIM as a tool to simulate soil P dynamics and crop responses in two long-term P studies in Vertisols, and (3) discusses future work needed to improve modelling of soil P dynamics. Soil P dynamics in Vertisols are not well understood, but evidence suggests that precipitation and dissolution of calcium-P minerals may play a significant role in plant P availability. These characteristics are substantially different to the strongly P-sorbing soils on which the APSIM model has previously been tested, so the ability of the model to simulate soil P dynamics and crop responses on these soils is untested. The APSIM model was able to predict the variation in crop yields due to seasonal variation of moisture availability within an acceptable modelling performance but was less able to predict P responses within seasons. This weakness was consistent with the lack of linear correlation between the plant-available P pools predicted by the P module and the measured Colwell-P that was correlated with crop responses. The non-linear relationships also differed between the two Vertisols. The poor understanding of P dynamics in Vertisols and the inability to partition soil P into measurable P pools is hindering the development of a mechanistic P module. The resulting lack of an appropriate mechanistic approach to soil P dynamics questions the predictive ability of the model. Further fundamental work is needed to predict the fate of applied P fertilisers in soils with contrasting physical and chemical characteristics, with this information integrated into an improved P modelling framework. [ABSTRACT FROM AUTHOR]

Highlights • APSIM-Oats provided a robust prediction of oats growth and yield under cool semi-arid climate conditions. • The warming trend under climate change has a negative effect on oats growth and yield. • Optimized sowing time windows enhance yield by better aligning crop requirements with temporal patterns of rainfall and temperature. • Climate risks for oats production are reduced by delaying sowing compared to current practice. Abstract Year to year variability in weather poses serious risks to crop production in the environmentally fragile agro-ecosystems of cool semi-arid areas, and future climate changes might further aggravate those risks. This study aims to quantify the contribution of altered sowing time windows to reduce climate risk for the production of oats (Avena sativa), a crop that is well adapted to short growing seasons and low rainfall. The APSIM-Oats model was calibrated and validated for phenology, above-ground dry matter and yield using data from field experiments with five sowing dates, conducted from 2009 to 2013 in Inner Mongolia, China. The model was used to determine yield trends and yield-limiting factors under rain-fed conditions using historical weather data. Changes in temperature had greater impact on crop production than changes in rainfall and the simulations indicated the importance of changed sowing windows to lengthen the growth duration and optimize water use. Delayed sowing of oats, 10 days later than current practice, ensured more secure temperature and rainfall conditions from emergence to flowering and substantially increased yields and decreased climate risk. Delayed sowing also reduced climate risk under two future climate scenarios, RCP4.5 (stabilize growth) and RCP8.5 (high greenhouse gas emission). We conclude that adaptation of sowing time of oats provides a practical strategy for enhancing yield and mitigating climate risk under climate change. [ABSTRACT FROM AUTHOR]

China produces more than 20 % of maize grain in the world, and Northeast China (NEC) accounts for ∼30 % of the nation's total maize production. Previous studies have used either climate data, satellite data, or crop growth model (CGM) to predict or forecast maize yield. However, maize is highly susceptible to the effect of extreme climate events (such as drought, heat) in NEC, and there is a lack of studies to predict/forecast maize yield by integrating climate data, satellite data, extreme climate events, and CGM-simulated data. We aim to develop a hybrid approach with machine learning to blend different sources of data (climate data, satellite data, extreme climate events) and process-based modelling results to improve predictive accuracy of maize yield in NEC. Using maize data from 44 sites during the period of 2000–2013 in NEC, we firstly optimized Agricultural Production System sIMulator (APSIM) using Differential Evolution Adaptive Metropolis combined with Gaussian likelihood function and Bayesian multiplication method. Next, we divided the growing season into five phases, and selected variables of different phases using exploratory data analysis and Random Forest. Then, we developed a hybrid model using Random Forest by blending of multiple sources of data and APSIM simulations to predict maize yield from the start to the end of the growing season, and quantified the relative contribution of predictors. A hybrid model developed with random forest by combining climate data, NDVI, extreme climate events and APSIM simulations can achieve high performance for predicting yield toward the end of the growing season. The accuracy of in-season yield prediction showed a linear increase with MAPE/KGE changing from 19 %/0.05 to 13 %/0.53 from start to end of the growing season. Yield forecasts are acceptable with RMSE/MAE of 1.20/1.01 Mg ha−1 (16 %/13 % of the observed mean yield) approximately one-month prior to harvest. The most important predictor that affect yield forecast was APSIM-simulated biomass or yield, and the most important extreme climate event was drought during early grain-filling stage. With the increasing availability of crop-related data, we expect that the in-season forecasting capacity of the proposed methodology could be further improved, and the methodology can be extended to other crops and other regions for yield forecast. [Display omitted] • A hybrid approach with biophysical modelling and machine learning developed to predict maize yield. • It integrates climate data, NDVI, extreme climate events, and APSIM-simulated biomass or yield. • It could be used for in-season yield prediction approximately one-month prior to harvest. • The most important predictor was APSIM-simulated biomass or yield. • The most important extreme climate event was drought during early grain-filling stage. [ABSTRACT FROM AUTHOR]

It is important to quantify the nitrogen (N) uptake and dinitrogen (N 2 ) fixation of legumes and estimate the N contribution that these crops make to subsequent crops for sustainable agricultural production. The growth and development of legumes and their impact on soil N fertility can be simulated by the Agricultural Production Systems Simulator (APSIM). However, the model performance has not been evaluated in simulating the dynamic processes of N accumulation and N 2 fixation. The parameterised model was tested for the simulation of N uptake and N 2 fixation in above-ground biomass of four crop legumes (lupin, chickpea, field pea and peanut). The simulations varied in location, cultivar, sowing date, climate, soil type, water regime (irrigated or dryland) and starting soil N and applied fertiliser N in tropical, subtropical, semiarid and Mediterranean environments across Australia. In general, the absolute amount of N uptake and N 2 fixation in above-ground biomass (unit: kg ha −1 ) were reasonably well simulated, with 92% of the variation in observed N accumulation in above-ground biomass and 84% in N 2 fixation being explained by APSIM. The model was also able to simulate the responses of N 2 fixation by chickpea and peanut to differences in soil mineral N status. However, the simulations of N 2 fixation efficiency (NFE, calculated as fixed N 2 per unit above-ground dry matter (DM; unit: g N kg −1 DM) were much less accurate, especially for lupin. Sensitivity analysis showed that improving the definition of the model parameter of crop N 2 fixing capacity (the potential to fix atmospheric N 2 per unit above-ground DM; unit: g N g −1 DM) would improve the simulations of NFE. We therefore propose that to successfully simulate the absolute amount of N accumulation and N 2 fixation, the above-ground biomass as the major driving factor must first be simulated well, and future work should focus on accurately determining the parameter of crop N 2 fixing capacity through optimisation of N 2 fixation data obtained from field or controlled experiments to fine-tune the simulations of the relative efficiency of N 2 fixation. [ABSTRACT FROM AUTHOR]

Modelling approaches have already been used to quantitatively assess the trade-offs between crop yield and N 2 O emissions as impacted by management practices. However, the model's performance in terms of predicting N 2 O emissions was mainly assessed against total emissions per growing season or year, which may reduce accuracy in modelling due to the uncertainties in total emissions estimated using the manual (static) chamber method. Here, a comparison between optimizations for Agricultural Production Systems sIMulator (APSIM) with total N 2 O emissions and daily N 2 O emissions was conducted. We further used the validated model to develop simple surrogate models for estimating total N 2 O emissions in different years and target potential opportunities to reduce N 2 O emissions while still maintaining the grain yield under long-term climatic conditions. Five parameters relating to denitrification and nitrification were optimized using differential evolution algorithm for global optimization based on two-year field experimental data at Huantai site in North China Plain, and comprehensive simulation experiments were further conducted under long-term climate variability in an irrigated wheat–maize rotation system. The method of Levenberg-Marquardt was implemented to fit simple surrogate models for estimating total N 2 O emissions in different years, and Analysis of Variance was used for model comparison. APSIM model optimized with daily N 2 O emissions could better simulate soil N 2 O and nitrate dynamics than that optimized with total N 2 O emissions. We obtained the posterior distributions of five key parameters to which N 2 O emissions are sensitive, and demonstrated that original model using default parameters could underestimate the rate of nitrification and denitrification and the subsequent N 2 O emissions. Total N 2 O emissions increased exponentially with nitrogen application rate and mean temperature, and IPCC (1% emission factor) could underestimate whole-year N 2 O emissions when N rate was higher than the optimized nitrogen rate for crop production. We also found that there was potential to optimize nitrogen fertiliser rate to reduce N 2 O emissions while still maintaining crop yield in the irrigated wheat–maize rotation system. This study demonstrated the necessity of optimization with daily N 2 O emissions in improving model accuracy, and the posterior distributions of five parameters relating to N 2 O emissions offered reference range for future model improvement and applications. [Display omitted] • Crop yield and soil N 2 O emissions were studied using a modelling approach. • APSIM could more accurately model soil N 2 O and nitrate dynamics when five related parameters were optimized with daily N 2 O emissions. • Posterior distributions of five parameters controlling N 2 O dynamics were obtained. • Total N 2 O emissions increased exponentially with N rate and mean temperature. • N rate could be reduced by 19% without sacrificing crop yield, which in turn led to a reduction of N 2 O emissions by 34%. [ABSTRACT FROM AUTHOR]