Explore the vast range of titles available.

The use of ML in agronomy has been increasing exponentially since the start of the century, including data-driven predictions of crop yields from farm-level information on soil, climate and management. However, little is known about the effect of data partitioning schemes on the actual performance of the models, in special when they are built for yield forecast. In this study, we explore the effect of the choice of predictive algorithm, amount of data, and data partitioning strategies on predictive performance, using synthetic datasets from biophysical crop models. We simulated sunflower and wheat data using OilcropSun and Ceres-Wheat from DSSAT for the period 2001-2020 in 5 areas of Spain. Simulations were performed in farms differing in soil depth and management. The data set of farm simulated yields was analyzed with different algorithms (regularized linear models, random forest, artificial neural networks) as a function of seasonal weather, management, and soil. The analysis was performed with Keras for neural networks and R packages for all other algorithms. Data partitioning for training and testing was performed with ordered data (i.e., older data for training, newest data for testing) in order to compare the different algorithms in their ability to predict yields in the future by extrapolating from past data. The Random Forest algorithm had a better performance (Root Mean Square Error 35-38%) than artificial neural networks (37-141%) and regularized linear models (64-65%) and was easier to execute. However, even the best models showed a limited advantage over the predictions of a sensible baseline (average yield of the farm in the training set) which showed RMSE of 42%. Errors in seasonal weather forecasting were not taken into account, so real-world performance is expected to be even closer to the baseline. Application of AI algorithms for yield prediction should always include a comparison with the best guess to evaluate if the additional cost of data required for the model compensates for the increase in predictive power. Random partitioning of data for training and validation should be avoided in models for yield forecasting. Crop models validated for the region and cultivars of interest may be used before actual data collection to establish the potential advantage as illustrated in this study.

The DSSAT model, integrated the calibrated Hargreaves ET model and dynamic crop coefficient, was run with the generated weather data by SDSM4.2 and CanESM2 to predict the potential yield and crop water requirement (ET C ) of winter wheat in the Huang-Huai-Hai Plain in China under RCP4.5 and RCP8.5 scenarios. The results showed that the spatial distribution of potential yield in the future under RCP4.5 and RCP8.5 were similar, characterized by an increasing trend from the northwest inland to the southeast coast. The spatial distribution of ET C decreased gradually from the Shandong Peninsula to the surrounding area, and the minimum ET C was observed in the southern part of Huang-Huai-Hai Plain. The potential yield, ET C, and effective precipitation during winter wheat growing seasons might increase in the future under RCP4.5, while irrigation water requirements (IWR) would decrease. Under RCP8.5, the effective precipitation during the wheat growing seasons decreased first and then increased. However, the potential yield, ET C , and IWR of winter wheat increased first and then decreased. This study can provide some scientific evidence to mitigate the negative effects of climate change on agricultural production and water use in the Huang-Huai-Hai Plain.

In this study assimilation of MODIS LAI (MOD15A2) into DSSAT-CERES-rice crop simulation model was used to develop advance yield estimates of rice crop during pre-harvest stage (F3) in Palakkad district of Kerala during Mundakan (September- January) season 2022-23 and 2023-24. The free parameters identified as inputs for the DSSAT-CERES-rice crop simulation model were adjusted and optimized sequentially during assimilation process until a minimum value of cost function is reached. This helped to minimize the deviation between MODIS- LAI and model generated LAI and the yield predicted by the model consequently is taken as the predicted yield. The average predicted yield during 2022-23 and 2023-24 was 5590 kgha-1 and 5124 kgha-1 respectively. The yield prediction by simulation model integrated with remote sensing products had higher accuracy than using simulation model alone during both the years with number of panchayats having the BIAS above ± 10 per cent reduced from 20 to 12 and 23 to 11 during 2022-23 and 2023-24 respectively. The findings clearly show that incorporating satellite data into crop simulation models can produce more accurate rice production forecasts than crop simulation techniques used alone.

Regional and global impact assessment tools are increasingly used to explore and evaluate the impact of climate change and extreme events on crop yield and environmental externalities. However, the large uncertainties associated with the inputs or the parameters in crop models within these tools, limits their predictive ability, exceeding the spatiotemporal variability of observed yields. The objective of this study is to explore and quantify different sources of uncertainties and assumptions made behind initial conditions (IC), soil input, meteorological forcing, management practices and model cultivar parameters by running regional simulations for the time period between 2009 and 2019. Simulations were performed for maize and soybean using the pSIMS platform across the U.S Midwest by incrementally accounting for five sources of uncertainty with a 7   k m × 7   k m resolution using the APSIM and DSSAT crop growth models. First, the relative contribution of different sources of uncertainty was estimated over time and space. Then, a series of nitrate leaching hotpots were identified and a regional maize yield productivity index was estimated by decomposing the uncertainty in the same scenario using a hierarchical Bayesian random-effect model. All factors showed a strong spatial pattern in their contribution to the total uncertainty and their contribution was found to be partially dependent on location. However, across the whole region, it was found that the uncertainty around management is larger than IC, soil and meteorological forcing while showing a strong correlation with each of these factors. Given the high spatial correlation, we hypothesize that constraining soil inputs and management uncertainty could allow for the largest reduction in predictive uncertainty for crop yield. Our results showed vast areas over northern IA, IL and IN with high potential for NO 3 leaching and southern IA, IL and east NE with lower maize productivity index compared to the regional average.

Climate change is no longer a distinct prognosis but become a reality and also proved to have its impacts on crop production. Blackgram, a C3 short duration pulse crop was considered for this study to clarify the controversial statements on adaptability of this crop for changing climate. Dynamically downscaled data (CCSM4 data using REGCM4.4 model) for 1971 to 2099 (RCP 4.5 scenario) was used for the study. The popular cultivars CO6 and VBN6 of Blackgram are employed for the study after calibration and validation of DSSAT model. The impact assessment was carried out with August 1st as sowing date. The yield of Blackgram was found to have beneficial stimulus towards the changing climate under enriched CO2. Considering cultivars, difference was noticed spatially and temporally. The average yield of VBN6 was less than CO6 during base and near century, but it got reverse with time.

Future frequent droughts threaten summer maize production in the North China Plain (NCP). A proper combination of irrigation and nitrogen (N) application can improve water and N use efficiency while maintaining summer maize yield. However, the optimal irrigation and N application strategies (OINASs) for summer maize during future drought years in the NCP require further exploration. This study applied the DSSAT‐CERES‐Maize model to investigate OINASs for summer maize for all drought years during 2021–2050 under three shared socioeconomic pathways (SSP1‐2.6, SSP2‐4.5, and SSP5‐8.5). The performance of the OINASs was subsequently evaluated against no irrigation and N application (CK) condition and a conventional irrigation and N application strategy (CINAS). The results highlight the following: (1) For all drought years under the three SSP scenarios, the base fertilizer rate should be 60 kg/hm2, after that the irrigation and N application are required during the jointing and heading periods. Under the SSP1‐2.6 scenario, the average values of irrigation and N application during each earlier period are 35.5 mm and 22 kg/hm2. Under the SSP2‐4.5 and SSP5‐8.5 scenarios, the average values are (34.5 mm, 23 kg/hm2) and (47.5 mm, 18 kg/hm2). (2) Under all SSP scenarios, the optimal irrigation amounts and N application rates are much lower than those under the CINAS. After applying OINASs for summer maize, an average of 1.16–1.22 billion kg of N and 2.98–5.19 billion m3 of freshwater will be saved per future drought year in the NCP. (3) Under all SSP scenarios, the summer maize yields under the OINASs are slightly and significantly greater than those under the CINAS and CK conditions. Moreover, both water and N use efficiencies improved under the OINASs compared with those under the CINAS, with more significant improvements in N use efficiency. The OINASs provide a practical way to ensure food security and environmental sustainability.

To reduce the risks of climate change, governments agreed in the Paris Agreement to limit global temperature rise to less than 2.0 °C above pre-industrial levels, with the ambition to keep warming to 1.5 °C. Charting appropriate mitigation responses requires information on the costs of mitigating versus associated damages for the two levels of warming. In this assessment, a critical consideration is the impact on crop yields and yield variability in regions currently challenged by food insecurity. The current study assessed impacts of 1.5 °C versus 2.0 °C on yields of maize, pearl millet and sorghum in the West African Sudan Savanna using two crop models that were calibrated with common varieties from experiments in the region with management reflecting a range of typical sowing windows. As sustainable intensification is promoted in the region for improving food security, simulations were conducted for both current fertilizer use and for an intensification case (fertility not limiting). With current fertilizer use, results indicated 2% units higher losses for maize and sorghum with 2.0 °C compared to 1.5 °C warming, with no change in millet yields for either scenario. In the intensification case, yield losses due to climate change were larger than with current fertilizer levels. However, despite the larger losses, yields were always two to three times higher with intensification, irrespective of the warming scenario. Though yield variability increased with intensification, there was no interaction with warming scenario. Risk and market analysis are needed to extend these results to understand implications for food security.

【Background】 More than 85% of sugar crops in China are sugarcane and more than 90% of its sugar is made using sugarcane. The double whammy for sugarcane production in China is the dwindling water and soil resources and the increased demand for sugar. Improving water use efficiency is hence critical to sustaining sugarcane production. 【Objective】 This paper is to investigate experimentally the impact of the level of deficit irrigation imposed at different stages on the growth and yield of sugarcane. 【Method】 The experiment was conducted at a field in Guanxi province. The crop was irrigated at seedling, tillering, elongation, and mature stage with 10 mm (T1), 20 mm (T2), 30 mm (T3) and 40 mm (T4) of water respectively, with irrigation of 50 mm taken as the control (CK). Overall, there were 17 treatments and they randomly arranged in the field. Crop growth and development in each treatment was simulated using the DSSAT model, from which we calculated the change in water productivity, irrigation water use efficiency (IWUE) and the ultimate crop yield. These results were used to optimize the deficit irrigation using the genetic algorithm. 【Result】 Reducing irrigation amount at seedling, tillering or mature stage improved both WUE and IWUE, although the yield difference between the treatments was not significant. Reducing irrigation amount at the elongation stage reduced the yield significantly, with T1—T4 reducing the yield by 12.33, 9.86, 5.4 and 3.53 t/hm2, respectively. The optimized deficit irrigation calculated by the genetic algorithm had the least impact on sugarcane growth, and it improved both yield and WUE as a result. Compared with CK, the optimized deficit irrigation at 30 mm and 40 mm mm increased yield by 2.5% and 8.7% respectively. Except 40 mm, all other deficit irrigations increased WUE. 【Conclusion】 Considering yield and WUE, the optimized deficit irrigation calculated by the genetic algorithm was 30 mm to 40 mm at each of the four growth stages.

This work aims to quantify the extent to which the cultivar parameters in the DSSAT model affect the simulation output at different water and N application levels. To be specific, the extended Fourier amplitude sensitivity test (EFAST) method is used to analyze the global sensitivity and uncertainty of the cultivar parameters on the physiological indices output from the crop growth model under six different treatments. According to the test results, P5 and P1D are the varietal parameters most sensitive to aboveground dry matter and dry matter nitrogen fertilizer utilization; G2, G1, and P1D are the varietal parameters most sensitive to yield and maximum nitrogen at maturity; P1D and P1V are the varietal parameters most sensitive to the maximum leaf area index and dry matter water utilization. Due to the dual stresses of water and nitrogen, there is a significant decrease in the sensitivity of the parameters, with a greater impact caused by water stress treatments. Moreover, uncertainty is introduced to examine the effect of differences among treatments on model simulation. It is discovered that the uncertainty of these parameters is minimized at 200 kg/hm 2 of N treatment with or without water stress. Therefore, the best simulation results are obtained at this level of nitrogen application. To sum up, this study provides crucial support for the parameterization and popularization of the DSSAT model under different water and N management treatments.

Agricultural field experiments are costly and time-consuming, and often struggling to capture spatial and temporal variability. Mechanistic crop growth models offer a solution to understand intricate crop-soil-weather system, aiding farm-level management decisions throughout the growing season. The objective of this study was to calibrate and the Crop Environment Resource Synthesis CERES-Maize (DSSAT v 4.8) model to simulate crop growth, yield, and nitrogen dynamics in a long-term conservation agriculture (CA) based maize system. The model was also used to investigate the relationship between, temperature, nitrate and ammoniacal concentration in soil, and nitrogen uptake by the crop. Additionally, the study explored the impact of contrasting tillage practices and fertilizer nitrogen management options on maize yields. Using field data from 2019 and 2020, the DSSAT-CERES-Maize model was calibrated for plant growth stages, leaf area index-LAI, biomass, and yield. Data from 2021 were used to evaluate the model's performance. The treatments consisted of four nitrogen management options, viz., N0 (without nitrogen), N150 (150 kg N/ha through urea), GS (Green seeker-based urea application) and USG (urea super granules @150kg N/ha) in two contrasting tillage systems, i.e., CA-based zero tillage-ZT and conventional tillage-CT. The model accurately simulated maize cultivar’s anthesis and physiological maturity, with observed value falling within 5% of the model’s predictions range. LAI predictions by the model aligned well with measured values (RMSE 0.57 and nRMSE 10.33%), with a 14.6% prediction error at 60 days. The simulated grain yields generally matched with measured values (with prediction error ranging from 0 to 3%), except for plots without nitrogen application, where the model overestimated yields by 9–16%. The study also demonstrated the model's ability to accurately capture soil nitrate–N levels (RMSE 12.63 kg/ha and nRMSE 12.84%). The study concludes that the DSSAT-CERES-Maize model accurately assessed the impacts of tillage and nitrogen management practices on maize crop’s growth, yield, and soil nitrogen dynamics. By providing reliable simulations during the growing season, this modelling approach can facilitate better planning and more efficient resource management. Future research should focus on expanding the model's capabilities and improving its predictions further.