Impact of water stress under ambient and elevated carbon dioxide across three temperature regimes on soybean canopy gas exchange and productivity.

The present study investigated the interactive effects of three environmental stress factors elevated CO2, temperature, and drought stress on soybean growth and yield. Experiments were conducted in the sunlit, controlled environment Soil-Plant-Atmosphere-Research chambers under two-level of irrigation (WW-well water and WS-water stress-35%WW) and CO2 (aCO2-ambient 400 µmol mol-1 and eCO2-elevated 800 µmol mol-1) and each at the three day/night temperature regimes of 24/18 °C (MLT-moderately low), 28/22 °C (OT-optimum), and 32/26 °C (MHT-moderately high). Results showed the greatest negative impact of WS on plant traits such as canopy photosynthesis (PCnet), total dry weight (TDwt), and seed yield. The decreases in these traits under WS ranged between 40 and 70% averaged across temperature regimes with a greater detrimental impact in plants grown under aCO2 than eCO2. The MHT had an increased PCnet, TDwt, and seed yield primarily under eCO2, with a greater increase under WW than WS conditions. The eCO2 stimulated PCnet, TDwt, and seed yield more under WS than WW. For instance, on average across T regimes, eCO2 stimulated around 25% and 90% dry mass under WW and WS, respectively, relative to aCO2. Overall, eCO2 appears to benefit soybean productivity, at least partially, under WS and the moderately warmer temperature of this study.

Phosphorus Nutrition Affects Temperature Response of Soybean Growth and Canopy Photosynthesis.

In nature, crops such as soybean are concurrently exposed to temperature (T) stress and phosphorus (P) deficiency. However, there is a lack of reports regarding soybean response to T × P interaction. To fill in this knowledge-gap, soybean was grown at four daily mean T of 22, 26, 30, and 34°C (moderately low, optimum, moderately high, and high temperature, respectively) each under sufficient (0.5 mM) and deficient (0.08 mM) P nutrition for the entire season. Phosphorus deficiency exacerbated the low temperature stress, with further restrictions on growth and net photosynthesis. For P deficient soybean at above optimum temperature (OT) regimes, growth, and photosynthesis was maintained at levels close to those of P sufficient plants, despite a lower tissue P concentration. P deficiency consistently decreased plant tissue P concentration ≈55% across temperatures while increasing intrinsic P utilization efficiency of canopy photosynthesis up to 147%, indicating a better utilization of tissue P. Warmer than OTs delayed the time to anthesis by 8-14 days and pod development similarly across P levels. However, biomass partitioning to pods was greater under P deficiency. There were significant T × P interactions for traits such as plant growth rates, total leaf area, biomass partitioning, and dry matter production, which resulted a distinct T response of soybean growth between sufficient and deficient P nutrition. Under sufficient P level, both lower and higher than optimum T tended to decrease total dry matter production and canopy photosynthesis. However, under P-deficient condition, this decrease was primarily observed at the low T. Thus, warmer than optimum T of this study appeared to compensate for decreases in soybean canopy photosynthesis and dry matter accumulation resulting from P deficiency. However, warmer than OT appeared to adversely affect reproductive structures, such as pod development, across P fertilization. This occurred despite adaptations, especially the increased P utilization efficiency and biomass partitioning to pods, shown by soybean under P deficiency.

Random Forests for Global and Regional Crop Yield Predictions.

Accurate predictions of crop yield are critical for developing effective agricultural and food policies at the regional and global scales. We evaluated a machine-learning method, Random Forests (RF), for its ability to predict crop yield responses to climate and biophysical variables at global and regional scales in wheat, maize, and potato in comparison with multiple linear regressions (MLR) serving as a benchmark. We used crop yield data from various sources and regions for model training and testing: 1) gridded global wheat grain yield, 2) maize grain yield from US counties over thirty years, and 3) potato tuber and maize silage yield from the northeastern seaboard region. RF was found highly capable of predicting crop yields and outperformed MLR benchmarks in all performance statistics that were compared. For example, the root mean square errors (RMSE) ranged between 6 and 14% of the average observed yield with RF models in all test cases whereas these values ranged from 14% to 49% for MLR models. Our results show that RF is an effective and versatile machine-learning method for crop yield predictions at regional and global scales for its high accuracy and precision, ease of use, and utility in data analysis. RF may result in a loss of accuracy when predicting the extreme ends or responses beyond the boundaries of the training data.