Detect and attribute the extreme maize yield losses based on spatio-temporal deep learning.

Providing accurate crop yield estimations at large spatial scales and understanding yield losses under extreme climate stress is an urgent challenge for sustaining global food security. While the data-driven deep learning approach has shown great capacity in predicting yield patterns, its capacity to detect and attribute the impacts of climatic extremes on yields remains unknown. In this study, we developed a deep neural network based multi-task learning framework to estimate variations of maize yield at the county level over the US Corn Belt from 2006 to 2018, with a special focus on the extreme yield loss in 2012. We found that our deep learning model hindcasted the yield variations with good accuracy for 2006-2018 (R2 = 0.81) and well reproduced the extreme yield anomalies in 2012 (R2 = 0.79). Further attribution analysis indicated that extreme heat stress was the major cause for yield loss, contributing to 72.5% of the yield loss, followed by anomalies of vapor pressure deficit (17.6%) and precipitation (10.8%). Our deep learning model was also able to estimate the accumulated impact of climatic factors on maize yield and identify that the silking phase was the most critical stage shaping the yield response to extreme climate stress in 2012. Our results provide a new framework of spatio-temporal deep learning to assess and attribute the crop yield response to climate variations in the data rich era.

Automatic monitoring of lettuce fresh weight by multi-modal fusion based deep learning.

Fresh weight is a widely used growth indicator for quantifying crop growth. Traditional fresh weight measurement methods are time-consuming, laborious, and destructive. Non-destructive measurement of crop fresh weight is urgently needed in plant factories with high environment controllability. In this study, we proposed a multi-modal fusion based deep learning model for automatic estimation of lettuce shoot fresh weight by utilizing RGB-D images. The model combined geometric traits from empirical feature extraction and deep neural features from CNN. A lettuce leaf segmentation network based on U-Net was trained for extracting leaf boundary and geometric traits. A multi-branch regression network was performed to estimate fresh weight by fusing color, depth, and geometric features. The leaf segmentation model reported a reliable performance with a mIoU of 0.982 and an accuracy of 0.998. A total of 10 geometric traits were defined to describe the structure of the lettuce canopy from segmented images. The fresh weight estimation results showed that the proposed multi-modal fusion model significantly improved the accuracy of lettuce shoot fresh weight in different growth periods compared with baseline models. The model yielded a root mean square error (RMSE) of 25.3 g and a coefficient of determination (R 2) of 0.938 over the entire lettuce growth period. The experiment results demonstrated that the multi-modal fusion method could improve the fresh weight estimation performance by leveraging the advantages of empirical geometric traits and deep neural features simultaneously.

Double cropping and cropland expansion boost grain production in Brazil.

Brazilian grain production increased more than fourfold from 1980 to 2016. The grain boom was achieved primarily by soybean-corn double cropping and cropland expansion-both show changing spatiotemporal patterns since the 1980s. Here, we quantified the contributions of these two strategies to corn and soybean production in Brazil using municipality-level data from 1980 to 2016. We found the contribution of double cropping to the grain boom steadily increased to 35% and the largest driving force was the increasing demand for grain export. While double cropping dominated the conventional agricultural regions, cropland expansion was still the major strategy in agricultural frontiers such as the Centre-West and Matopiba. The implementation of double cropping offset the equivalent of 76.7 million ha of Brazilian arable land for grain production from 2003 to 2016. Double cropping in Brazil has the potential to help alleviate land burdens in other pantropical countries with increasing global food demand.

A deep learning approach to conflating heterogeneous geospatial data for corn yield estimation: A case study of the US Corn Belt at the county level.

Understanding large-scale crop growth and its responses to climate change are critical for yield estimation and prediction, especially under the increased frequency of extreme climate and weather events. County-level corn phenology varies spatially and interannually across the Corn Belt in the United States, where precipitation and heat stress presents a temporal pattern among growth phases (GPs) and vary interannually. In this study, we developed a long short-term memory (LSTM) model that integrates heterogeneous crop phenology, meteorology, and remote sensing data to estimate county-level corn yields. By conflating heterogeneous phenology-based remote sensing and meteorological indices, the LSTM model accounted for 76% of yield variations across the Corn Belt, improved from 39% of yield variations explained by phenology-based meteorological indices alone. The LSTM model outperformed least absolute shrinkage and selection operator (LASSO) regression and random forest (RF) approaches for end-of-the-season yield estimation, as a result of its recurrent neural network structure that can incorporate cumulative and nonlinear relationships between corn yield and environmental factors. The results showed that the period from silking to dough was most critical for crop yield estimation. The LSTM model presented a robust yield estimation under extreme weather events in 2012, which reduced the root-mean-square error to 1.47 Mg/ha from 1.93 Mg/ha for LASSO and 2.43 Mg/ha for RF. The LSTM model has the capability to learn general patterns from high-dimensional (spectral, spatial, and temporal) input features to achieve a robust county-level crop yield estimation. This deep learning approach holds great promise for better understanding the global condition of crop growth based on publicly available remote sensing and meteorological data.

The quantitative models for broiler chicken response to monochromatic, combined, and mixed light-emitting diode light: A meta-analysis.

Although many experiments have been conducted to clarify the response of broiler chickens to light-emitting diode (LED) light, those published results do not provide a solid scientific basis for quantifying the response of broiler chickens. This study used a meta-analysis to establish light spectral models of broiler chickens. The results indicated that 455 to 495 nm blue LED light produced the greatest positive response in body weight by 10.66% (BW; P < 0.001) and 515 to 560 nm green LED light increased BW by 6.27% (P < 0.001) when compared with white light. Regression showed that the wavelength (455 to 660 nm) was negatively related to BW change of birds, with a decrease of about 4.9% BW for each 100 nm increase in wavelength (P = 0.002). Further analysis suggested that a combination of the two beneficial light sources caused a synergistic effect. BW was further increased in birds transferred either from green LED light to blue LED light (17.23%; P < 0.001) or from blue LED light to green LED light (17.52%; P < 0.001). Moreover, birds raised with a mixture of green and blue LED light showed a greater BW promotion (10.66%; P < 0.001) than those raised with green LED light (6.27%). A subgroup analysis indicated that BW response to monochromatic LED light was significant regardless of the genetic strain, sex, control light sources, light intensity and regime of LED light, environmental temperature, and dietary ME and CP (P > 0.05). However, there was an interaction between the FCR response to monochromatic LED light with those covariant factors (P < 0.05). Additionally, green and yellow LED light played a role in affecting the meat color, quality, and nutrition of broiler chickens. The results indicate that the optimal ratio of green × blue of mixed LED light or shift to green-blue of combined LED light may produce the optimized production performance, whereas the optimal ratio of green/yellow of mixed or combined LED light may result in the optimized meat quality.

Artificial light and biological responses of broiler chickens: dose-response.

Light intensity is an important aspect for broiler production. However, previous results do not provide a solid scientific basis for quantifying the response of broilers to light intensity. This study performed a meta-analysis to model the response of broilers to 0.1-200 lux of light intensity. Meta-analysis was used to integrate smaller studies and increase the statistical power over that of any single study and explore new hypotheses. The results indicated that light intensity <5 lux caused welfare concern (P < 0.05) and light intensity <1 lux induced productivity loss of broiler (P < 0.05), whereas greater level of light intensity >10 lux led to increased mortality (P < 0.01) and decreased uniformity (P < 0.05). Meta-regression showed that 30-200 lux light intensity was negatively related to BW (P = 0.047) and feed intake change (P = 0.054), whereas a quadratic relationship was observed between feed conversion ratio change and 50-180 lux light intensity (R2 = 0.95). In addition, the majority of carcass characteristics (abdominal fat weight and wing weight) and metabolic indicators (K+, Ca2+, and T3) were affected by light intensity >5 lux. To conclude, this meta-analysis based on published data quantitatively identified that 5 lux of light intensity during grow-out period should be the minimum level to maintain a well productivity and welfare of broiler chickens.