Multi-source information fusion-driven corn yield prediction using the Random Forest from the perspective of Agricultural and Forestry Economic Management.

The objective of this study is to promptly and accurately allocate resources, scientifically guide grain distribution, and enhance the precision of crop yield prediction (CYP), particularly for corn, along with ensuring application stability. The digital camera is selected to capture the digital image of a 60 m × 10 m experimental cornfield. Subsequently, the obtained data on corn yield and statistical growth serve as inputs for the multi-source information fusion (MSIF). The study proposes an MSIF-based CYP Random Forest model by amalgamating the fluctuating corn yield dataset. In relation to the spatial variability of the experimental cornfield, the fitting degree and prediction ability of the proposed MSIF-based CYP Random Forest are analyzed, with statistics collected from 1-hectare, 10-hectare, 20-hectare, 30-hectare, and 50-hectare experimental cornfields. Results indicate that the proposed MSIF-based CYP Random Forest model outperforms control models such as support vector machine (SVM) and Long Short-Term Memory (LSTM), achieving the highest prediction accuracy of 89.30%, surpassing SVM and LSTM by approximately 13.44%. Meanwhile, as the experimental field size increases, the proposed model demonstrates higher prediction accuracy, reaching a maximum of 98.71%. This study is anticipated to offer early warnings of potential factors affecting crop yields and to further advocate for the adoption of MSIF-based CYP. These findings hold significant research implications for personnel involved in Agricultural and Forestry Economic Management within the context of developing agricultural economy.

Analysis of Corn Yield Prediction Potential at Various Growth Phases Using a Process-Based Model and Deep Learning.

Early and accurate prediction of grain yield is of great significance for ensuring food security and formulating food policy. The exploration of key growth phases and features is beneficial to improving the efficiency and accuracy of yield prediction. In this study, a hybrid approach using the WOFOST model and deep learning was developed to forecast corn yield, which analysed yield prediction potential at different growth phases and features. The World Food Studies (WOFOST) model was used to build a comprehensive simulated dataset by inputting meteorological, soil, crop and management data. Different feature combinations at various growth phases were designed to forecast yield using machine learning and deep learning methods. The results show that the key features of corn's vegetative growth stage and reproductive growth stage were growth state features and water-related features, respectively. With the continuous advancement of the crop growth stage, the ability to predict yield continued to improve. Especially after entering the reproductive growth stage, corn kernels begin to form, and the yield prediction performance is significantly improved. The performance of the optimal yield prediction model in flowering (R2 = 0.53, RMSE = 554.84 kg/ha, MRE = 8.27%), in milk maturity (R2 = 0.89, RMSE = 268.76 kg/ha, MRE = 4.01%), and in maturity (R2 = 0.98, RMSE = 102.65 kg/ha, MRE = 1.53%) were given. Thus, our method improves the accuracy of yield prediction, and provides reliable analysis results for predicting yield at various growth phases, which is helpful for farmers and governments in agricultural decision making. This can also be applied to yield prediction for other crops, which is of great value to guide agricultural production.

[Effects of Nitrogen Fertilizer Application Times and Nitrification Inhibitor on N2O Emission from Potted Maize].

Rational application of nitrogen is an important strategy for increasing yield while reducing environmental pollution due to nitrogen. Pot experiments were conducted to study the effects of different application times on maize yield and soil N2O emission under conditions of equal nitrogen content, and to explore the relationship between the abundance of nitrogen conversion functional genes and N2O emission. Four treatments were used, namely a control (CK, no urea), one-time application (S1, one application of 0.5 g·kg-1 urea+nitrification inhibitor), two separate applications ï¼»S2, two applications of 0.5 g·kg-1 urea (40% and 60% respectively)ï¼½ and three separate applications (S3, 0.5 g·kg-1 urea was divided into three different applications: 20%, 40% and 40% respectively). The results showed that: ① nitrogen application promoted soil acidification, and the degree of soil acidification varied significantly with different application times. More applications of nitrogen led to stronger soil acidification. Nitrogen application significantly increased the ear yield and stem biomass of fresh table maize, but different nitrogen application times may alter soil pH, leading to differences in the degree of nitrogen uptake and utilization in plants. While the S3 treatment significantly reduced soil pH, it also reduced the cumulative nitrogen uptake and utilization in the plants, resulting in a high cumulative N2O emission. Compared with the S3 treatment, the yield was 40.21% and 42.55% higher in the S1 and S2 treatments, and the cumulative N2O emission decreased by 79.4% and 20.9%, respectively. ② N2O emission was positively correlated with the abundance of AOB and nirK genes, which were the main contributors to N2O emission. S1 significantly decreased the abundance of AOB and nirK genes and N2O emissions, while S2 and S3 significantly increased the abundance of nirK and nirS genes and decreased the abundance of nosZ genes after fertilization, promoting N2O emissions. Nitrogen application times affect the functional genes of the nitrogen transformation process, and thus affect N2O emissions. In conclusion, a one-time application of urea combined with DCD only guarantees high maize yield and improves the efficient use of nitrogen, but also reduces greenhouse gas emissions. Thus, it is the recommended nitrogen fertilization mode for the cultivation of fresh corn in Hainan.

Machine Learning-Based Modeling of Spatio-Temporally Varying Responses of Rainfed Corn Yield to Climate, Soil, and Management in the U.S. Corn Belt.

Better understanding the variabilities in crop yield and production is critical to assessing the vulnerability and resilience of food production systems. Both environmental (climatic and edaphic) conditions and management factors affect the variabilities of crop yield. In this study, we conducted a comprehensive data-driven analysis in the U.S. Corn Belt to understand and model how rainfed corn yield is affected by climate variability and extremes, soil properties (soil available water capacity, soil organic matter), and management practices (planting date and fertilizer applications). Exploratory data analyses revealed that corn yield responds non-linearly to temperature, while the negative vapor pressure deficit (VPD) effect on corn yield is monotonic and more prominent. Higher mean yield and inter-annual yield variability are found associated with high soil available water capacity, while lower inter-annual yield variability is associated with high soil organic matter (SOM). We also identified region-dependent relationships between planting date and yield and a strong correlation between planting date and the April weather condition (temperature and rainfall). Next, we built machine learning models using the random forest and LASSO algorithms, respectively, to predict corn yield with all climatic, soil properties, and management factors. The random forest model achieved a high prediction accuracy for annual yield at county level as early as in July (R 2 = 0.781) and outperformed LASSO. The gained insights from this study lead to improved understanding of how corn yield responds to climate variability and projected change in the U.S. Corn Belt and globally.

Agronomic performance of polyethylene and biodegradable plastic film mulches in a maize cropping system in a humid continental climate.

Plastic polyethylene mulch has been widely used in crop production, but also causes environmental pollution if plastic residues accumulate in soil. Biodegradable plastic mulches (BDM) are a potential solution to problems caused by polyethylene mulches, as BDMs are designed be tilled into the soil after the growing season and then biodegrade. However, the agronomic performance of BDMs still needs to be tested for comparison to polyethylene mulch. We carried out a two-year field experiment in 2018 and 2019 in a typical humid continental climate in Northeast China. Maize was planted in a ridge-furrow pattern, with mulching treatments consisting of no mulch (control), clear BDM, black BDM, clear polyethylene, and black polyethylene. Clear mulches increased soil temperature when compared to no mulch control treatments, while black mulches decreased or did not change soil temperature during the early growing season. Soil temperature and root morphology were similar between BDM and polyethylene mulches for a given type of plastic color. Maize yield did not differ across all the treatments. Maize protein, fat, N and P contents were generally higher for black BDM than other treatments, suggesting that maize quality benefited especially from black BDM. Overall, these results show that, in a humid continental climate, the agronomic performance of clear and black BDMs was equivalent to, or better than, that of polyethylene plastic mulches for maize production.

Effect of biodegradable film mulching on crop yield, soil microbial and enzymatic activities, and optimal levels of irrigation and nitrogen fertilizer for the Zea mays crops in arid region.

Biodegradable film mulching (BM) is considered as the best alternative to plastic film mulching (PM) since it can prevent pollution caused due to plastic residues. However, the differences in soil microbial biomass and enzymatic activities between BM and PM, especially for different soil water and nitrogen contents remain ambiguous. In this study, the effects of BM, PM, and no film mulching (NM) on soil microbial biomass C (Cmic), N (Nmic), soil enzymes, and soil C/N ratio in a cornfield were evaluated using experimental data from 2018 and 2019. Additionally, different irrigation depths (30 mm, 22.5 mm, and 15 mm) and N-fertilizer application levels (280 kg ha-1 and 210 kg ha-1) were used in BM. The experimental results demonstrated no apparent differences between the Cmic, Nmic, and soil enzymes between BM and PM in the early stage (elongation stage), but these values under BM were significantly lower than that of PM in the middle stage of crop growth (tasseling and filling stages). Soil sucrase, catalase, and urease under PM were increased by 20.2%, 0.6%, and 12.0%, respectively, compared to BM. The analysis of Cmic, Nmic, soil enzymes, and crop yield under different irrigation and N-fertilizer application levels demonstrated the preponderance of BM22.5, 280, showing the highest yield of 14,110.1 kg ha-1 and NUE of 61.7.

Understanding the impact of sub-seasonal meteorological variability on corn yield in the U.S. Corn Belt.

Rain-fed corn system has varied optimal environmental requirements by growth phases and regions. Understanding spatiotemporal characteristics of such requirements are important to ensure food security. To capture the stage-variant growing requirements, we develop and compare statistical models with various spatial and temporal resolutions to quantify the relationships between corn yield and meteorological factors. Multilinear regression models are trained using cross-sectional datasets pooled at three magnitudes (state, district, county) with temperature and precipitation related predictors according to three temporal resolutions (growing season, fixed month, growing phase). The models are applied to the U.S. Corn Belt for the time period of 1981-2016. The results show that average corn yield variation explained by meteorological factors can be improved to 50.2% at the agricultural district scale with growth phase resolution from ~30% at the state-level with growing season resolution. The results reveal that corn yield is most sensitive to extreme heat stress during the grain filling phase. From a spatial perspective, the northern counties in the U.S. Corn Belt are less limited by precipitation resources but are more vulnerable to extreme heat. The spatiotemporal explicit statistic modeling approach quantifies the impact and adaptation potential of changing the planting date for production. Appropriate adaptions by changing plant dates can increase the potential of corn production by 0.87 million Mg year-1 in the Corn Belt.

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.

Effects of Ridge-Furrow System Combined with Different Degradable Mulching Materials on Soil Water Conservation and Crop Production in Semi-Humid Areas of China.

In China, the ridge-furrow water conservation planting (RC) system is advantageous for improving crop yields and rainwater use efficiency. In RC planting system, plastic film-mulched ridges are employed for water harvesting while the furrows serve as infiltration and planting belts. To optimize the RC system and to overcome problems due to the lack of water in semi-humid areas at risk of drought, we mulched the furrows with 8% biodegradable film (RCSB), liquid film (RCSL), or no mulching in the furrows (RCSN), while conventional flat planting (CF) was employed as the control. After 4 year (2007-2010) consecutive field study, the results showed that the soil water storage level in the 0-100 cm layer with four treatments was ranked as follow: RCSB > RCSL > RCSN > CF, while the RCSB and RCSL were 26.3 and 12.2 mm greater than RCSN, respectively. Compared with CF, the average soil temperature was significantly (P < 0.05) higher by 3.1, 1.7, and 1.5°C under the RC planting treatments (RCSB, RCSL, and RCSN) during each year, respectively. The average ET rate of RC treatments were all lower than CF in each experimental year, and the average decreased by 8.0% (P < 0.05). The average yields with RCSB, RCSL, and RCSN increased by 2,665, 1,444, and 1,235 kg ha-1, respectively, and the water use efficiency (WUE) increased by 51.6, 25.6, and 21.1%, compared with CF. RCSB obtained the highest economic benefit, the average net income was higher than CF by 4,020 Yuan ha-1. In conclusion, we found that RC planting with biodegradable film mulching in the furrows is the best cultivation pattern in the semi-humid areas of China in terms of both environmental and economic benefits.

A Weekly Indicator of Surface Moisture Status from Satellite Data for Operational Monitoring of Crop Conditions.

The triangle method has been applied to derive a weekly indicator of evaporative fraction on vegetated areas in a temperate region in Northern Italy. Daily MODIS Aqua Land Surface Temperature (MYD11A1) data has been combined with air temperature maps and 8-day composite MODIS NDVI (MOD13Q1/MYD13Q1) data to estimate the Evaporative Fraction (EF) at 1 km resolution, on a daily basis. Measurements at two eddy covariance towers located within the study area have been exploited to assess the reliability of satellite based EF estimations as well as the robustness of input data. Weekly syntheses of the daily EF indicator (EFw) were then derived at regional scale for the years 2010, 2011 and 2012 as a proxy of overall surface moisture condition. EFw showed a temporal behavior consistent with growing cycles and agro-practices of the main crops cultivated in the study area (rice, forages and corn). Comparison with official regional corn yield data showed that variations in EFw cumulated over summer are related with crop production shortages induced by water scarcity. These results suggest that weekly-averaged EF estimated from MODIS data is sensible to water stress conditions and can be used as an indicator of crops' moisture conditions at agronomical district level. Advantages and disadvantages of the proposed approach to provide information useful to issue operational near real time bulletins on crop conditions at regional scale are discussed.

Semeadura direta de milho com dois tipos de sulcadores de adubo, em nível e no sentido do declive do terreno / No tillage of corn with two types of fertilizer furrow across and towards the slope of land surface

O presente estudo teve como objetivo avaliar parâmetros operacionais de um conjunto mecanizado envolvendo trator e semeadora, assim como o rendimento da cultura do milho semeada nas diferentes configurações das máquinas e combinações com o ambiente de produção. Os tratamentos consistiram de tipos de sulcadores (discos duplos e hastes), os quais foram avaliados em experimentos em que a operação de semeadura direta do milho foi efetuada transversalmente ao declive (em nível) e em aclive e declive. A semeadura contra o declive e o uso de haste sulcadora implicaram maior demanda de esforço de tração, patinagem do trator e consumo de combustível por área trabalhada e não influenciaram o volume de solo mobilizado, a população de plantas e a produtividade de grãos do milho, em relação à operação realizada em declive e uso de sulcador de discos duplos, respectivamente. A utilização de sulcador do tipo haste resultou em menor número de plantas acamadas e quebradas, em relação ao uso de discos duplos, independentemente do sentido da operação. A principal diferença entre semear em nível ou em declive é a formação de sulcos orientados no sentido do terreno, pela ação de sulcadores do tipo haste e elevada patinagem dos rodados do trator, já que o consumo de combustível por área trabalhada e capacidade operacional não foram afetados por aquelas variáveis.

This study aimed to evaluate the operational parameters of a series of mechanized tractor and seed drill, and corn yield planted in different settings and combinations of machines with the production environment. Treatments consisted of types of fertilizer furrow (double discs and shanks), which were tested in experiments in which the operation of corn direct seeding was performed across the slope (in level) and direction of slope (for and against). Sowing against the slope and the use of shank implied a higher demand of traction effort, skating the tractor and fuel consumption per area worked, and didn't affect the volume of soil mobilized, population of plants and grain yield of corn in relation to the operation for the use of slope and furrow discs doubles, respectively. The use of furrow shank type t resulted in fewer plants broken and lodged plants in relation to the use of double discs, regardless of the direction of operation. The main practical difference between level sow or direction of the slope is the formation of grooves oriented in the same direction, by the action of furrow shank and the high shaft of the tractor wheel slip since the fuel consumption per area worked and operation capacity were not affected by those variables.