Soil organic matter (SOM) is one of the important parameters in agriculture management, thus estimating its distribution on the land will be essential. Remote sensing can be utilized to map the SOM distribution in the large-scale area. The objective of this research was to determine the estimation of SOM distribution on the paddy field in Sukoharjo Regency, Indonesia using Landsat 8 OLI imagery. The sampling points were determined by purposive sampling based on an overlay of land use classification map of paddy field, NDSI (Normalized Difference Soil Index) map, and soil type map. The analysis method was used simple linear regression (SLR) and multiple linear regression (MLR) between SOM content and a digital number of Landsat 8 OLI imagery. The SLR analysis resulted that all band except band 1 and 5 of Landsat 8 OLI Imagery have the capability to estimating SOM. The MLR model based on best subset analysis resulted in the combination of bands 3, 4, 6, and 7 was the best model for estimating SOM distribution (R2=0.399). The MLR model was used to create SOM distribution map on paddy field in Sukoharjo Regency and resulted in the SOM range of the area is distributed from very low (<1%) to moderate (2.1–4.2%) with the largest area was on low level (1–2%) about 11,028 ha. The result indicates that Landsat 8 OLI Imagery could be used for mapping the SOM distribution.
Silicon (Si) is recognized as one of the beneficial nutrients to enhance rice ( Oryza sativa L.) growth, yield, and quality. However, the practical knowledge of the needs for Si fertilization in real‐world farming remains unclear. Farmer‐centered experiments focusing on soil‐available Si are essential to improve productivity and profitability. This study aimed to explore factors affecting rice yield and quality using the Bayesian approach along with the economic feasibility of Si fertilization. On‐farm experimentation was conducted over 3 years (2020–2022) across 169 fields with varying nitrogen (N) fertilization rates (48–75 kg N ha −1 ) in Gero City, Japan. The yield and protein content of rice were measured using commercial yield monitors, and soil properties such as mineralizable N and available Si were chemically analyzed. A Bayesian linear regression model was employed to explore the impact of multiple factors on rice yield and quality. The results revealed substantial field‐level yield variability, with expected deviation ranging from −194 to 742 kg ha −1 , and seasonal differences, with the highest yield observed in the third year. However, high N applications reduced rice grain quality by increasing protein content by an average of 0.34% without increasing yield. Furthermore, single‐season Si fertilization showed a consistently low probability of profit, even under favorable conditions. Therefore, it might be necessary to explore the long‐term effect of Si multi‐seasonal applications on yield and profitability. The Bayesian approach provided a probabilistic evaluation of yield variability and economic feasibility, offering valuable insights for optimizing Si fertilization strategies in rice production.
Dividing soybean (Glycine max (L.) Merr) yield into several yield components, including the seeds per area and seed weight, offers better identification of the driver of yield variation, especially that is affected by environmental factors. The objectives of this study are to understand the relationship among yield, yield components, and environmental factors using a hierarchical Bayesian model, and to determine the potential limiting factors for soybean yield production. A hierarchical Bayesian approach offers a natural mechanism of the eco-biological system through a multi-level model. Precipitation data was used to represent the environmental factors during the key stages of soybean development. Yield in seven soybean environments, defined as the combination of location and year, was surveyed from 2018 to 2023. The results indicated that soybean yield varied between environments. Seed numbers per area was the main driver of the soybean yield. Moreover, precipitation during the early reproductive stages, where the seed was being developed, also significantly affected the final yield. Seed weights also contributed to the increase in soybean yield, even though the environmental factors during the seed-filling stage were not substantial. In summary, this study provides evidence of environmental conditions as a potential limiting factor of soybean yield.