A Spatial Distribution Empirical Model of Surface Soil Water Content and Soil Workability on an Unplanted Sugarcane Farm Area Using Sentinel-1A Data towards Precision Agriculture Applications

Abstract

Obtaining soil water content and soil workability data using remote sensing technology with passive sensors has some limitations due to cloud cover, cloud shadow, haze and smoke. This study proposes a method for computing soil water content and soil workability over large areas, faster and in near real-time based on Sentinel-1A (SAR) data. Sample data collected from sugarcane plantations in the Kediri and Sidoarjo districts in East Java, Indonesia, were used to develop a mathematical model of the proposed method using multi-polynomial regression. The performance indicators of the model (RMSE, MAPE and accuracy) were calculated with the results of RMSE = 0.213 and 0.250, MAPE = 16.39% and 18.79%, and accuracy = 83.6% and 81.2% for the training and testing models, respectively. The distribution of soil water content and soil workability can be computed and visualized using a spatial map. The future contribution of this work is to develop a decision support system for the selection of appropriate machinery for sugarcane field operations based on the principles of precision agriculture.

Exploring the Potential Use of Sentinel-1 and 2 Satellite Imagery for Monitoring Winter Wheat Growth under Agricultural Drought Conditions in North-Western Poland

“…However, radar data require different interpretation than optical data, which are influenced by various factors such as moisture and the physical structure of the plant [14], so many ongoing studies are exploring the possibility of interpreting the Agriculture 2023, 13, 1798 2 of 17 results obtained from S-1 using S-2 data. Therefore, S-1 may be useful for, e.g., monitoring phenological phases of plants [15], where the authors of the study, using neural networks, obtained an average accuracy of 93.5% for the separation of rice phenological phases, and the average error between the calculated and actual phenological date was 3.08 days; crop classification based on temporal signatures with a supervised approach [16], where the author achieved an overall accuracy higher than 70%, or crop classification using the Random Forest method [17], where the 48 crop groups could be classified with an overall accuracy of 93.4%; monitoring crop height [18], where the authors showed a strong relationship between maize height and SAR parameters, with the coefficient of determination for VV + VH (R 2 = 0.82), VV (R 2 = 0.81), and VH (R 2 = 0.80); selection of the optimal machinery type for sugarcane field cultivation [19], where authors developed a mathematical model and received an accuracy of 83.6% and 81.2% for the training and testing models, respectively; monitoring plant development [20][21][22], where authors showed a high sensitivity of the indicators provided by S-1 to the detection of phenological growth stages for different crops; or testing sensitivity to agricultural drought [23,24], where authors found a correlation between backscatter as well as interferometric data and crop water stress. On the other hand, S-2 is more suitable for yield prediction [25,26] since authors reported a strong correlation between the obtained yield and vegetation indices with R 2 values ranging from 0.6 to 0.9; precision nitrogen fertilization [27], where authors showed that NDVI data can be used for field-scale optimal nitrogen management models; or detailing the soil-agricultural map [28].…”