Abstract
This study conducted the soil moisture estimation using remote sensing data and ground soil moisture sensors in the mango plantation. The applied methodology is the spatial and statistical analysis to determine the relationship between the measured soil moisture using ground sensors and the remote sensing indices generated from Sentinel-2A satellite images. The sensors measured the soil moisture ground data from Nov. 2019 to Feb. 2020. However, we used only the data on seven dates. This is because the cloud-free satellite images are available only on these dates to generate the remote sensing indices. The used indices are NDVI, Normalized Water Moisture Index (NDWI) for vegetation water content monitoring, Normalized Soil Moisture Index (NSMI) for data visualization and analysis. In the implementation, we first visualized the soil moisture trend compared with the remote sensing indices value at the image pixels of sensor location on each observation day. Next, we statistically analyzed the spatial data to establish the relationship between the soil moisture from all the ground sensors and the remote sensing indices. However, the output R2 is very low; then, it brings us to have an idea to apply in-depth analysis based on the ground sensor performance. This method shows an interesting result. We found that only the NDWI for monitoring vegetation water content has a similar trend with the soil moisture. Secondly, we performed the linear regression correlation between soil moisture and remote sensing indices values of each sensor as time-series analysis. The result show that the correlation between soil moisture and NDWI, NSMI and NDVI are classified into 3 groups, which are 0.7 < R2 < 0.9, 0.6 < R2 < 0.7, and R2 < 0.5, where their corresponding p-value ranges are 0.001 < p-value < 0.02, 0.01 < p-value < 0.03, and 0.08 < p-value < 0.9, respectively. Lastly, we investigated the reason that causes the very high correlation between the soil moisture value of the first group of sensors and NDWI and NSMI. The result shows that these sensors are in a sparse vegetation cover area, where NDVI ≤ 0.3. Therefore, according to this, we can conclude that remote sensing indices NDWI and NSMI can be applied for soil moisture estimation in a sparsely vegetated study area, where the NDVI value should be less than or equal to 0.3.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
