DefinePK

Research Objective

To record soil moisture and temperature values in cotton production areas using data loggers, determine the relationships between these values and SAR/optical satellite images, and investigate the usability of satellite images for estimating soil moisture and temperature.

Methodology

The study was conducted in 8 villages and 27 cotton parcels in Mardin Province. Data logger devices were installed to record soil temperature and humidity every 6 hours. These data were compared with pre-processed Landsat-8 and Sentinel-1 images in Google Earth Engine (GEE). Correlation and regression analyses were performed to investigate the relationships.

Methodology Flowchart

                        graph TD
    A[Install Data Loggers in Cotton Parcels] --> B[Record Soil Temperature & Humidity];
    B --> C[Collect Landsat-8 & Sentinel-1 Images];
    C --> D[Pre-process Images in GEE];
    D --> E[Apply Speckle/Radiometric/Atmospheric Corrections];
    E --> F[Analyze Data Logger & Satellite Data];
    F --> G[Perform Correlation & Regression Analysis];
    G --> H[Determine Relationships & Predictability];
    H --> I[Formulate Conclusions & Recommendations];                    

Discussion

The study demonstrates the potential of Sentinel-1 VV and VV-VH bands, along with the Landsat-8 SMI index, for estimating soil moisture and temperature in cotton production areas. The lower correlation with Landsat-8 LST band was attributed to soil cover and lower resolution, suggesting its use for larger-scale estimations.

Key Findings

- A significant and high correlation was found between soil moisture (TN) and Sentinel-1 VV (R² = 0.67), VV-VH (R² = 0.65), and Landsat-8 SMI (R² = 0.85) values.
- A significant and high correlation was found between soil temperature (TS) and Sentinel-1 VV (R² = 0.57), VV-VH (R² = 0.54), and Landsat-8 SMI (R² = 0.75) values.
- Soil moisture was primarily affected by irrigation, with no significant impact from precipitation or air humidity.
- Soil temperature was affected by air temperature at the beginning of the season and irrigation.

Conclusion

Sentinel-1 VV and VV-VH bands, and the Landsat-8 SMI index are recommended for soil moisture and temperature estimation studies. The Landsat-8 LST band is recommended for larger-scale land areas.

Fact Check

- The study was conducted in 8 villages and 27 cotton parcels in Mardin Province. (Confirmed by text)
- Sentinel-1 was launched in 2014 (Sentinel-1 A) and 2016 (Sentinel-1 B). (Confirmed by text)
- Landsat-8 was launched in 2013. (Confirmed by text)

Mind Map