Integration of Sentinel-1 and -2 imagery through advanced cloud computing improves hillside vineyard soil moisture analysis
Grape yield and quality are tightly linked to soil moisture (SM), making SM monitoring critical, especially as climate change increases reliance on irrigation in rain-fed areas. Sentinel-1 (S1) radar and Sentinel-2 (S2) optical satellites offer valuable high-resolution, frequent data streams ideal f...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-06-01
|
| Series: | Agricultural Water Management |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0378377425002550 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Grape yield and quality are tightly linked to soil moisture (SM), making SM monitoring critical, especially as climate change increases reliance on irrigation in rain-fed areas. Sentinel-1 (S1) radar and Sentinel-2 (S2) optical satellites offer valuable high-resolution, frequent data streams ideal for measuring surface SM dynamics. However, despite the recognized potential of satellite remote sensing, Change Detection (CD) techniques have not been widely applied for SM monitoring within more challenging hillside grapevine vineyard environments. To address this gap, this study developed and validated two CD methods for retrieving SM at 20 m resolution using Google Earth Engine. The first method (CDS2) used S2 optical bands (Red, NIR, SWIR), while the second (CDS1S2) combined S1’s C-band radar (VV polarization) with S2 optical data. The methods were tested in two hillside vineyards (northern/southern Italy) and validated using independent reference data from flat bushlands (TxSON, Texas). Comparisons with in-situ measurements showed both methods effectively captured SM dynamics. However, combining S1 and S2 data (CDS1S2) provided significantly more accurate estimates than using S2 alone (CDS2), achieving higher R² (0.23–0.41 vs. 0.16–0.21) and lower RMSE (3.6–5.7 % vs. 3.8–7.1 % [m³/m³]). This improvement is attributed to S1's ability to penetrate vegetation and operate under various atmospheric conditions. This research demonstrates a scalable, user-friendly, and reproducible geospatial approach for precision viticulture. It highlights the potential of integrating advanced remote sensing technologies to enhance vineyard water management and sustain agricultural productivity amidst environmental changes. |
|---|---|
| ISSN: | 1873-2283 |