The Optimal Image Date Selection for Evaluating Cultivated Land Quality Based on Gaofen-1 Images

Area of Science:

• Agricultural Remote Sensing
• Geospatial Analysis
• Land Management

Background:

• Accurate cultivated land quality (CLQ) evaluation is vital for food security and sustainable agriculture.
• Optimizing satellite image acquisition dates can significantly enhance the precision of CLQ assessments.
• Existing methods often lack a systematic approach to identify the ideal timing for remote sensing-based CLQ studies.

Purpose of the Study:

• To develop and validate a method for selecting the optimal image acquisition date for improved CLQ evaluation.
• To compare the performance of different vegetation indices and regression models in assessing CLQ at various rice growth stages.
• To identify the most suitable growth stage and remote sensing-derived parameters for accurate CLQ mapping.

Main Methods:

• Retrieval of five vegetation indices (LAI, DVI, EVI, NDVI, RVI) using the PROSAIL model and Gaofen-1 (GF-1) satellite imagery.
• Application of four regression models incorporating vegetation indices at distinct rice growth stages to estimate CLQ.
• Determination of the optimal image date based on minimizing the root mean square error (RMSE) between measured and estimated CLQ.

Main Results:

• The heading to flowering stage yielded the lowest RMSEs across all vegetation index-based regression models, identifying it as the optimal time for CLQ evaluation.
• The leaf area index (LAI)-based logarithm model demonstrated the highest accuracy in estimating CLQ compared to other vegetation index models.
• A regional CLQ map generated using the optimal model and GF-1 imagery at the heading to flowering stage achieved a relative RMSE of 14.09%.

Conclusions:

• The heading to flowering stage is identified as the optimal period for acquiring satellite imagery to accurately assess cultivated land quality.
• The LAI-based logarithm model, utilizing data from the optimal growth stage, provides a robust and accurate method for regional CLQ estimation.
• This study presents a novel and applicable approach for optimizing image date selection, advancing the field of remote sensing for agricultural land assessment.