Razan Elnour
A dedicated Water Resources Engineer with a specific interest in irrigation water management and wastewater treatment. Motivated to achieve tangible results and cross-team collaboration. Proactive and excited to partner with like-minded individuals to achieve goals. I hold a BSc in Civil Engineering from the University of Khartoum, Sudan and an MSc in Water and Sustainable Development from the IHE Delft Institute for Water Education, the Netherlands. I am currently enrolled at a special tailored program at the IHE, focusing on strategic planning for large-scale irrigation schemes.
Sessions
This study reflects the opportunity to utilize QGIS for validation of crop maps in agriculture, focusing on Gezira Scheme in Sudan. Due to its vast size and access limitations to some areas, making direct ground-based crop mapping across all fields is impractical. Leveraging the capabilities of the Google Earth Engine (GEE) code editor, Sentinel-2 imagery was employed to generate a crop map. However, validation of this map necessitated comparison with an existing map from local authorities, who utilized a different remote sensing-based software for its creation. Given the shared remote sensing basis of both maps, pixel by pixel comparison emerged as the most effective method. Furthermore, the outcomes of this comparison served to enhance the precision of the GEE crop map.
Initially, the Map Swipe tool, a QGIS plugin, was employed to visually compare the two maps, pinpointing smaller areas exhibiting significant differences. This approach streamlined the process by focusing detailed analysis on these specific regions, alleviating the need for exhaustive comparison across the entire boundary, such as the expansive Gezira scheme. The Map Swipe tool facilitates seamless comparison by allowing users to swipe over two active layers, facilitating direct visual contrast. Following the identification of regions with notable disparities, they were isolated from the broader scheme boundary using the clipped raster by mask layer tool. Subsequently, utilizing the raster calculator, pixels classified similarly in both maps were extracted. Finally, each resulting map underwent polygonization, and the polygons were merged to construct the new input training dataset.
Significantly accurate crop mapping within schemes like Gezira is paramount for optimizing agricultural productivity, water resource management, and sustainable land use planning. This QGIS-enhanced validation approach contributes to informed decision-making processes in agricultural management and policy formulation within the context of schemes with no ground-obtained crop maps.