Actual Evapotranspiration from UAV Images: A Multi-Sensor Data Fusion Approach

Multispectral imaging using Unmanned Aerial Vehicles (UAVs) has changed the pace of precision agriculture. Actual evapotranspiration (ETa) from very high spatial resolution UAV images over agricultural fields can help farmers increase their production at lowest possible cost. ETa estimation UAVs requires a full package sensors capturing visible/infrared and thermal portions spectrum. Therefore, this study focused on multi-sensor data fusion approach for (MSDF-ET) independent sensors. The method was based sharpening Landsat 8 pixels to by considering relationship between reference fraction (ETrf) Vegetation Index (VI). Four were processed calculate three almond fields. Two flights coincided with overpasses one in two consecutive images. ETrf chosen instead interpolate 8-derived obtain an image flight. defined as ratio grass (ETr), VIs tested included Normalized Difference (NDVI), Soil Adjusted (SAVI), Enhanced (EVI), Water (NDWI), Land Surface (LSWI). NDVI performed better under conditions. MSDF-ET-derived showed strong correlations against measured ETa, UAV- 8-based METRIC ETa. Also, visual comparison MSDF-ET maps indicative promising performance method. In sum, resulting had higher compared thermal-based without need Albedo hot/cold selection procedure. However, wet soils poorly detected, cases continuous cloudy long interval may cause biases Generally, reduces information ground, calculations be conducted moderate-performance computer system because main processing is applied coarser resolutions.