Inundation is increasing due to extreme weather conditions severely impacting the agricultural sector. Hence, inundation impact on crops and its monitoring is essential for policy and decision-makers to provide timely and precise reimbursement to farmers. This study attempts to estimate paddy crop (a major staple food cultivation of the world) yield under different inundations using multispectral imageries acquired by Unmanned Aerial Vehicle (UAV) in the subtropical region. A field experiment was carried out on paddy crop (MTU-1010) during the Kharif (monsoon) season of 2018 in the research farm of the Agricultural and Food Engineering (AgFE) Department, IIT Kharagpur, India. The experimental design consists of five different types of treatments i.e., treatment 1-20 cm standing water level was maintained for 10 days, treatment 2-20 cm (20 days), treatment 3-30 cm (10 days), treatment 4-30 cm (20 days) and treatment-5 (control)-5 cm standing water level maintained up to maturity stage with three replications. An in-house-quad copter Unmanned Aerial Vehicle (UAV) equipped with a multispectral camera was used to acquire the high-resolution imageries at different inundation periods. The acquired images were radiometrically calibrated and pre-processed using Pix4d-mapper software. Various spectral indices (such as NDVI, NGRDI, RVI, GRVI, NDRE, TNDVI etc.) were evaluated and compared with the different ground truth parameters (SPAD, green seeker). The yield of different treatments was also compared and correlated with the spectral indices. The yield was increased from treatment 1 to treatment 3 but decreased for treatment 4. The highest yield of 5.02 t/ha was observed for treatment 3, and treatment 1 was the lowest yield of 4.55 t/ha compared to the control treatment of 4.92 t/ha. The spectral reflectance of RVI and GRVI were observed to have similar response variations for different treatments with an increase in Days After Transplantation (DAT). The variations of yield and spatial maps generated using UAV-based multispectral imageries for the treatments will be helpful to government agencies for early estimation of yield due to flood inundation within the small farming fields. Ask a question or comment on this session (not intended for technical support questions). Have a question or comment? Enter it here.

Evaluation of spatially distributed crop coefficient (Kc) for estimating evapotranspiration (ETc) based on remotely sensed imagery has become an essential topic in managing the demand for agricultural water. Currently, satellite (MODIS, Landsat, etc.) imageries are not insufficient to detect variability within the small agricultural field due to its lack of desired spatial and temporal resolutions. Unmanned Aerial Vehicle (UAV) equipped with various sensors like Multispectral (MS), Thermal, and Hyperspectral cameras is becoming an emerging technology to overcome these limitations over small agricultural fields. A field experiment is carried out in the Agricultural and Food Engineering (AGFE) Department, IIT Kharagpur, to estimate Kc over the small Agri. Field using UAV-based MS cameras during Kharif (monsoon) 2019-2020 season. Lysimeters are used for estimating daily ETc for conventionally irrigated paddy crops. Reference evapotranspiration (ET0) is also calculated using the weather data of the study area. High-resolution multispectral imageries are acquired using a quad-copter UAV. The imageries are pre-processed using Pix4Dmapper software, and various vegetation indices (such as NDVI, TNDVI, NDRE, RVI, GNDVI, and LCI) are evaluated. The vegetation indices (VIs) are correlated with ground truth Kc values and spatially distributed Kc maps for the whole study area are generated based upon the excellent correlation between the VIs and ground Kc. The spatial Kc maps clearly show the variation in Kc within the plots and will be helpful for the calculation of Kc for any field without a lysimeter experiment. Generated Kc maps describe the crop water demand by visual color variations within the field. This approach may be helpful in understanding the variability in crop water requirements within the field Keywords: UAV, Crop Coefficient (Kc), Crop Evapotranspiration (ETc), Vegetation Indices, Remote Sensing.

Water stress mapping in crops and its spatial disparity study at field scale is important for precise management of irrigation. Results obtained from conventional airborne practice (balloons, airplanes, and satellites) are less acceptable for timely irrigation management due to lack in spatial and temporal resolutions. Unmanned Aerial Vehicle (UAV) equipped with multispectral (MS) and thermal cameras with higher spectral and temporal resolutions can be used as a promising tool for preparing water stress maps under different water deficit conditions. In this study, Water Deficit Index (WDI) maps are generated at different days after sowing (DAS) in wheat crops under three different water conditions (WI (well water), WS1(irrigation at 5 days’ interval), and WS2 (irrigation at 11 days’ interval)) using the concept of Vegetation Index Trapezoid (VIT) using UAV based thermal and MS imageries. The UAV is flown at 60m altitude during the Rabi season 2018-19. After pre-processing of images in Pix4dMapper, nine vegetation indices are calculated from MS images and one of the indices, Normalized Green Red Difference Index (NGRDI) is selected based on the higher correlation with ground truth data (R2 greater than 0.5) and visual interpretation according to the real field condition to construct the VIT. Vegetation index and temperature values are calculated for four points of VIT by using four boundary conditions such as bare soil with (1) dry and (2) wet conditions, and full vegetation with (3) well-watered and (4) water stress conditions. By using the ArcGIS, geo-referencing of thermal images with respect to MS images is done to get the exact overlap of both images, and resampling of thermal and MS images are also carried out to get the same pixel size. WDI values are estimated using VIT of the surface-air temperature difference and NGRDI, and WDI maps are generated from the UAV-based thermal and MS imageries for potential detection of crop water stress. The conventional Crop Water Stress Index (CWSI) which is solely based on the crop canopy temperature is outperformed by the WDI, which is integration of composite land surface temperature (LST) and degree of greenness, and could be effective enough for irrigation water management. Keywords: UAV, Multispectral and Thermal imageries, NGRDI, WDI, and Wheat crop.