Acharya, UmeshDaigh, Aaron L.M.Oduor, Peter G.2023-04-212023-04-212021-03-16https://hdl.handle.net/10388/14610Soil moisture is an important variable in hydrology and climate studies and has been a vital factor for various processes such as water infiltration, runoff, evaporation, dryness. The use of remote sensing technology has achieved a varying degree of success in mapping soil properties for frequent temporal and greater area coverage. Soil moisture estimation using satellite image needs information on dynamic nature of actual field circumstances and micrometeorological variability in real time. The objective of this study is to (a) predict field soil moisture with Optical Trapezoidal Model (OPTRAM) using Landsat 8 images (b) use cumulative rainfall (CR), standardized precipitation index (SPI), clay content and OPTRAM in Random Forest Model to estimate field soil moisture. The use of vegetative indices to estimate soil moisture was not effective because they are affected by the growth stages and crop type. We used google earth engine to process Landsat 8 image and predict soil moisture using OPTRAM model. ArcGIS was used to make moisture maps using pixel by pixel method and R software for modeling random forest regression. The soil moisture estimated using OPTRAM model showed low correlation with field soil moisture. Soil factors, rainfall patterns might have affected the correlation. Random Forest Model was used to predict soil moisture using OPTRAM soil moisture, clay percent, four-day CR, SPI as predictor variables. This model provides promising result of r2=0.67 and RMSE= 0.053. This study proposed model that includes soil properties, meteorological information, satellite image to predict soil moisture in the Red River Valley. Link to Video Presentation: https://youtu.be/REU13yTtG8UenAttribution-NonCommercial-NoDerivs 2.5 CanadaSoil Moisture, Remote Sensing, MappingConference Presentation