Abstract

hisstudy aims to estimate the soil moisture content (SMC) of topsoils over theNorthern Central Highland using the Sentinel 2A image basing on the developmentof a regional emprirical model for the estimation. Total 8 soil samplesrepresentating for 6 soil types in the study area were collected andsub-sampled into 80 samples with 10 interval SMC levels for each sample.Resultant soil spectral analysis indicated a high correlation of SMC and theratio of S2A band 8 (NIR, 865 nm) versus band 12 (SWIR2, 2190 nm), thereforethe SMC can be accurately estimated from the ratio with a logarithm function.Applications of the model into estimating SMC from 30 S2A scenes acquired overthe study area in March 5 years, since 2016 to 2020, show that the 2016 droughtis severer than the other year droughts. In March 2016, a one-third of thestudy area was coverered by topsoils with SMC lower than 25 %. The attention onthe correlation between SMC in topsoils which are coverered by vegetation andun-vegetation should be focused in the future for futher application in theagricultural drough management.