• RAMISETTI SRI DEVI BHARGAVI
• G.VENKATA SUBBARAO
• NUKALA ALEKHYA
• POOJITHA TONTA
• MUMMAREDDY RAJESH

Soil moisture is the key factor that regulates agricultural productivity, ecosystem health, and hydrologically driven natural hazards. The most significant timescales of accurate forecasting with respect to soil moisture are over the leads from days to months ahead of time; however, instrumental noise, extremes variability, or nonlinear responses to rainfall are strongly non-stationarities, especially in highly dynamic landscapes such as post-wildfire sites. This research explores the potential of machine learning algorithms, namely Random Forest(RF) and Gradient Boost(GB), in improving soil moisture forecasting. We have designed and evaluated RF and GB models with soil moisture and rainfall time series data collected at a steep gradient, post-wildfire site. The area offers some of the most demanding conditions for soil moisture predictions due to the highly dynamic landscape changes concerning rainfall. By employing these machine learning algorithms, we will be able to reduce the weaknesses of conventional models, which are usually not very good at providing an accurate forecast of more than a few hours. Our RF and GB models include features derived from both historical soil moisture and meteorological data, capturing complex nonlinear relationships between input variables and soil moisture levels. We compare the performance of these models with state-of-the-art forecasting methods and find out their ability to generate long-term forecasts of soil moisture with high accuracy, up to a period of 10–24 hours.

Machine Learning, ·Random Algorithm, ·Gradient Boost Algorithm, Precision Agricul·ture ,LIBS(Laser- induced breakdown spectroscopy) , Farm productivity ,Crop productivity Water usage optimization , Plant growth.

"MACHINE LEARNING TECHNIQUES FOR SOIL MOISTURE PREDICTION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 12, page no.e553-e561, December-2024, Available :http://www.jetir.org/papers/JETIR2412462.pdf

"MACHINE LEARNING TECHNIQUES FOR SOIL MOISTURE PREDICTION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 12, page no. ppe553-e561, December-2024, Available at : http://www.jetir.org/papers/JETIR2412462.pdf