Development of data-driven statistical models for daily water table depth prediction and variable selection: Two case studies in coastal plain forested wetlands

Abstract

Accurately predicting water table dynamics is vital for sustaining groundwater resources, ecological functions, and anthropogenic activities. This study evaluates autoregressive model with a) prediction under sparsity assumption within model coefficients, b) allows lags present in both dependent and independent variables for estimating daily water table depth using hydroclimatic data from the USDA Forest Service Santee Experimental Forest (SC) and D1 (NC). Data from 2006–2019 (SC) and 1988–2008 (NC) were used, with predictors including soil and air temperature, precipitation, wind, and radiation. For WS80, RMSE during the dormant season was 10.09cm, with daily testing phase RMSE 14.94 cm. The model achieved an R² 0.93 for 2019 (dry year) and 0.96 for 2016 (wet year). Solar radiation, rainfall, and wind direction were among the most influential variables. This predictive model can aid forest managers and hydrologists in using water tables for assessing wetland hydrology and related ecosystem functions in management decisions and provide out-of-sample prediction with reasonable accuracy.

Citation

Manna, Alokesh; Mehan, Sushant; Amatya, Devendra M. 2026. Development of data-driven statistical models for daily water table depth prediction and variable selection: Two case studies in coastal plain forested wetlands. Communications in Statistics: Case Studies, Data Analysis and Applications: 1-47. https://doi.org/10.1080/23737484.2026.2630287