Predicting Inundation Dynamics and Hydroperiods of Small, Isolated Wetlands Using a Machine Learning Approach

Abstract

The duration of inundation or saturation (i.e., hydroperiod) controls many wetland functions. In particular, it is a key determinant of whether a wetland will provide suitable breeding habitat for amphibians and other taxa that often have specific hydrologic requirements. Yet, scientists and land managers often are challenged by a lack of sufficient monitoring data to enable the understanding of the wetting and drying dynamics of small depressional wetlands. In this study, we present and evaluate an approach to predict daily inundation dynamics using a large wetland water-level dataset and a random forest algorithm. We relied on predictor variables that described characteristics of basin morphology of each wetland and atmospheric water budget estimates over various antecedent periods. These predictor variables were derived from datasets available over the conterminous United States making this approach potentially extendable to other locations. Model performance was evaluated using two metrics, median hydroperiod and the proportion of correctly classified days. We found that models performed well overall with a median balanced accuracy of 83% on validation data. Median hydroperiod was predicted most accurately for wetlands that were infrequently inundated and least accurate for permanent wetlands. The proportion of inundated days was predicted most accurately in permanent wetlands (99%) followed by frequently inundated wetlands (98%) and infrequently inundated wetlands (93%). This modeling approach provided accurate estimates of inundation and could be useful in other depressional wetlands where the primary water flux occurs with the atmosphere and basin morphology is a critical control on wetland inundation and hydroperiods.

Inundation–Desiccation State Prediction for Salt Pans in the Western Pannonian Basin Using Remote Sensing, Groundwater, and Meteorological Data

“…Using data until the end of March of each year, we predict whether a salt pan dries out ('desiccated') or remains 'inundated' during JASO of the same year. This binary classification scheme has already been used as the basis for modeling WE in a number of studies [87,90,91]. The simplicity of the inundation state in summer/fall, meaning its low temporal resolution, its low number of classes, and its low degree of mathematical abstraction (Section 3.1.2), in combination with our predictor setup and the RF algorithm, leads to relatively good model performance and model interpretability.…”

“…Advances in ML [70,71] have boosted the relevance of stochastic modeling for predicting lake WH [81][82][83][84][85]. Past research in modeling wetland inundation dynamics using ML methods is often restricted to using in situ measurements for identifying the presence of water [86,87]. Greater data availability provided by EO [39,88,89] has contributed to studies utilizing WE for modeling wetlands, although, to our knowledge, not for salt pans and in different temporal resolutions.…”

Spatiotemporal Dynamics and Driving Factors of Small and Micro Wetlands in the Yellow River Basin from 1990 to 2020

“…For instance, Cheng et al [19] discovered the significant role of independent small wetlands in nutrient retention in the USA. Riley et al [20] investigated the impact of hydrological cycle changes in small isolated wetlands on ecological functions. Atkinson et al [5] found that seasonal small wetlands at the landscape scale play a crucial role in maintaining biodiversity, with their hydrological characteristics influencing amphibian productivity and community dynamics.…”

Analysis of Dynamic Changes and Driving Factors of Small Wetlands Around Chaohu Lake Based on Remote Sensing Images

“…The prerequisite for the successful protection and management of small wetlands is the construction of an accurate inventory of small wetlands using accurate maps (Aldous et al, 2021;Riley and Stillwell, 2023). Existing studies generally categorize and extract small wetlands using different classification methods based on low resolution remote sensing image data (Guan et al, 2022;Hibjur Rahaman et al, 2023;Wang et al, 2021).…”