Wetland mapping at the landscape scale using remotely sensed data requires both affordable data and an efficient accurate classification method. Random forest classification offers several advantages over traditional land cover classification techniques, including a bootstrapping technique to generate robust estimations of outliers in the training data, as well as the capability of measuring classification confidence. Though the random forest classifier can generate complex decision trees with a multitude of input data and still not run a high risk of over fitting, there is a great need to reduce computational and operational costs by including only key input data sets without sacrificing a significant level of accuracy. Our main questions for this study site in Northern Minnesota were: (1) how does classification accuracy and confidence of mapping wetlands compare using different remote sensing platforms and sets of input data; (2) what are the key input variables for accurate differentiation of upland, water, and wetlands, including wetland type; and (3) which datasets and seasonal imagery yield the best accuracy for wetland classification. Our results show the key input variables include terrain (elevation and curvature) and soils descriptors (hydric), along with an assortment of remotely sensed data collected in the spring (satellite visible, near infrared, and thermal bands; satellite normalized vegetation index and Tasseled Cap greenness and wetness; and horizontal-horizontal (HH) and horizontal-vertical (HV) polarization using L-band satellite radar). We undertook this exploratory analysis to inform decisions by natural resource managers charged with monitoring wetland ecosystems and to aid in designing a system for consistent operational mapping of wetlands across landscapes similar to those found in Northern Minnesota.

Accurate and current wetland maps are critical tools for water resources management, however, many existing wetland maps were created by manual interpretation of one aerial image for each area of interest. As such, these maps do not inherently contain information about the intra- and interannual hydrologic cycles of wetlands, which is important for effective wetland mapping. In this paper, several sources of remotely sensed data will be integrated and evaluated for their suitability to map wetlands in a forested region of northern Minnesota. These data include: aerial photographs from two different times of a growing season, National Elevation Dataset and topographical derivatives such as slope and curvature, and multitemporal satellite-based synthetic aperture radar (SAR) imagery and polarimetric decompositions. We identified the variables that are most important to accurately classify wetland from upland areas and discriminate between wetland types for a forested region of northern Minnesota using the d... Les cartes précises et à jour des milieux humides sont des outils essentiels pour la gestion des ressources en eau; toutefois, de nombreuses cartes des milieux humides existantes furent créées à l'aide de l'interprétation manuelle d'une seule image aérienne pour chaque zone d'intérêt. Comme tel, ces cartes ne contiennent pas d'information inhérente sur les cycles hydrologiques intra- et interannuels des milieux humides qui constitue une information essentielle pour la cartographie efficace des milieux humides. Dans cet article, diverses sources de données de télédétection seront intégrées et évaluées pour leur capacité à cartographier les milieux humides dans une région boisée située dans le nord du Minnesota. Ces données incluent : des photographies aériennes acquises à deux périodes différentes de la saison de croissance, un ensemble de données « National Elevation Dataset » et des dérivées topographiques comme la pente et la courbure, des images satellite multi-temporelles radar à synthèse d'ouverture ...