No abstract

No abstract

In the presented methodology, multitemporal Landsat images were used to develop enhanced information about complex assemblages of vegetation and patterns of seasonal land cover variability, thereby facilitating improved land use and land cover (LULC) classification of urbanized areas among sensitive environments along the Mississippi Gulf Coast. For Landsat-5 and Landsat-7 images acquired for leaf-off and leaf-on conditions for 1991 and 2000, exploratory spectral analyses and field studies were conducted to detect and analyze patterns of spectral variability in land cover observed in the multitemporal image data. Patterns were identified of seasonal spectral data changes associated with seasonal vegetation changes for known land cover and land use types, thus characterizing patterns of seasonal LULC thematic change for the area. Detected seasonal variability for known land use and land cover types were used to develop formal classification rules based upon a thematic-change logic table. An image subset area based on United States Geological Survey (USGS) 1 : 24,000 quadrangles was used to develop a class-learning area within which unsupervised classification results were grouped into thematic classes. Signature files from the unsupervised classification results were applied to classify the balance of the study area. Individual images were classified for leaf-off and leaf-on conditions and thematic-change analyses were conducted. The formal class rules based on thematic-change logic were applied resulting in a classification that provided a level one accuracy exceeding 90% and a level two accuracy exceeding 85%. Index Terms-Coastal remote sensing, multitemporal, urban remote sensing, wetlands. I. INTRODUCTION L ULC classification of urbanized areas within lushly vegetated, sensitive coastal environments is a challenging task to accurately accomplish using synoptic Landsat imagery. Multitemporal imagery provides increased information about urbanized areas within complex assemblages of seasonally variable vegetation in temperate coastal regions. However, the com-Manuscript

No abstract

The Community Rating System (CRS) was introduced to encourage community-level flood mitigation and increase household-level flood insurance uptake through the National Flood Insurance Program (NFIP). Using historical data of policies-in-force and flood damage claims from 1998-2014 for all NFIP communities in Alabama and Mississippi, we estimate the relationship between community participation in the CRS and the number of policies-in-force, as well on flood damage claims. We find a significant, positive, and generally increasing effect of CRS participation on insurance uptake. Reduced flood damage claims are found to be limited to communities with a very high level of CRS participation (Class 5) only.