Interior Least Tern Sandbar Nesting Habitat Measurements from Landsat TM Imagery: NAIP Validation Images
Dates
Publication Date
2018-10-05
Citation
Edward A. Bulliner, 20170215, Interior Least Tern Sandbar Nesting Habitat Measurements from Landsat TM Imagery: NAIP Validation Images: U.S. Geological Survey, http://dx.doi.org/10.5066/F7CV4GNG.
Summary
Sandbars of large sand-bedded rivers of the Central U.S. serve important ecological functions to many species, including the endangered Interior Least Tern (Sternula antillarum) (ILT). ILT are colonial birds which feed on fish and nest primarily on riverine sandbars during their annual breeding season of approximately May through July, depending on region. During this time, ILT require bare sand of sufficient elevation so as not to be inundated during the period between nest initiation and fledging of hatchlings. ILT were originally listed as endangered due in part to decreases in available sandbar habitat from river channelization and impoundment. Sandbars in Central U.S. rivers used by ILT are highly dynamic, with substantive changes [...]
Summary
Sandbars of large sand-bedded rivers of the Central U.S. serve important ecological functions to many species, including the endangered Interior Least Tern (Sternula antillarum) (ILT). ILT are colonial birds which feed on fish and nest primarily on riverine sandbars during their annual breeding season of approximately May through July, depending on region. During this time, ILT require bare sand of sufficient elevation so as not to be inundated during the period between nest initiation and fledging of hatchlings. ILT were originally listed as endangered due in part to decreases in available sandbar habitat from river channelization and impoundment. Sandbars in Central U.S. rivers used by ILT are highly dynamic, with substantive changes occurring across a wide range of temporal and spatial scales. River hydrology is the primary driver of sandbar morhpodynamics in these systems. Better characterization of sandbar area over time accounting for varying flow regimes allows for a better understanding of landscape-scale ecology for sandbar-dependent species such as the ILT. This dataset seeks to quantify sandbar area which may be used by ILT at the landscape scale and how it has changed over time using remote sensing techniques. The assessment of landscape scale trends in sandbar area over time requires datasets with relatively high temporal resolution and long periods of record covering large geographic areas, and remotely-sensed datasets must also consider changes in river stage. Land cover classification datasets within active channel masks were developed using all available images from the Landsat Thematic Mapper (TM) series of satellites meeting cloud-free (40% or less) and ice-free criteria. Landsat imagery was selected because of its long period of record, spatial coverage, and regular reimaging cycle, making it well suited to monitoring ILT sandbar habitat over time. These datasets were also attributed by discharge or stage using a new database integrating U.S. Geological Survey and U.S. Army Corps of Engineers river data with Landsat metadata. This dataset documents development of these riverine classification datasets with a focus on applicability to the ILT. This framework may be used for continuing monitoring of ILT sandbar nesting habitat.
The purpose of this dataset is to present methods used to create riverine land cover classification data (simply referred to as classification data) used to quantify sandbar area potentially used by the ILT across large rivers of the Central U.S. The database and classification products presented are intended to provide a systematic calculation of sandbar area across all useable images in the Landsat TM archive while accounting for varying river stage. Classified land cover datasets are presented for multiple flow segments from eight rivers, along with a database containing Landsat metadata and discharge records for all relevant Landsat images. The database encompasses all imagery for the eight classified rivers along with five additional rivers less commonly used by the ILT and not investigated as part of this dataset. These rivers consist of a very small fraction of the overall ILT population habitat and were deemed less important for this analysis. However, they were included in the database as a framework for future analysis if they should later be deemed important. This dataset was compiled using a combination of Landsat Climate Data Record (CDR) surface reflectance data products developed and processed by the USGS Earth Resources Observation Systems (EROS) Data Center (U.S. Geological Survey, 2014), discharge data from the USGS national stream gauging network, and dam release and river-stage records from the U.S. Army Corps of Engineers (USACE). The intent of this dataset is to provide a historic summary of sandbars visible from the Landsat TM satellites on large rivers used by the ILT given all available images in the Landsat TM period of record. Additionally, this dataset provides an analysis framework amenable for ongoing monitoring of sandbars useable by the ILT as nesting habitat. The associated files present manual classification of two NAIP images overlapping the Missouri and Red Rivers and compare the resulting classifications with the Landsat classifications presented in this dataset. Images, classified vectors, and classified rasters along with associated FCDC metadata for each file are contained in the .zip file.
