2002 Long Island South Shore Estuary Benthic Habitat Polygon Data Set

In June 2002, 200 1:20,000 scale conventional-color metric film diapositives for Long Island, New York were collected as part of an effort to map submerged aquatic vegetation (SAV) in Long Islands South Shore bays. They were provided by New York State Department of State's Division of Coastal Resources. Photographs were taken at low tide and during times that the growth stage of the SAV allowed for clear identification. Care was taken to minimize the effects of turbidity, sun glint, wind, and haze on the photos. The photos were scanned at a resolution of 15 microns. Ground control points were collected primarily from NYSDS 2 ft orthophotos. Additional control points were collected from USGS DOQQs where coverage from the primary source was lacking. All elevations were derived from USGS digital elevation models. A bundle block adjustment was performed using Albany and exterior orientation parameters were calculated. Boeing/Autometric's Softplotter was used to orthorectify the photos. The images were then dodged and mosaicked using Z/I's Orthopro. No additional color-balancing was performed as the mosaic's intended purpose was the delineation of benthic habitats. The mosaic was then output into 1000m by 1000m tiles with a 0.5m pixel resolution. The naming convention uses the first 3 numbers of the UTM x coordinate followed by the first 4 numbers in the UTM y coordinate of the southwest corner. Stereo digital images were created and the habitat features were interpreted and digitized on screen using softplotter microstation resulting in accurate and efficient 3D extraction of the data. Habitats were delineated with a high level of detail with the minimum mapping unit (MMU) being 0.01 hectares(approx.10m x 10m).The digitized polygons have the following specifications: Vertex Distance less than 1.0 m Node Snap Distance less than 4.0 m Arc Snap Distance less than 4.0 m During August 2002, NOAA staff collected 95 field observations throughout the study area and this information was incorporated into the map. In June 2003, after reviewing the photography, questionable areas were visited by Greenhorne and O'Mara staff and the findings were subsequently applied to the map. The map layers show delineated polygons and lines representing benthic habitat data. Each polygon feature is given a 1,2,3 or 4 digit number representing 11 habitats. The item numbers are stored in the attribute table under Text. The benthic data is classified according to the System for Classification of Habitats in Estuarine and Marine Environments (SCHEME). This system is fully described in "Development of a System for Classification of Habitats in Estuarine and Marine Environments (SCHEME) for Florida, Report to U.S. EPA - Gulf of Mexico Program, Florida Fish and Wildlife Conservation Commission, Florida Marine Research Institute. Review Draft 12/04/02."The collected data was converted to an ARCGIS format for quality control and delivery. The data was assessed for horizontal spatial accuracy and thematic agreement during 2003.

Original contact information:

Contact Name: Richard Eastlake

Contact Org: Greenhorne and O'Mara,Inc.

Title: Technical Director

Phone: 727 576 0402

Email: reastlake@g-and-o.com

At the urging of Long Island citizens, the New York State Legislature passed the Long Island South Shore Estuary Reserve Act in 1993. The Act recognized the South Shore Estuary as an unparalleled resource and provided for development of a Comprehensive Management Plan (CMP) to improve and protect the health of the estuary's ecosystem while increasing public access opportunities and sustaining economic activities. The Act also created a South Shore Estuary Reserve Council to prepare the plan and promote its implementation. The Council represents local governments, commercial and recreational fishing businesses, business associations, conservation and scientific organizations, education institutions and three State agencies including the New York State Department of State?s Division of Coastal Resources.One of the critical needs identified by the CMP is a baseline inventory of Submerged Aquatic Vegetation (SAV) resources, one of the Reserve?s most important living marine resources. Following the establishment of a comprehensive baseline SAV data set, the CMP calls for a long term monitoring program The monitoring will be designed to evaluate the level of success in attainment of specific site goals and the cumulative achievement of projected improvements in estuarine resources from habitat restoration and protection activities. The ultimate goal is to support productivity of commercially and ecologically important estuarine species by sustaining existing habitats of high functional quality and restoring degraded habitats. This data set will serve as the baseline inventory of benthic habitats for the Reserve.

10097

The study area covers approximately 443 km2 from the west end of Long Beach Island in Nassau County to Heady Creek at the east end of Shinnecock Bay in Suffolk County.

While every effort has been made to ensure that these data are accurate and reliable within limits of the state of the art, NOAA will not assume any liability for damages caused by any inaccuracies in these data, or failure in their use.

None

None

Data is available upon request

Data should not be changed or modified by anyone other than NOAA

Microsoft Windows 2000 Version 5.0 (Build 2195) Service Pack 2; ESRI ArcCatalog 8.2.0.700

A quality control inspection of the polygon labels was manually and digitally performed. A quality control AML verified that there are no unattributed polygons, label errors, node errors or invalid codes found in the coverage.Thematic accuracy assessment methods: The thematic accuracy of the benthic habitat data set was assessed via sampling of 99 field points generated through a random selection process. This was accomplished by randomly selecting ten percent of the polygons in each class of interest at the two digit SCHEME classification level ? Unconsolidated bottom, Submerged Rooted Vascular (SRV or SAV), and Algae. The random point generation was accomplished using the MILA extension in ESRI ArcView 3.X. The point generation process used the entire polygon as the sample unit. The actual geographic coordinate is derived from the polygon centroid. These points are then used as the target to navigate to in the field. Due to depth limitations, the actual centroid coordinate is often not accessible. In these cases, a representative area is surveyed as far into the polygon, from the deep edge into the shallows toward shore, as is possible. Navigation to each field point is accomplished using a digital NOAA nautical chart, real time Wide Area Augmentation System (WAAS) differentiated GPS (Garmin 76 unit) and a ruggedized field lap top PC. The PC displays the chart of the area with each ?target? point as well as a real time symbol display of the boat location via the GPS. This system allows navigation precisely to the ?target? point.The field observation of habitat type at each point was conducted in one of two ways. In shallow or extremely clear water direct observations were made from the boat. In deeper areas or areas of unclear water a towed underwater video camera was deployed. The camera provides a live video feed to a monitor. The camera is towed long enough (~2-4 minutes) at each station to provide a complete assessment of the dominate habitat type. The ?field? classes were recorded to at least the two digit SCHEME classification level and where possible to the three digit level.The ?field? classification for each point is compared to the ?lab? classification of each point in an error matrix. The resulting accuracy at the 2 digit SCHEME classification code level is 87%. At the 3 digit SCHEME classification code level the accuracy is 80%.

The accuracy of this digital data is based on the compilation from the georeferenced scanned images. The digital boundary in respect to the scanned images exceeds the 1:24000 scale National Map Accuracy Standards. The horizontal accuracy was assessed according to the FGDC National Standard for Spatial Data Accuracy (NSDDA) methodology. This approach compares the Root Mean Square Error (RMSE) value for a set of paired points, one point from the test imagery and one point from a source of higher spatial accuracy, to verify the imagery has met the contract specifications. For this project, 20 spatial accuracy points were collected using a Trimble GeoExplorer 3 GPS at terrestrial features spaced throughout the study area. According to the NSDDS methodology, the polygonal boundaries are +/- 5.5 meters.

The accuracy of this digital data is based on the compilation from the georeferenced scanned images. The digital boundary in respect to the scanned images meet the 1:24000 scale National Map Accuracy Standards.

All polygons and lines have been digitized and labeled from the orthorectified geotiff images provided by Greenhorne and O'Mara,inc. Coded features were cross-referenced with the SCHEME (Florida System for Classification of habitats in Estuarine and Marine Environments). A quality control inspection has been performed to eliminate any missing or duplicate polygon labels in the coverage. All inconsistent data was corrected. Six images in flightline 2 (21, 23, 25, 27, 29, and 31) were not included in the AT solution because these frames consisted entirely of water. These images were warped using an affine transformation to be included in the mosaic. For 2 of the images (25 and 27) there were not enough photo-identifiable locations to accurately perform the affine transformation. The transformation was still attempted to prevent a hole in the mosaic. The accuracy for these images is unknown. The affected mosaic tiles are: 6584505, 6584506, 6594504-6594507, 6604505-6604507, 6614505-6614508, 6624506-6624508, 6634506-6634508, 6644506-6644508.

The tests of logical consistency were performed using ESRI's ARC/INFO ARCGIS software. The dataset was built for internal arcs and polygons. The coverage has not been edited since the last build or clean.