Mapping prairie remnants on the Hempstead Plains, Long Island, New York

Carole Neidich-Ryder, Patrick Joseph Kennelly

Research output: Contribution to journalArticle

Abstract

The Hempstead Plains, located in Nassau County, New York, contains remnants of the only naturally occurring prairie east of the Appalachian Mountains. It originally encompassed approximately 12,500 ha. Although the area receives higher amounts of rainfall for a typical tall-grass prairie ecosystem, approximately 114 cm of precipitation per year, its well-drained, dark-colored soil sited above glacial outwash, available natural seed bank, and history of fires enabled development of a tall-grass prairie. This study identified prairie remnants within the historical extent of the Hempstead Plains delineated by the 1928 soil survey from the United States Department of Agriculture. Image analysis of infrared color 8-bit orthophotographs was used for an unsupervised classification on a 156-ha primary study area containing a known prairie remnant, centered on the Red Golf Course at Eisenhower Park in East Meadow. The resulting 16 classes were combined into six more general classes before undergoing an error assessment based on field and orthoimagery ground-truthing of 1,000 random points. As confirmed by site visits, analysis was generally able to distinguish prairie grass from non-native grasses using remote sensing, as native warm-season prairie grasses are dormant from late fall to early spring. Overall accuracy for the six classes was 89 %. Accuracy of the warm-season grass class was 81 % for producer's accuracy and 83 % for user's accuracy. This study identified additional sites containing warm-season grasses and potential prairie remnants in Nassau County.

Original languageEnglish (US)
Pages (from-to)3011-3022
Number of pages12
JournalEnvironmental Monitoring and Assessment
Volume186
Issue number5
DOIs
StatePublished - Jan 1 2014

Fingerprint

Soil surveys
Ecosystems
prairie
Agriculture
Image analysis
Rain
Seed
Remote sensing
Fires
Color
Infrared radiation
Soils
grass
golf course
unsupervised classification
outwash
plain
soil survey
seed bank
image analysis

All Science Journal Classification (ASJC) codes

  • Environmental Science(all)
  • Pollution
  • Management, Monitoring, Policy and Law

Cite this

@article{8a930f3ac9704dfa99cf332d738abe32,
title = "Mapping prairie remnants on the Hempstead Plains, Long Island, New York",
abstract = "The Hempstead Plains, located in Nassau County, New York, contains remnants of the only naturally occurring prairie east of the Appalachian Mountains. It originally encompassed approximately 12,500 ha. Although the area receives higher amounts of rainfall for a typical tall-grass prairie ecosystem, approximately 114 cm of precipitation per year, its well-drained, dark-colored soil sited above glacial outwash, available natural seed bank, and history of fires enabled development of a tall-grass prairie. This study identified prairie remnants within the historical extent of the Hempstead Plains delineated by the 1928 soil survey from the United States Department of Agriculture. Image analysis of infrared color 8-bit orthophotographs was used for an unsupervised classification on a 156-ha primary study area containing a known prairie remnant, centered on the Red Golf Course at Eisenhower Park in East Meadow. The resulting 16 classes were combined into six more general classes before undergoing an error assessment based on field and orthoimagery ground-truthing of 1,000 random points. As confirmed by site visits, analysis was generally able to distinguish prairie grass from non-native grasses using remote sensing, as native warm-season prairie grasses are dormant from late fall to early spring. Overall accuracy for the six classes was 89 {\%}. Accuracy of the warm-season grass class was 81 {\%} for producer's accuracy and 83 {\%} for user's accuracy. This study identified additional sites containing warm-season grasses and potential prairie remnants in Nassau County.",
author = "Carole Neidich-Ryder and Kennelly, {Patrick Joseph}",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/s10661-013-3597-1",
language = "English (US)",
volume = "186",
pages = "3011--3022",
journal = "Environmental Monitoring and Assessment",
issn = "0167-6369",
publisher = "Springer Netherlands",
number = "5",

}

Mapping prairie remnants on the Hempstead Plains, Long Island, New York. / Neidich-Ryder, Carole; Kennelly, Patrick Joseph.

In: Environmental Monitoring and Assessment, Vol. 186, No. 5, 01.01.2014, p. 3011-3022.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Mapping prairie remnants on the Hempstead Plains, Long Island, New York

AU - Neidich-Ryder, Carole

AU - Kennelly, Patrick Joseph

PY - 2014/1/1

Y1 - 2014/1/1

N2 - The Hempstead Plains, located in Nassau County, New York, contains remnants of the only naturally occurring prairie east of the Appalachian Mountains. It originally encompassed approximately 12,500 ha. Although the area receives higher amounts of rainfall for a typical tall-grass prairie ecosystem, approximately 114 cm of precipitation per year, its well-drained, dark-colored soil sited above glacial outwash, available natural seed bank, and history of fires enabled development of a tall-grass prairie. This study identified prairie remnants within the historical extent of the Hempstead Plains delineated by the 1928 soil survey from the United States Department of Agriculture. Image analysis of infrared color 8-bit orthophotographs was used for an unsupervised classification on a 156-ha primary study area containing a known prairie remnant, centered on the Red Golf Course at Eisenhower Park in East Meadow. The resulting 16 classes were combined into six more general classes before undergoing an error assessment based on field and orthoimagery ground-truthing of 1,000 random points. As confirmed by site visits, analysis was generally able to distinguish prairie grass from non-native grasses using remote sensing, as native warm-season prairie grasses are dormant from late fall to early spring. Overall accuracy for the six classes was 89 %. Accuracy of the warm-season grass class was 81 % for producer's accuracy and 83 % for user's accuracy. This study identified additional sites containing warm-season grasses and potential prairie remnants in Nassau County.

AB - The Hempstead Plains, located in Nassau County, New York, contains remnants of the only naturally occurring prairie east of the Appalachian Mountains. It originally encompassed approximately 12,500 ha. Although the area receives higher amounts of rainfall for a typical tall-grass prairie ecosystem, approximately 114 cm of precipitation per year, its well-drained, dark-colored soil sited above glacial outwash, available natural seed bank, and history of fires enabled development of a tall-grass prairie. This study identified prairie remnants within the historical extent of the Hempstead Plains delineated by the 1928 soil survey from the United States Department of Agriculture. Image analysis of infrared color 8-bit orthophotographs was used for an unsupervised classification on a 156-ha primary study area containing a known prairie remnant, centered on the Red Golf Course at Eisenhower Park in East Meadow. The resulting 16 classes were combined into six more general classes before undergoing an error assessment based on field and orthoimagery ground-truthing of 1,000 random points. As confirmed by site visits, analysis was generally able to distinguish prairie grass from non-native grasses using remote sensing, as native warm-season prairie grasses are dormant from late fall to early spring. Overall accuracy for the six classes was 89 %. Accuracy of the warm-season grass class was 81 % for producer's accuracy and 83 % for user's accuracy. This study identified additional sites containing warm-season grasses and potential prairie remnants in Nassau County.

UR - http://www.scopus.com/inward/record.url?scp=84898859244&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898859244&partnerID=8YFLogxK

U2 - 10.1007/s10661-013-3597-1

DO - 10.1007/s10661-013-3597-1

M3 - Article

VL - 186

SP - 3011

EP - 3022

JO - Environmental Monitoring and Assessment

JF - Environmental Monitoring and Assessment

SN - 0167-6369

IS - 5

ER -