Author(s):
Kevin P. Price, Dr.* - University of Kansas
Jonathan B. Thayn - University of Kansas
Fangfang Yu - University of Kansas
Matthew E. Ramspott, Dr. - University of Kansas
Derrick W. Voisey - University of Kansas
Jude H. Kastens - University of Kansas
Loretta C. Johnson, Dr. - Kansas State University
Abstract:
Conifer red cedar (Juniperus Virginiana) is a native and widely distributed tree species in the eastern Great Plains. Of all the woody species invading into the tallgrass prairie, red cedar may pose a greater threat than other invading species since it thrives on both lowland and upland area, and its evergreen growth-form allows it to more effectively compete for soil moisture and nutrients. Over the past several decades, invasion of red cedar into the tallgrass prairie has been widely documented and has raised concerns of land managers because the growth of the woody species can adversely affect grazing productivity. Furthermore, its invasion into the grasslands can alter biogeochemical and biophysical processes, trace gas fluxes, biodiversity, and consequently change vegetation climate interactions.
The goal of this study was to develop remote sensing methods for mapping changing tree distribution patterns and estimating percent cover. Methods used for estimating cover, and animations of tree distributions over a 15 year period will be discussed and demonstrated. A linear spectral mixing model approach was used to map tree cover. Our model was validated using e-Cognition classification of multispectral high spatial resolution airborne imagery. This approach was found promising for mapping the cover of this species over large geographic regions. A leaf-off imaging period was best for discriminating red cedar from other land cover types. Our findings support the thesis that red cedar geographic distribution is expanding and the density of the trees is increasing at an alarming rate.
