Environ Monit Assess (2011) 177:353–373
DOI 10.1007/s10661-010-1639-5
Monitoring and identification of spatiotemporal landscape
changes in multiple remote sensing images by using
a stratified conditional Latin hypercube sampling
approach and geostatistical simulation
Yu-Pin Lin · Hone-Jay Chu · Yu-Long Huang ·
Chia-Hsi Tang · Shahrokh Rouhani
Received: 10 January 2010 / Accepted: 29 July 2010 / Published online: 17 August 2010
© Springer Science+Business Media B.V. 2010
Abstract This study develops a stratified con-
ditional Latin hypercube sampling (scLHS) ap-
proach for multiple, remotely sensed, normalized
difference vegetation index (NDVI) images. The
objective is to sample, monitor, and delineate spa-
tiotemporal landscape changes, including spatial
heterogeneity and variability, in a given area. The
scLHS approach, which is based on the variance
quadtree technique (VQT) and the conditional
Latin hypercube sampling (cLHS) method, selects
samples in order to delineate landscape changes
from multiple NDVI images. The images are then
Y.-P. Lin (B ) · H.-J. Chu · Y.-L. Huang · C.-H. Tang
Department of Bioenvironmental Systems
Engineering, National Taiwan University,
1, Sec. 4, Roosevelt Rd., Da-an District,
Taipei City 106, Taiwan
e-mail: yplin@ntu.edu.tw
H.-J. Chu
e-mail: honejaychu@gmail.com
Y.-L. Huang
e-mail: morris0109@hotmail.com
C.-H. Tang
e-mail: jesst17@hotmail.com
S. Rouhani
NewFields, 1349 W. Peachtree St. Suite 2000,
Atlanta, GA 30309, USA
e-mail: SRouhani@newfields.com
mapped for calibration and validation by using
sequential Gaussian simulation (SGS) with the
scLHS selected samples. Spatial statistical results
indicate that in terms of their statistical distribu-
tion, spatial distribution, and spatial variation, the
statistics and variograms of the scLHS samples
resemble those of multiple NDVI images more
closely than those of cLHS and VQT samples.
Moreover, the accuracy of simulated NDVI im-
ages based on SGS with scLHS samples is sig-
nificantly better than that of simulated NDVI im-
ages based on SGS with cLHS samples and VQT
samples, respectively. However, the proposed ap-
proach efficiently monitors the spatial characteris-
tics of landscape changes, including the statistics,
spatial variability, and heterogeneity of NDVI im-
ages. In addition, SGS with the scLHS samples
effectively reproduces spatial patterns and land-
scape changes in multiple NDVI images.
Keywords Stratified conditional Latin hypercube
sampling · Sequential Gaussian simulation ·
Landscape change · Remotely sensed images
Introduction
Patterns of landscape change are attributed to
complex interactions between physical, biolog-
ical, and social forces (Turner 1987), such as
fires, hurricanes, earthquakes, urbanization, and