information for discriminating deciduous rubber planta-
tions—that is, from single-date (Dong
et al.
2013) to bi-tem-
poral (Li
et al.
2015) to tri-window. Therefore, we believe that
NBR
has wide expandability for mapping deciduous rubber
plantations. However, two factors should be considered when
our phenology-based
NBR
method is applied in other regions.
One is the limit of elevation set for detecting deciduous
rubber plantations, especially in the mountain area. In our
previous study, we put particular emphasis on the elevation
ranges between 600 and 1100 m to detect and map deciduous
rubber plantations. However, the small proportion of decidu-
ous rubber plantations below 600 m or above 1100 m might
have been neglected, as we noticed during the field trips in
Xishuangbanna. The other factor is that patches of newly
established or young rubber plantations at higher elevations
tend to be small, and the mixed-spectral-pixel issue usu-
ally leads to misclassification. Thus, the use of finer spatial
resolution images, such as Sentinel-2, for detecting deciduous
rubber plantations will improve cla
Second, the continuous expansio
plantations in the border region of
Since the 1990s, over 30 bi- and mu
cal cooperative mechanisms and su
been established between China,
MSEA
nations, and other
developed countries (Li
et al.
2018). Thus, this area has
undergone, and will continue to experience, extensive and
rapid land use/cover changes in borderlands (Liu
et al.
2013),
including expansion of rubber plantations. What makes this
study possible is explicitly understanding the applicability
and limitations of remotely sensed data of optical satellites,
such as
LTS
imagery (i.e.,
TM
,
ETM+
, and
OLI
). Lately we have
proven that historical
LTS
data, especially acquired during
the dry season, hold great potential in monitoring deciduous
rubber plantations (Li
et al.
2018; Xiao
et al.
2018). With the
resultant updated and historical-epoch maps of deciduous
rubber plantations derived from Landsat-based tri-window
CRNBR
, the trans-boundary expansion over the past 26 years
was clearly investigated in this study. In the geoeconomic
context, more studies on the spatial and temporal change
patterns, including trans-boundary expansion of deciduous
rubber plantations in the national border regions, should be
conducted. In our upcoming work we plan to comprehensive-
ly reveal the spatio-temporal expansion features of deciduous
rubber plantations in the borderlands of China, Laos, and
Vietnam since the 1990s.
Conclusions
In this study, a modified phenology-based tri-window algo-
rithm, the Change Rate of Normalized Burn Ratio (
CRNBR
),
was proposed and applied to map updated and historical
information on deciduous rubber plantations in Xishuang-
banna in 2016 and other five past epochs, namely the early
1990s, late 1990s, early 2000s, late 2000s, and early 2010s.
The area of deciduous rubber plantations increased more than
6.6 times in the past 26 years, from 463 to 3074 km
2
. At the
horizontal level, the expansion of deciduous rubber planta-
tions varied from concentrated to dispersed, before gradually
expanding into the southern border region with Myanmar
and Laos as well as into the eastern border region with Laos.
At the vertical level, more than 86.5% of the deciduous rub-
ber plantations were concentrated at elevations between 600
and 1000 m, although plantations expanded into higher and
lower elevations from 1991 to 2016. These plantations broke
through their previous upper elevation limits of about 1000 m
in the early 1990s, about 1100 m in the late 1990s, about 1200
m in the late 2000s, and about 1300 m in the early 2010s, and
also expanded to their lower limit of around 500 m in the
early 2010s. We believe that updated and historical maps of
deciduous rubber plantations form a solid database for future
investigation into the economic, social, and environmental
impacts of this dominant artificial landscape in Xishuang-
banna.
Acknowledgments
We acknowledge financial support for this project from the
Strategic Priority Research Program of Chinese Academy of
Sciences (Grants XDA20010203 and XDA20010201), and the
Program for BINGWEI Excellent Young Talents of Institute
of Geographic Sciences and Natural Resources Research,
Chinese Academy of Sciences (2018RC201). We also wish to
thank the Landsat Science Team for their support in techni-
cal and scientific input. We specially thank the editors and
anonymous reviewers for their comments that helped to
improve this manuscript.
g rubber plantation age information
TM
&
ETM+
time series data and pixel-
based image compositing. Remote Sensing of Environment 196:
89–100.
Chen, B., X. Li, X. Xiao, B. Zhao, J. Dong, W. Kou, Y. Qin, C. Yang,
Z. Wu, R. Sun, G. Lan and G. Xie. 2016. Mapping tropical
forests and deciduous rubber plantations in Hainan Island,
China by integrating PALSAR 25-m and multi-temporal Landsat
images. International Journal of Applied Earth Observation and
Geoinformation 50: 117–130.
Chen, H., Z.-F. Yi, D. Schmidt-Vogt, A. Ahrends, P. Beckschäfer, C.
Kleinn. S. Ranjitkar and J. Xu. 2016. Pushing the limits: The
pattern and dynamics of rubber monoculture expansion in
Xishuangbanna, SW China. PLoS One 11 (2):e0150062.
Dong, J., X. Xiao, B. Chen, N. Torbick, C. Jin, G. Zhang and C. Biradar.
2013. Mapping deciduous rubber plantations through integration
of PALSAR and multi-temporal Landsat imagery. Remote
Sensing of Environment 134: 392–402.
Fan, H., X. Fu, Z. Zhang and Q. Wu. 2015. Phenology-based
vegetation index differencing for mapping of rubber plantations
using Landsat
OLI
data. Remote Sensing 7 (5):6041–6058.
Grogan, K., D. Pflugmacher, P. Hostert, R. Kennedy and R. Fensholt.
2015. Cross-border forest disturbance and the role of natural
rubber in mainland Southeast Asia using annual Landsat time
series. Remote Sensing of Environment 169: 438–453.
Kou, W., X. Xiao, J. Dong, S. Gan, D. Zhai, G. Zhang, Y. Qin and L. Li.
2015. Mapping deciduous rubber plantation areas and stand ages
with PALSAR and Landsat images. Remote Sensing 7 (1):1048–
1073.
Kou, W., J. Dong, X. Xiao, A. J. Hernandez, Y. Qin, G. Zhang, B. Chen,
N. Lu and R. Doughty. 2018. Expansion dynamics of deciduous
rubber plantations in Xishuangbanna, China during 2000–2010.
GIScience & Remote Sensing 55 (6):909–925.
Li, H., T. M. Aide, Y. Ma, W. Liu and M. Cao. 2007. Demand for
rubber is causing the loss of high diversity rain forest in SW
China. Biodiversity and Conservation 16: 1731–1745.
Li, H., Y. Ma, T. M. Aide and W. Liu. 2008. Past, present and future
land-use in Xishuangbanna, China and the implications for
carbon dynamics. Forest Ecology and Management 255: 16–24.
Li, P., Z. Feng and C. Xiao. 2018. Acquisition probability differences
in cloud coverage of the available Landsat observations over
mainland Southeast Asia from 1986 to 2015. International of
Journal Digital Earth 11 (5):437–450.
Li, P., J. Zhang and Z. Feng. 2015. Mapping rubber tree plantations
using a Landsat-based phenological algorithm in Xishuangbanna,
southwest China. Remote Sensing Letters 6: 49–58.
Li, Z. and J. M. Fox. 2012. Mapping rubber tree growth in mainland
Southeast Asia using time-series MODIS 250 m
NDVI
and
statistical data. Applied Geography 32: 420–432.
696
September 2019
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
