| Summary: | The area of vegetation cover indicates the amount ofcarbon storagein certain
area. The objective of this studyis to discuss theestimationof carbon storagebased
onland coverof urban area of Yogyakartacity and its surroundings by
usingmediumspatialresolutionsatellite imagery of ALOSAVNIR-2. The urban area of
Yogyakarta developed after the outer ringroad access opened. The expansion
ofresidential and commercialarea tookover the landvegetation area, bothforest
landandagricultural land.
This study used allometry method and regression model using satellite imagery
spectral value and field data to get the total urban carbon. Spectral value of ALOS
AVNIR-2 single bands and vegetation indices: simple ratio (SR), normalized
difference vegetation index (NDVI), triangular vegetation index (TVI), modified soiladjusted
vegetation index - 2 (MSAVI2), enhanced vegetation index (EVI), EVI-2,
visible atmospherically resistant vegetation index (VARIGreen), global environment
monitoring index (GEMI), and atmospherically resistant vegetation index
(ARVI).Field data was taken based on vegetation cover percentage in various land
covers.
The study results that ALOS AVNIR-2 single bands, SR and ARVI at Atsensor
Reflectance and At-surface Reflectance radiometric correction shows a weak
correlation between spectral value and carbon storage (r<0.76). Regression model then
inversed between the field data and field carbon storage to estimate carbon storage in
unsampled pixels. Accuracy assessment used the independent field set based on
standard error method. This accuracy assessment shows that EVI-2 is the best
vegetation index to estimate urban vegetation carbon in the study area than other
vegetation indices. δest EVI-2 is 0,862tonC/pixel. The largest carbon storage was
found in vegetation area such as vegetation area along the riverside and urban forest.
This study suggest that the use of medium spatial resolution satellite imagery for
estimating urban vegetation carbon storage needs to consider mixed pixels, because
the spectral value in the high density built-up area affects the spectral value of
vegetation.
|
|---|