| Summary: | Based on the importance of rice as a staple food most of the people of Indonesia,
the volume of rice production every year should be controlled, through the estimation of
production. Various methods have been developed to estimation production of rice, one
of which by using a remote sensing method. The aims of the study are : (1). To determine
a suitable the transformation of the vegetation index is used as the base for production
estimation of rice, (2). To asses the ability of Landsat 7 ETM+ imagery in the extraction
data of rice fields area to estimate the production of rice, and (3). To find our the spatial
distribution of production estimation of rice in Sragen regency based on transformation of
vegetation index of Landsat 7 ETM+ imagery and analysis of Geographic Information
Systems (GIS).
The main data source is Landsat 7 ETM+ imagery recorded on 30 August 2002,
which is used for extraction rice fields data and production estimation of rice. Land cover
maps and landforms maps wich derived from Landsat 7 ETM+ imagery is used as the
basis for production estimation of rice. Transformation of vegetation index method used
in this study is the transformation of NDVI and SAVI. Productivity of rice data derived
from measurements field. The data analysis used the spatial descriptions using analysis of
Geographic Information Systems(GIS).
Results of this research show that : (1). Based on the results of correlation
between the productivity of rice with the brightness value, the correlation coefficient (r)
between variables SAVI with productivity is 0.841 and the value of r ² = 0.707, while the
value of the correlation coefficient (r) between variables NDVI with productivity is 0.844
and the value of r ²= 0, 712, so between SAVI and NDVI variables with productivity
variables have positive correlation and strong correlation. Thus, the transformation of
NDVI and SAVI indices can be used as the basis for estimation production of rice in the
research area. (2), Based on the interpretation of test matrices, obtained accuracy of
image interpretation of Landsat 7 ETM + is 90.22%. Thus, the Landsat 7 ETM+ imagery
have a high accuracy for the extraction of rice fields of data required for estimation
production of rice. The accuracy of test results obtained at 89.063%, thus the Landsat
ETM+ imagery have a high accuracy for the production estimation of rice, so an
alternative method that can be used to estimate rice production. (3). Based on the results
of the analysis of Landsat 7 ETM +, the estimated production of rice in Sragen regency of
125,110.11 tons for a single harvest, or amounting to 375.060,45 tons in 2002. The
highest rice production found on alluvial plains with volcanic material landforms, on the
foot slope landforms, and alluvial plains with sedimentary material landforms. Subdistrict
Sidoharjo, Ngrampal, Sragen and Plupuh a district that has a substantial production of
rice, while the district that has a relatively sma ll rice production is Subdistrict Jenar,
Tangen, and Miri.
|
|---|