In order to join virtual sessions, you must be registered and logged-in(Were you registered for the in-person meeting in Denver? if yes, just log in.) 
Note: All session times are in Mountain Daylight Time.

Mapping rice-based cropping patterns using Sentinel-1 SAR multi-temporal imagery: A case study in the district of Indramayu, West Java Province.

Authors: Josue Gabriel Yarleque Ipanaque*, Clark University, John Rogan, Graduate School of Geography, Clark University
Topics: Remote Sensing, Agricultural Geography, Geographic Information Science and Systems
Keywords: Sentinel-1, SAR, Support Vector Machine, Google Earth Engine, Remote Sensing, Sustainable Agriculture, Indonesia, Food Security.
Session Type: Guided Poster
Presentation Link: Open in New Window
Presentation File: Download

Sustainable rice production on a regional scale requires regular information on where, when and how rice is grown. Rice can be grown more than once per year in the same field, or in a system of alternating crops such as wheat, maize or vegetables. In Southeast Asia, some regions switch between rice and pond aquaculture in the same year establishing unique rice-based cropping patterns. A cropping pattern refers to the seasonal sequence and spatial arrangement of crops in a particular area for a given year. It is important to identify cropping patterns to estimate current rice production and provide insights for implementation and management to improve sustainable crop yield.

Synthetic Aperture Radar (SAR) imagery is proven to be reliable for characterization of rice-cropping patterns due to its sensitivity to water content in rice fields. Sentinel-1 data provide high spatial and temporal resolution C-SAR information. This research explores the use of 10 meters Sentinel-1 imagery to help characterize where and what cropping patterns exist in the district of Indramayu, located in the province of West Java, Indonesia. The objectives are: (1) track trends in the temporal signatures of rice-based fields using Sentinel-1 imagery from September 2017 to August 2018 to optimize data selection for rice-cropping patterns; and (2) classify different rice-based cropping patterns in the study area. A multi-band composite was created using metrics such as the 10th percentile, 50th percentile and standard deviation. Support vector machine classification was used in Google Earth Engine to identify two distinctive rice-based cropping patterns.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login