Mapping rice in the Philippines - where

  • About

  • Acknowledgement

  • Related items

rice extent map of the PhilippinesRice is cultivated in almost all of the provinces in the Philippines. Information on the location and extent of rice-growing areas is critical to helping quantify rice production and potential import needs and export opportunities. It can therefore help policy-makers with their decisions and be used to prioritize research and extension activities to help rice farmers.

In order to assess where rice is usually grown, IRRI and DA-PhilRice developed an automated system using multi-temporal MODIS (Moderate Resolution Imaging Spectroradiometer) satellite images, which is a modified version of the methodology developed by Xiao et al (2006) for mapping rainfed lowland and irrigated rice areas.

Our rice extent map of the Philippines is based on MODIS images from 2000-2012. This can be used as a baseline representation of the spatial distribution of rice across the major rice-growing areas in the Philippines.

The map has been validated following an extensive field campaign by DA-PhilRice, IRRI, and GIZ staff following a protocol devised by sarmap. Staff visited 24 provinces and collected field data in 2,770 locations. Based on this fieldwork, the overall accuracy of the map is 78.8%. In remote sensing, the target for a high quality classification is 85%. For this reason, this product is described as ‘research grade’.

rice Kappa index

Overall accuracy is calculated as the total number of correctly classified pixels (diagonal elements) divided by the total number of ground control points. This represents the proportion of the correctly classified features (rice, non-rice) on the map (predicted) and features on the ground (actual).

Kappa index is an assessment of the accuracy of the map compared to the accuracy that could have been achieved purely by chance. Higher is better. (click table to enlarge)

The rice extent map 2000-12 shows the general distribution of where rice is grown across the nation in a year. It is the best available public information on where rice is grown at the start of the 21st century.

MODIS images were used because they are free and available in near real time – every 8 days. The drawbacks are that the spatial resolution is limited to approximately 500m x 500m, or 25 hectares per pixel and that pervasive cloud cover can limit the amount of good quality images in the wet season. Field sizes are substantially smaller than 25ha, but contiguous areas of rice cultivation can still be detected. The method has resulted in a rice map that represents the general distribution of rice across the nation, but it cannot show each and every rice paddy.

To address these limitations, RIICE and the new DA-supported Philippines Rice Information System (PRISM) projects are using Synthetic Aperture Radar (SAR) images that can detect and map rice at higher detail regardless of cloud conditions. Maps based on SAR images will be developed and published over the next few years to build up an accurate and detailed map of rice growing areas for each region of the country.

This map was developed and validated by the Department of Agriculture (DA) Philippine Rice Research Institute (PhilRice), the International Rice Research Institute (IRRI), sarmap, and the German Society for International Cooperation (GIZ) as part of the DA-supported Rice Self-sufficiency Project (RSSP) and Remote Sensing-based Information and Insurance for Crops in emerging Economies (RIICE) projects.

Partners

  • DA-Philippines Rice Research Institute (DA-PhilRice)
  • International Rice Research Institute (IRRI)
  • sarmap
  • German Society for International Cooperation / Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ)

Donors

For more information, please contact:

Samarendu Mohanty
Social Sciences Division Head, IRRI
Nasreen Islam Khan
Senior Scientist and GIS Head, IRRI

Alice Laborte
GIS and Market Specialist, IRRI