A rice yield estimation system was developed based on the crop growth model ORYZA and SAR-derived key information such as start of season (SOS) and leaf area growth rate. Results from study sites in ...South and South-east Asian countries suggest that incorporating remote sensing data, specifically Synthetic aperture radar (SAR), into a process-based crop model improves the spatial distribution of yield estimates. This article highlights the detailed methodology of SAR data incorporation into crop yield simulation and comprehensive validation of yield forecast and estimates in the Philippines, Vietnam, Cambodia, Thailand, and Tamil Nadu, India. Remote sensing data assimilation into a crop model effectively captures the responses of rice crops to environmental conditions over large spatial coverage, which otherwise is practically impossible to achieve. A process-based crop simulation model is used in the system to ensure that climate information is captured, and this provides the capacity to deliver a mid-season yield forecast for national planning and policy for rice. Good agreement between SAR-based yield and crop-cut-based yield and official yield statistics and ensuring efficiency of the processing suggest that the system is a promising solution for the needed timely information on rice yield for application in food security and policies, climate disaster management, and crop insurance programs.
Conventional supervised classification of satellite images uses a single multi-band image and coincident ground observations to construct spectral signatures of land cover classes. We compared this ...approach with three alternatives that derive signatures from multiple images and time periods: (1) signature generalization: spectral signatures are derived from multiple images within one season, but perhaps from different years; (2) signature expansion: spectral signatures are created with data from images acquired during different seasons of the same year; and (3) combinations of expansion and generalization. Using data for northern Laos, we assessed the quality of these different signatures to (a) classify the images used to derive the signature, and (b) for use in temporal signature extension, i.e., applying a signature obtained from data of one or several years to images from other years. When applying signatures to the images they were derived from, signature expansion improved accuracy relative to the conventional method, and variability in accuracy declined markedly. In contrast, signature generalization did not improve classification. When applying signatures to images of other years (temporal extension), the conventional method, using a signature derived from a single image, resulted in very low classification accuracy. Signature expansion also performed poorly but multi-year signature generalization performed much better and this appears to be a promising approach in the temporal extension of spectral signatures for satellite image classification.
Rice is the most important food security crop in Asia. Information on its seasonal extent forms part of the national accounting of many Asian countries. Synthetic Aperture Radar (SAR) imagery is ...highly suitable for detecting lowland rice, especially in tropical and subtropical regions, where pervasive cloud cover in the rainy seasons precludes the use of optical imagery. Here, we present a simple, robust, rule-based classification for mapping rice area with regularly acquired, multi-temporal, X-band, HH-polarized SAR imagery and site-specific parameters for classification. The rules for rice detection are based on the well-studied temporal signature of rice from SAR backscatter and its relationship with crop stages. We also present a procedure for estimating the parameters based on "temporal feature descriptors" that concisely characterize the key information in the rice signatures in monitored field locations within each site. We demonstrate the robustness of the approach on a very large dataset. A total of 127 images across 13 footprints in six countries in Asia were obtained between October 2012, and April 2014, covering 4.78 m ha. More than 1900 in-season site visits were conducted across 228 monitoring locations in the footprints for classification purposes, and more than 1300 field observations were made for accuracy assessment. Some 1.6 m ha of rice were mapped with classification accuracies from 85% to 95% based on the parameters that were closely related to the observed temporal feature descriptors derived for each site. The 13 sites capture much of the diversity in water management, crop establishment and maturity in South and Southeast Asia. The study demonstrates the feasibility of rice detection at the national scale using multi-temporal SAR imagery with robust classification methods and parameters that are based on the knowledge of the temporal dynamics of the rice crop. We highlight the need for the development of an open-access library of temporal signatures, further investigation into temporal feature descriptors and better ancillary data to reduce the risk of misclassification with surfaces that have temporal backscatter dynamics similar to those of rice. We conclude with observations on the need to define appropriate SAR acquisition plans to support policies and decisions related to food security.
Reliable and regular rice information is essential part of many countries' national accounting process but the existing system may not be sufficient to meet the information demand in the context of ...food security and policy. Synthetic Aperture Radar (SAR) imagery is highly suitable for detecting lowland paddy rice, especially in tropical region where pervasive cloud cover in the rainy seasons limits the use of optical imagery. This study uses multi-temporal X-band and C-band SAR imagery, automated image processing, rule-based classification and field observations to classify rice in multiple locations across Tropical Asia and assimilate the information into ORYZA Crop Growth Simulation model (CGSM) to generate high resolution yield maps. The resulting cultivated rice area maps had classification accuracies above 85% and yield estimates were within 81-93% agreement against district level reported yields. The study sites capture much of the diversity in water management, crop establishment and rice maturity durations and the study demonstrates the feasibility of rice detection, yield monitoring, and damage assessment in case of climate disaster at national and supra-national scales using multi-temporal SAR imagery combined with CGSM and automated methods.
To meet the growing demand for rice and to ease the pressure on mountain ecosystems in northern Laos, it has been proposed to reduce upland rice cultivation and to expand the area under paddy rice. ...We used Random Forest, a classification and decision-tree-based method, to characterize the areas currently under paddy cultivation, and to predict which areas are suitable for paddy. Topographic variables and accessibility to villages and roads were the most important predictors for the presence of paddy cultivation. There appears to be much land available that is suitable for expanding paddy areas in central and southern Laos but not in the north, where more than 40% of the rice area is on sloping land, and much less area is suitable. We conclude that expanding paddy-based rice production will be difficult in most parts of northern Laos.
The remote sensing-based information and Insurance for Crops in Emerging Economics (RIICE) project aims to develop a national rice information system that provides timely and accurate information on ...rice area production, yield estimates, and production losses due to calamities to address food security and crop insurance purposes. This project makes use of remote sensing imagery from Synthetic Aperture Radar (SAR) platforms to generate reliable rice area maps and derives crop status information from the imagery as input to a crop growth simulation model (CGSM) to estimate yield. All image analysis is performed with MAPscape-RICE in a fully automatic way and yield estimation is performed with Oryza2000 CGSM. First, a rice extent map is generated based on 350 ENVISAT Wide Swath images (75-100m resolution) acquired between 2003 and 2010. This provides a national level baseline estimate of the physical rice extent. Second, the actual rice area is generated using high resolution (3 m) COSMO-SkyMed images acquired approximately every 16 days from June-October 2012 over the Province of Leyte. This provides a detailed rice area map for the season as well as information on the start of season and the crop status on a bi-monthly basis. Leaf Area Index (LAI) measurements from fieldwork conducted at selected sites on the same day as the satellite pass are correlated with the satellite imagery to generate LAI maps. Third, rice yield per barangay village and municipality is estimated using the CGSM with input data from SAR imagery (planting dates and LAI), weather stations, soil maps, fieldwork (crop management and variety). These yield estimates are being assessed by RIICE partners to develop a yield index which can later be made the triggering basis for an insurance product. Preliminary result of the project in Leyte shows a promising result.