Crop area estimation using high and medium resolution satellite imagery in areas with complex topography

[1] Reliable estimates of cropped area (CA) in developing countries with chronic food shortages are essential for emergency relief and the design of appropriate market-based food security programs. Satellite interpretation of CA is an effective alternative to extensive and costly field surveys, which fail to represent the spatial heterogeneity at the country-level. Bias-corrected, texture based classifications show little deviation from actual crop inventories, when estimates derived from aerial photographs or field measurements are used to remove systematic errors in medium resolution estimates. In this paper, we demonstrate a hybrid high-medium resolution technique for Central Ethiopia that combines spatially limited unbiased estimates from IKONOS images, with spatially extensive Landsat ETM+ interpretations, land-cover, and SRTM-based topography. Logistic regression is used to derive the probability of a location being crop. These individual points are then aggregated to produce regional estimates of CA. District-level analysis of Landsat based estimates showed CA totals which supported the estimates of the Bureau of Agriculture and Rural Development. Continued work will evaluate the technique in other parts of Africa, while segmentation algorithms will be evaluated, in order to automate classification of medium resolution imagery for routine CA estimation in the future.

[1]  A L Hammond,et al.  Crop forecasting from space: toward a global food watch. , 1975, Science.

[2]  Marvin E. Bauer,et al.  Sampling Landsat classifications for crop area estimation , 1981 .

[3]  P. McCullagh,et al.  Generalized Linear Models , 1992 .

[4]  F. J. Gallego.  Remote sensing and land cover area estimation , 2004 .

[5]  B. Brisco,et al.  Rice monitoring and production estimation using multitemporal RADARSAT , 2001 .

[6]  Jeanne B. Etheridge,et al.  AREA ESTIMATION OF CROPS BY DIGITAL ANALYSIS OF LANDSAT DATA. , 1978 .

[7]  Sushil Pradhan,et al.  Crop area estimation using GIS, remote sensing and area frame sampling , 2001 .

[8]  C. Woodcock,et al.  Resolution dependent errors in remote sensing of cultivated areas , 2006 .

[9]  Cheryl Christensen.  The new policy environment for food aid: the challenge of sub-Saharan Africa. , 2000 .

[10]  Ashutosh Kumar Singh,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2010 .

[11]  Hongling Fang,et al.  Rice crop area estimation of an administrative division in China using remote sensing data , 1998 .

[12]  N. A. Quarmby,et al.  Linear mixture modelling applied to AVHRR data for crop area estimation , 1992 .

[13]  John A. Nelder,et al.  Generalized Linear Models , 1972, Predictive Analytics.

[14]  Alfred Stein,et al.  Bayesian classification and class area estimation of satellite images using stratification , 1998, IEEE Trans. Geosci. Remote. Sens..

[15]  V. N. Sridhar,et al.  Wheat production forecasting for a predominantly unirrigated region in Madhya Pradesh )India) , 1994 .

[16]  F. Hall,et al.  Global Crop Forecasting , 1980, Science.

[17]  R. B. Erb,et al.  The Use of LANDSAT Data in a Large Area Crop Inventory Experiment (LACIE) , 1975 .

[18]  M. E. Bauer,et al.  Evaluation of a Segment-Based Landsat Full-Frame Approach to Crop Area Estimation , 1981 .

[19]  Edward Clay,et al.  Responding to Change: Wfp and the Global Food Aid System , 2003 .

[20]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[21]  John A. Nelder,et al.  Generalized linear models. 2nd ed. , 1993 .

[22]  F. J. Gallego.  CROP AREA ESTIMATION IN THE MARS PROJECT , 1999 .

[23]  P. McCullagh,et al.  Generalized Linear Models , 1984 .

[24]  James P. Verdin,et al.  Grid‐cell‐based crop water accounting for the famine early warning system , 2002 .