On the Ground or in the Air? : A Methodological Experiment on Crop Residue Cover Measurement in Ethiopia

Kosmowski, Frederic ; Stevenson, James ; Campbell, Jeff ; Ambel, Alemayehu ; Tsegay, Asmelash H.

World Bank, Washington, DC

关键词: agricultural survey;    crop residue;    mulch;    remote sensing;    NDTI;   

DOI  :  10.1596/1813-9450-7813
RP-ID  :  WPS7813

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

Maintaining permanent coverage of thesoil using crop residues is an important and commonlyrecommended practice in conservation agriculture. Measuringthis practice is an essential step in improving knowledgeabout the adoption and impact of conservation agriculture.Different data collection methods can be implemented tocapture the field level crop residue coverage for a givenplot, each with its own implications for the survey budget,implementation speed, and respondent and interviewer burden.This study tests six alternative methods of crop residuecoverage measurement among the same sample of ruralhouseholds in Ethiopia. The relative accuracy of thesemethods is compared against a benchmark, the line-transectmethod. The alternative methods compared against thebenchmark include: (i) interviewee (respondent) estimation;(ii) enumerator estimation visiting the field; (iii)interviewee with visual-aid without visiting the field; (iv)enumerator with visual-aid visiting the field; (v) fieldpicture collected with a drone and analyzed withimage-processing methods; and (vi) satellite picture of thefield analyzed with remote sensing methods. Results of themethodological experiment show that survey-based methodstend to underestimate field residue cover. When quantitativedata on cover are needed, the best estimates are provided byvisual-aid protocols. For categorical analysis (such asgreater than 30 percent cover or not), visual-aid protocolsand remote sensing methods perform equally well. Amongsurvey-based methods, the strongest correlates ofmeasurement errors are total farm size, field size,distance, and slope. The results deliver a ranking ofmeasurement options that can inform survey practitioners and researchers.