Combining satellite remote sensing and surveys to understand persistent yield variation--- a case study in North China Plain

Abstract:

A large gap between maize yields on average farmers’ fields and the highest yields achieved by either experiment or farmers is typical throughout the developing world, including in the North China Plain (NCP). This maize yield gap as identified by previous studies indicates large opportunities for raising yield by improving agronomy. Quzhou county is typical of the winter-wheat summer-maize system in NCP where the average plot size is as small as 0.25 hectares. To analyze this cropping system amidst the challenge of substantial heterogeneity, we identified fields that were either persistently higher or lower yielding according to the remote sensing yield estimates, and then conducted detailed field surveys. We found irrigation facility to be a major constraint to yield both in terms of irrigation water quality and farmers’ access to wells. In total, improving the access to unsalty water would be associated with a 0.32t/ha (4.2%) increase in multi-year average yield. In addition, farmers’ method of choosing cultivar, which likely relates to their overall knowledge level, significantly explained yield variation. In particular, those choosing cultivars according to technician advice, personal experiences and high yielding neighbors’ advice had on average higher yield than farmers that either followed seed sellers’ advice or collectively purchased seeds. Overall, the study presents a generalizable methodology of assessing yield gap as well as its persistent factors using a combination of satellite and survey data.