Publications

Sha, S; Wang, XP; Li, QZ; Li, WD (2016). Study on Yield Estimation of Spring Wheat basing on Hyperspectral data under Different Meteorological Condition in Semi-Arid Rain Fed Region. 2016 FIFTH INTERNATIONAL CONFERENCE ON AGRO-GEOINFORMATICS (AGRO-GEOINFORMATICS), 242-247.

Abstract
Statistically relating vegetation index to yield is a common wheat yield estimation method. In this paper, we investigate the relationship between a variety of spectral vegetation indices and yield factors for spring wheat in a semiarid, rain-fed, agricultural region under different meteorological conditions on the basis of relevant ground observations. We also analyze the yield estimation factor of remote sensing for spring wheat by regression method under different meteorological conditions. Results are as follows. 1) The theoretical yield per unit area, thousand-kernel weight, grains per ear, and number of productive tillers per square meters at the milk stage of maturity are relatively small. These data exhibit flat variation trends with spectral vegetation indices under drought conditions. By contrast, the trends under non-drought conditions are significantly changing. 2) Meanwhile, the spectral vegetation indices under drought and non-drought conditions are appreciably associated with the theoretical yields at the booting (0.01) and heading stages (0.001). 3) In the meteorological droughts, the aggregate value of the semi-arid water index at the booting and heading stages is suitable for use as the yield estimation factor for spring wheat. However, under the meteorological non-droughts, the RVI(p 780/p1750) at the booting stage is used as the yield estimation factor for spring wheat. The mean absolute percentage errors of the yield estimations in the two cases are 70.9% and 84.2%, respectively.

DOI:

ISSN:
2334-3168