Organic Corn In California



Organic Corn In California

Abstract

The objective of this study was to present a data fusion framework used to calibrate a crop growth model at the plot scale and to estimate yield at the regional scale on the basis of two types of data fusion algorithms, which reduces the uncertainty of regional yield estimations. First, based on local intensive observation, the simulated annealing algorithm was applied to obtain a parameter vector that was suited to the local crop variety. This scheme reduces model parameter uncertainty. Then, the ensemble Kalman filter (EnKF), a sequence filter algorithm, was adopted to integrate the areal crop growth information that was derived from remote sensing technologies into a crop growth model for precise regional yield estimation, which reduces uncertainties in the model structure or input data related to meteorological, soil, or filed management information. This proposed scheme and technology will provide an operational method for precisely estimating crop yields at regional scales.

Introduction

Accurate estimations of regional crop yields play an important role in food security (Macdonald and Hall, 1980 and Hutchinson, 1991). Two methods have been used to estimate crop yields, i.e., yield monitoring and model yield simulations (Maselli et al., 1992, Priya and Shibasaki, 2001 and Palosuo et al., 2011). In early studies, sampling-based investigations and statistical analyses were used to monitor yields. For this approach, crop yields from sample locations are surveyed and aggregated into a regional yield value. This method is laborious; therefore, sampling-based investigations have gradually been abandoned. With the development of remote sensing technology, satellite sensors can continuously obtain terrain surface information on a regional scale, which provides an attractive alternative method of monitoring the crop status (Tennakoon et al., 1992, Moriondo et al., 2007 and Claverie et al., 2012).

In addition to observational means, crop growth models can chart the evolution of crop growth, and they have proven to be valuable for yield forecasting and climate scenario analyses (Batchelor et al., 2002, Donatelli et al., 2002, Bouman et al., 2006 and Varella et al., 2010). Crop growth models mainly simulate biogeophysical processes in soil-crop-atmospheric systems to provide a continuous estimate of growth over time, and the actual crop yield is the result of complex interactions among such factors as the soil, atmosphere, plant genotype, and management practices (Moen et al., 1994 and Doraiswamy et al., 2005). A series of mechanistic crop growth models have been developed in recent decades. Of these models, the WOrld FOod STudies model (WOFOST) with its characteristic of genericness has been adopted worldwide (Van Keulen and Wolf, 1986, Supit et al., 1994 and Confalonieri et al., 2009).

However, crop growth models are simplifications of the agri-ecological systems they represent, and model structure factors are a significant source of uncertainity (Gao et al., 2011). Moreover, the use of crop models is also limited by uncertainties in the parameters or input data (Fang et al., 2008, Lizumi et al., 2009, Niu et al., 2009, Ceglar et al., 2011 and Challinor et al., 2012). Additionally, crop simulation models are very useful in evaluating crop growth at the field scale, but their implementation at a larger scale is ...