, the #1 open access web portal for PhD theses...

Why PhD theses...

PhD thesis is the result of years of hard work.

keyword researchMeasured by download count PhD theses are one of the most popular items world wide on open access repositories. But unless a thesis is published, it is very difficult for other researchers to find out about it and get access to it. Theses are often under-used by other researchers. attempts to address this issue by making it easy to identify and locate copies of many theses in various disciplines.

Best management practice development with the CERES-maize model for sweet corn production in north Florida

Best management practice development with the CERES-maize model for sweet corn production in north Florida
Jianqiang He


Institute of Food and Agricultural Sciences, University of Florida, Gainesville, FL 32611, UNITED STATES OF AMERICA.


Increasing nitrogen loads within the Suwannee River Basin of North Florida has become a major concern. Nitrogen fertilizer application in field crop production is proved to be the most import nitrogen contribution in this region. Florida ranks highest in the nation in the production and value of fresh market sweet corn. Thus it is necessary to develop research based nitrogen best management practices (N-BMPs) to reduce nitrogen leaching while keeping an acceptable yield in sweet corn production.

This study is an attempt to utilize the CERES-Maize mode of the Decision Support System for Agrotechnology Transfer (DSSAT) model as a platform to develop potential BMPs for sweet corn production in North Florida.

The results show that the non-restricted and restricted one-at-a-time (OAT) method can be used to conduct global sensitivity analysis for the CERES-Maize so as to select the most influential parameters for model calibration. The generalized likelihood uncertainty estimation (GLUE) method was proved to be a powerful tool for model parameter estimation, since the uncertainties in model input parameters were significantly reduced after GLUE was used to estimate the model input parameters. The uncertainties in model outputs were reduced correspondingly.

The comparison between the model simulated and field observed results of the seven treatments in a field plot experiment of sweet corn in 2006, shows that the model did a good job in predicting dry yield and phenology dates.

The results of BMP development with the calibrated CERES-Maize model show that if the growers could apply both irrigation water and nitrogen fertilizer more frequently but with smaller amounts in each application, this would result in an acceptable yield and a lower level of nitrogen leaching. The results showed a total nitrogen amount between 196 and 224 kg N ha-1 would be enough for sweet corn production in North Florida, which confirmed that the recommendation nitrogen amount (224 kg N ha-1) by Institute of Food and Agricultural Sciences (IFAS, Univerisity of Florida, was reasonable.

The results of uncertainty analysis of the CERES-Maize model for sweet corn simulation show that the weather was the dominant uncertainty contributor. This was because after two rounds of GLUE parameter estimation procedure, the uncertainties existing in input parameters were minimized.