Identification and Characterization of Soil Limiting Factors to Rice Yield through Remote Sensed Information

博士 === 國立中興大學 === 土壤環境科學系所 === 98 === In Taiwan, rice is the most important food, and is also the biggest area crop. Even in the same climate and field management, spatial distribution diverse always appear on the field paddy rice yield. It demonstrated that there were in the existence of limiting p...

Full description

Bibliographic Details
Main Authors: Yi-Ping Wang, 王依蘋
Other Authors: Yuan Shen
Format: Others
Language:zh-TW
Published: 2010
Online Access:http://ndltd.ncl.edu.tw/handle/55391033438367657214
Description
Summary:博士 === 國立中興大學 === 土壤環境科學系所 === 98 === In Taiwan, rice is the most important food, and is also the biggest area crop. Even in the same climate and field management, spatial distribution diverse always appear on the field paddy rice yield. It demonstrated that there were in the existence of limiting paddy rice yield factors in soil, we called it “soil limiting factors”. Because field research and soil characteristics surveying consume time and money, it is hard to identify or characterize soil limiting factors in rice yield. Farmers always use excessive fertilizer to raise rice yield, but some studies prove that the lost of fertilizer not only pollute the ecological environment, but also harm the human health indirectly. Our purpose is using the establishment of remote sensed information to identify the soil limiting factors, to suggest farmers appropriate field management methods, to maintain high yield and reduce pollution. The study is divided into five chapters. In chapter 1, we explain our purpose. In chapter 2, we study about atmospheric correction protocols using FLAASH to retrieve effectively surface reflectance from SPOT imageries for regions which have large extends of paddy rice fields are presented. Examples are given to demonstrate that the proposed protocols work well under various atmospheric and surface conditions. In chapter 3, we use SPOT satellite imageries to forecast large-area rice yield, the possibility of forecasting rice yield rate in Taiwan with a high degree of accuracy using existing optical remote sensing data sources in combination with an empirical regression type yield model has been proven. In chapter 4, through remote sensed yield information, spatiotemporal trend maps of yield classification were first determined for a 200ha paddy rice fields under conventional two-cropping system in central Taiwan. Soil and plant samples were then collected from areas of different yield classes. Through statistical analysis and interpretations based on the observed differences in soil characteristics and rice yield component performances. In chapter 5 we believed that the systematic approach developed in this study has the potential to expedite the work of identification and characterization of the yield limiting factors in other paddy rice grown area because multiple year/crop season yield maps, usually were not available, can now be retrieved from historical satellite images.