Development of models to estimate leaf wetness duration combining empirical methodology and physical concepts
Date
2000
Authors
Kim, Kwang-Soo
Major Professor
Advisor
Taylor, S. Elwynn
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Foliar diseases can reduce yield in many crops. One of the most important factors that govern the development of foliar diseases is the duration of environmental wetness on a leaf. Therefore, a reliable method to assess leaf wetness duration (LWD) is desirable. Although electronic wetness sensors are inexpensive and widely available, their application is limited by the cost and effort required to maintain a monitoring system. Utilization of models to estimate wetness duration is one way to avoid problems associated with wetness sensors. In this study, non-standard weather data (e.g. cloud cover and air temperature at a surface level) were incorporated into an empirical model for LWD estimation to enhance its accuracy, and a LWD model using site-specific data was designed and evaluated. First, a new empirical model, built on air temperature at 0.3 m high above ground, wind speed, and relative humidity (RH), showed similar performance to the CART/SLD model proposed by Gleason et al. (1994). The model predicted more accurately occurrence of a wetting event than the original model because of the air temperature data at the surface level.
Even though the modified model can predict wetness from both rain and dew, considering the need for installation of additional sensors, the new model does not appear to offer a clear advantage over the CART/SLD model. Second, LWD was estimated by joining CART/SLD model with a wind speed correction to 1-m or 0.3-m height. This model estimated LWD with less error than other models that used either a different wind speed correction algorithm or only site-specific data. This model also estimated LWD with acceptable accuracy under daytime or rain conditions. The CART/SLD model with wind speed correction showed little variation in accuracy of estimation of LWD over the study area, suggesting that the model may be useful to estimate LWD throughout the midwestern United States.
Series Number
Journal Issue
Is Version Of
Versions
Series
Academic or Administrative Unit
Type
thesis