Laserfiche WebLink
Methods Used to Estimate Effective Precipitation <br />Effective precipitation is the amount of rainfall that occurs during the growing season, is not lost to run- <br />off or deep percolation, and becomes available to meet the evapotranspiration needs of the crops. It is <br />important that a model includes this water since the amount of water reduces the irrigation water <br />requirements. <br />Can Effective Precipitation in Excess of Evapotranspiration Be Carried Over to the Next Month? <br />If the amount of effective precipitation is more than the plant can use during one month, then that <br />amount should be carried over to the next month. Otherwise, the model may over-estimate CIR in the <br />following month. <br />In Areas of High Water Tables, Does the Model Reduce the CIR to Account for the Portion of <br />Evapotranspiration Met By Ground Water? <br />How Easy Will the Model Integrate with the CRDSS? <br />Will it be easy to include the model into CRDSS by using the portions needed and excluding the portions <br />not needed? The factors that determine these criteria are the language in which the model is written; <br />whether the model is interactive or data driven; the quality of the programming, including whether the <br />model is modularly designed; and the availability of source code. <br />Is the Model in the Public Domain? <br />Only models for which the source code can be obtained and modified can be incorporated into CRDSS <br />because of the need to customize the models as well as incorporate them into the overall system. <br />How Well Documented is the Model? <br />Some of the models have very good manuals that make evaluating and using them easier. Users? <br />manuals as well as technical documentation of a model makes it easier in the long run to incorporate and <br />use the code for a model. <br />The Quality of the Programming Code <br />How the model's source code is organized and written can affect the portability of the code, the ease of <br />integration with other codes, and the ease of use and modification of the code; in the long run, this will <br />affect the maintainability of the code. <br />3.0 CONCLUSIONS AND RECOMMENDATIONS <br />It was not deemed necessary to evaluate every model by actually running it on a computer with a sample <br />data set. Therefore, a long list of models was developed and evaluated based on the criteria in this task <br />memorandum. The long list was then refined to a short list and the short list was evaluated more <br />thoroughly. The next two task memorandums describe each model short-listed in detail. <br />3 <br />A275 01.09.95 1.14-3 CSU IDS <br />