透過您的圖書館登入
IP:3.17.150.89
  • 期刊

日射量估計模式比較之研究-以嘉義地區為例

Comparison of Model Estimation on Solar Radiation-A Case Study in Chiayi Area of Taiwan

摘要


日射量在農業生產或模擬作物生長模式的研究中,是不可或缺的氣象資料,但常受限於量測儀器不足或歷史資料的不完整,必須以估計方式來彌補資料的欠缺。日射量估計模式依其利用之氣象資料種類,可概分為日照型、氣溫型、雲量型及其他等類型,不同類型之模式各有其使用條件上的侷限及優缺點,通常以地理(緯度、海拔)及氣候等因素對於估計的影響最大。本研究以2003-2010年農業試驗所嘉義農業試驗分所與中央氣象局嘉義氣象站的氣象資料,對於日照型模式Angström-Prescott model及El-Metwally model,氣溫型模式Hargreaves model及Richardson model,以及雲量型模式Badescu model等5個估計模式進行參數計算及配適性指標R^2之比較。另以嘉義農業試驗分所之2011-2013年資料,及嘉義氣象站2011-2012年資料進行估計準確度指標RMSE、MBE、MABE、相關係數r及t統計量的驗證比較。結果顯示,2處氣象站一致以日照型模式之配適性及估計準確度的表現最好,其次為雲量型模式,而氣溫型模式最差。

並列摘要


Solar radiation is an indispensable climatic data for the use in estimating agricultural production or simulation model of crop growth. However, in case of malfunction of measurement devices, the missing data need to be provided by estimated values yielded from simulation models. Basically there are three types of simulation models for the estimation of solar radiation, sunshine-based model, temperature-based model and cloud-based model, by climatic data used. Each type of model has its own advantages and disadvantages, but is greatly affected by factors of geography (e.g., latitude and elevation) and surroundings. In this study, weather data from both Chiayi Agricultural Experiment Branch (CAEB) of Taiwan Agricultural Research Institute and Chiayi Weather Station (CWS) of Center Weather Bureau in 2003-2010 were used for parameters calculation and goodness of fit (R^2). Five simulation models, Angström-Prescott model, El-Metwally model, Hargreaves model, Richardson model and Badescu model, were adopted. Weather data of 2011-2013 from CAEB and 2011-2012 from CWS were used for validation the accuracy of the estimated values, by the indices of RMSE, MBE and MABE, correlation coefficient (r), and t-statistic. Results of validation indices and goodness of fit indicated that sunshine-based models had the better performance in these two weather stations, followed by cloud-based model. On the other hand, temperature-based models had the least performance.

被引用紀錄


黃勁嘉(2017)。應用紅外線熱影像技術檢測磁磚溫差與自然環境因子關係之研究〔碩士論文,國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU201701246

延伸閱讀