کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
4978334 | 1452261 | 2017 | 6 صفحه PDF | دانلود رایگان |
- A daily precipitation reconstruction method is compiled in an Open Source R package.
- Three functions are provided to apply quality control, fill gaps and create grids.
- Users are able to obtain complete datasets and estimate new data at ungauged locations.
Daily precipitation datasets are usually large, bulky and hard to handle, but they are of key importance in many environmental studies. We developed a tool to create custom datasets from observed daily precipitation records. Reference values (RV) are computed for each day and location using multivariate logistic regression with altitude, latitude and longitude as covariates. The operations were compiled in an Open Source R package called reddPrec. The reddPrec package consists of a set of functions used to: i) apply a comprehensive quality control over original daily precipitation datasets, flagging suspect data based on five predefined criteria; ii) fill missing values in original data series by estimating precipitation values using the 10 nearest observations for each day; and iii) create new series and gridded datasets in locations where no data were recorded.
Journal: Environmental Modelling & Software - Volume 89, March 2017, Pages 190-195