To what extent does the large temporal variability in Irish precipitation mask underlying trends?
![Thumbnail](/themes/Mirage2/images/white_rectangle.jpeg)
File Type:
PDFItem Type:
reportDate:
2015Access:
openAccessCitation:
Emily Gleeson, Ciara O'Hara, Séamus Walsh, Sinéad Duffy, 'To what extent does the large temporal variability in Irish precipitation mask underlying trends?', [report], Met Éireann, Technical Note, 63, 2015Download Item:
Description:
A pertinent question about precipitation trends is how to distinguish a climate change signal from natural variability. This is particularly important for Ireland where, in addition to the high spatial variability, precipitation also has a high level of temporal variability. Future precipitation projections for Ireland are highly uncertain due to conflicting results from climate models. Due to this, using an observation-derived dataset we determined the magnitude of the changes in the spatially-averaged precipitation over Ireland that need to occur, and the length of the record that needs to be available, to separate an externally forced anthropogenic change from changes due to natural variability. We used a 71-year (1942-2012) 1km gridded dataset, derived from precipitation observations over Ireland, to generate a de-trended dataset of annual average precipitation for the country. Using this data, we then created artificial time series of annual precipitation of varying length in terms of years. These datasets had the same underlying natural variability as the original dataset but in addition to this we applied predefined external forcings. We then used statistical testing to determine the minimum length of the time series of annual precipitation and the magnitude of the external forcing required to separate the external trend from the natural variability with statistical confidence. We examined trends in annual, Boreal summer and Boreal winter precipitation using the Mann-Kendall test. For example, in the case of annual precipitation we found that an increase of at least 20% over a period of 30 years (for a monotonically increasing external forcing of 20%) was the minimum needed for the result to be statistically significant at the p<0.1 level. The number of years required to statistically decouple external from natural forcings increases by a factor of 3 or 4 when the external forcing is applied non-montonically, and also when the results are considered by seasonCorporate name:
Met ÉireannDublin Institute of Technology. Department of Physics
Publisher:
Met ÉireannType of material:
reportCollections
Availability:
Full text availableKeywords:
Meteorology, Precipitation, Temporal variability, External forcings, Climate change, Mann-Kendal, Trend analysisISSN:
1393905xMetadata
Show full item recordLicences: