Transcript Slide 1
Scientific benefits from undertaking data rescue activities: some examples of what can be achieved with long records Phil Jones Climatic Research Unit University of East Anglia Norwich, UK Summary of scientific benefits • Longer records for analysis and assessment of trends • Able to place recent records in a much longer context • Provide longer records for assessing impacts of • • • climate change Provide longer records for the calibration of natural and documentary proxies Provide longer records for Reanalysis* extensions further back in time Ensures NMSs are aware of early data in their country and their possible shortcomings *model-based consistent assimilation of the past – so internally Placing in a longer context • Some long European records • European summer of 2003 • Central England temperature extremes since 1772 Long European regional temperatures Heat waves in Central Europe (35-50ºN, 0-20ºE) in JJA Extreme Heat Wave Summer 2003 Europe: Trend plus variability? Max temps in Basle (1961-90 and 2003) Counts of daily temperature extremes in the CET Series Defined as > 90%ile (warm) and < 10%ile (cold) – relative to the period 1881-1910 and to the time of year Annual number of warm days/cold days Could have used 196190 as the base period. Changing the base period just changes the level Longer records for assessment of proxy evidence • Documentary Records (N. Sweden – where the • • instrumental record has been extended back from the 1860s to 1800, Klingbjer and Moberg, 2003) Natural Proxy Records (N. Fennoscandian trees and Greenland examples – the latter extending the instrumental record from the 1870s to the 1780s) Both extend the instrumental record beyond that readily available from the local NMS Klingbjer and Moberg (2003) Int. J. Climatol. 23, 1465-1494. N. Fennoscandia (Haparanda, Klingbjer and Moberg, 2003) Comparison of instrumental, documentary and natural proxy data SW Greenland temperatures Seasonal temperatures for SW Greenland From Vinther et al (2006) JGR Running correlations (30-year window) between Greenland ice cores (winter oxygen isotope values) and SW Greenland winter temperatures Assessment of changes in extremes • Require long series of daily data (earlier CET example) • Consistent measures of extremes • Extremes are how the public and governments perceive climate change Warm nights are increasing; cold nights decreasing 1979-2003 1951-1978 1901-1950 fewer more fewer 10th (left) and 90th (right) percentiles more Frequency of occurrence of cold or warm temperatures for 202 global stations with at least 80% complete data between 1901 and 2003 for 3 time periods: 1901 to 1950 (black), 1951 to 1978 (green) and 1979 to 2003 (orange). Longer records to look at the influence of the circulation on surface temperature and precipitation series • Paris – daily pressure (MSLP) data can be digitized back to 1670s with only a few missing years • London – back to the 1690s • Together these two sites produce a useful winter measure of westerly flow (Paris minus London ≈ NAO) The Royal Society's Meteorological Record (1774 - 1842) - London • Appendix to Phil.Trans publication • Twice daily observations • Very nearly complete, but gap from 1782 - 1786 • Extracted Internal/External Temperature & Pressure [SOURCE: Philosophical Transactions of the Royal Society of London, 1775] James Jurin's Weather Diary (1728 - 1750) London • Once daily observation of temperature, pressure & wind direction • Observation time included Joseph de L'Isle's Weather Diary (1747 1760) - Paris • Three or four observations per day • Temperature, pressure, state of the sky • Mostly readings from 2+ barometers • Temperature scale problematic Running 30-year correlation between winter Paris minus London Pressure Index and several temperature series and one precipitation series Correlations generally consistent except for England and Wales precipitation and NH temperatures north of 20ºN From Jones et al. (2003) in AGU book on the NAO by Hurrell et al. Conclusions • Longer records than currently available to • • each NMS are generally available across Europe and the Mediterranean Records need finding, digitizing and then assessing for homogeneity Records often located in libraries/archives, sometimes in other countries. They may take some finding, but they generally exist