Where is the current climate best matched to the UK climate of the future?

 

The identification of locations which currently experience climates similar to those predicted for Britain has the potential to provide guidance on the likely effect of climate change on tree growth, and whether a given species is likely to prove suitable for hardwood timber production in the climate of the future. In addition, this information will identify those areas from which seed material could be collected for provenance trials focussing on climate change adaptation. However, some caution should be applied to the climate mapping analysis presented here, as it is highly unlikely that future climates predicted for specific locations can be matched exactly to the full range of variables of existing climates, including daylength and solar radiation input.

Data sources and analysis

To maintain data consistency for current and future coverage and ensure the availability of climate data across Europe, a global gridded data-set at a resolution of ten minutes was used. The interpolated data-set is based on the climate records from between 12783 (temperature) and 27075 (precipitation) weather stations at which data were available for the period 1961-1990. The data-set has been developed and made available by the Climatic Research Unit of the University of East Anglia (CRU: New et al., 2002). The predicted future climate is based upon the UK Climate Impacts Programme 2002 scenarios (UKCIP02: Hulme et al., 2002), with the ‘Low’ and ‘High’ scenarios for the 2050s and 2080s covered. The UKCIP02 Low scenario corresponds to the IPCC SRES B2 scenario, and the UKCIP02 High, to the IPCC SRES A1F1 scenario (IPCC, 2000; Hulme et al., 2002).

Climate matching analysis was carried out for five of the twenty sites comprising the UK Intensive Forest Health Monitoring network. The five sites (Brechfa: south Wales; Kelty: central lowland Scotland; Rannoch: Scottish uplands; Alice Holt: southern England; Thetford: East Anglia) were chosen to provide an indication of the range of future climates that may be experienced across the range of current climates in which broadleaf species are currently grown for timber.

For each site, the UKCIP02 climate change scenarios, taken from the 50 km gridded data-set, were applied to the CRU monthly mean (1961-90) climatology for the grid-square for mean temperature, precipitation and diurnal temperature range (DTR). The location of the gridsquare currently best matched to the predicted climate of the future was identified by minimising the sum of squares of differences between the current (1961-1990) and predicted future monthly mean climate variables, creating a climatic difference index. The climate variables were weighted according to the range of monthly values under the current climate. Temperature and DTR were both weighted on the basis of mean temperature, as both contribute equally to describing the climatology of a given location. For precipitation, values were natural log transformed to normalise the distribution of annual rainfall totals across the data-set.

Results of climate matching analysis

The location of current climates best matched to the predicted climate for the five Intensive Forest Health Monitoring Sites have been identified:

View map - Brechfa ^(11K)

View map - Kelty ^(11K)

View map - Rannoch ^(11K)

View map - Alice Holt ^(11K)

View map - Thetford ^(11K)

Under the 2050s Low scenario, the climate of Brechfa is predicted to be similar to that of much of the south west peninsula and south west of Ireland, reflecting the moist climate of south Wales. However, under the 2050s High scenario, this area is predicted to move southwards into areas of Cornwall experiencing lower rainfall, and also the Brittany peninsula.

Under the most extreme of the scenarios (2080s High), the combination of warm winters, and very high winter rainfall (~200 mm per month) compared to summer rainfall (~50 mm per month) results in the climate being best matched to that of northern Spain at relatively high elevation (~500 m). For Kelty, implications of predicted climate change appear less significant than for Brechfa, with the Welsh borders (2050s Low) and southern England (2050s High) currently having similar climates to those predicted. However, under the 2080s High scenario, the climate is best matched to that of southern Brittany.

For Rannoch, the climate of the future is predicted to match that currently experienced in areas of Wales, and under the most extreme scenarios, upland areas of the English southwest peninsula.

The climate of Alice Holt is predicted to be similar to that of southern Brittany and the northern Loire under both scenarios by the 2050s, while the predicted low summer rainfall and warm winters result in the climate being best matched to that of central southern Italy, Sardinia and northern Greece under the 2080s High scenario, but again at high elevation.

The consequences for Thetford appear even more extreme as a result of the already low rainfall (< 600 mm per year) with a climate similar to that of central southern Italy and northern Greece predicted by the 2050s for the High emissions scenario.

However, the interpretation given above should be viewed with extreme caution as energy balance is significantly different at lower latitudes, and a number of important variables including rainfall intensity and relative humidity, which would be expected to differ, have not been included in the climate matching analysis.

Potential evapotranspiration (PET) and maximum soil moisture deficit (SMD) provide a test of how well matched a climate may be, since PET is dependent on a more extensive range of climate variables including humidity, solar radiation input and wind speed, as well as temperature and rainfall. PET has been calculated for the four UKCIP02 scenarios (2050s Low and High, 2080s Low and High) for two contrasting sites (Alice Holt and Kelty), and these values are compared with PET and SMD (cumulative PET less precipitation) calculated for the best matched gridsquare under each of the scenarios.

It is apparent that PET of the matched gridsquares is higher than the predicted future climate of Alice Holt and, in general, for matched sites at lower latitudes. However, when maximum SMD is calculated the fit is improved for the lower latitude sites. The generally good agreement for both PET and SMD indicates that this climate matching exercise has identified regions with broadly similar climates across a wide range of climatic variables.