The IPCC was set up in 1988 by the World Meteorological Organization (WMO) and the United Nations Environment Programme (UNEP).  The role of the IPCC is to assess on a comprehensive, objective, open and transparent basis the scientific, technical and socio-economic information relevant to understanding the basis of risk of human-induced climate change, its potential impacts and options for adaptation and mitigation.

The IPCC released its Fourth Assessment Report in 2007 and this describes the current understanding of the climate system and provides estimates of its projected future changes and the associated uncertainties.  The IPCC developed and approved a range of emissions scenarios, published in the IPCC Special Report on Emission Scenarios (SRES).  The SRES scenarios replace earlier emission scenarios and are based on four descriptions on how the world may develop over the next century.

The UKCIP 2002 (or UKCIP02) scenarios are the currently available description of UK future climate based on the four different emission scenarios recently published by the IPCC SRES.  The UKCIP02 scenarios represent an advance in the description of future UK climate when compared to the scenarios published by UKCIP in 1998.  The 2002 scenarios were developed using the latest climate models from the Met Office’s Hadley Centre for Climate Prediction and Research, namely HadCM3, HadAM3H and HadRM3.  These models, which are run on super-computers, are more sophisticated than the HadCM2 model which was used to produce the UKCIP98 scaenarios. 

The UKCIP scenarios describe four alternative future climates for the UK, based on the following categorisation of greenhouse gas emissions:

• Low Emissions
• Medium-Low Emissions
• Medium-High Emissions
• High Emissions

The UKCIP02 data is available as a 50km by 50km grid over the whole of the UK, compared to just four grid boxes in UKCIP98, and so it provides far greater spatial resolution. 

The more sophisticated computing approach also allows better analysis of extreme daily weather events.  Compared to the 1998 scenarios the 2002 scenarios show for the UK:

• Slightly greater rates of warming;
• Slightly lower rates of sea level rise; and
• Drier summers.

The various emission scenarios were used by UKCIP to produce the average weather statistics over a thirty year period for three future time epochs, namely: 2011 to 2040 (labelled the 2020s); 2041 to 2070 (the 2050s); and 2071 to 2100 (the 2080s).  These results are produced to provide changes in comparison to the 1961 to 1990 (the 1970s) climate, which is used as the ‘baseline’. 

To provide an indication of the extent of climate change anticipated by the 2050s across the Region in the present study, use of the Medium-High scenario has been recommended by experts at the Met Office and its Hadley Centre because it represents a reasonable estimate of future changes over this timescale.

This Environment Agency Rainfall and Weather Impacts Generator (EARWIG) was developed by Newcastle University for the Environment Agency’s use in climate impact assessments and agricultural and water system management.  The generator produces internally consistent series of meteorological variables including:

• Temperature;
• Rainfall;
• Humidity;
• Wind; and
• Sunshine.

The system produces results at a daily time resolution, using two stochastic models in series.  The first, for rainfall, produces and outputs data which is then used as input to a second model which generates the other variables that are dependent on rainfall.  The series are intended for single sites defined nationally at 5km resolution, but can be generated to be representative across catchments.  Scenarios can be generated for the UKCIP baseline control period (1961-1990) based on observed data, as well as for the three future UKCIP time horizons (2020s, 2050s and 2080s).

To assess the changes in daily rainfall extremes which are most relevant for analyses of fluvial and small and medium catchments, the following measures have been considered:

• 50% Annual Exceedance Probability (AEP) - the daily rainfall which has a 50% chance of being exceeded in any given year. This is equivalent to the 2-year return period event, and is referred to in the Flood Estimation Handbook (FEH) as the Rmed, or median annual maximum rainfall, representing a “typical” extreme rainfall event;

• 5% Annual Exceedance Probability (AEP) - the daily rainfall which has a 5% chance of being exceeded in any given year. This is equivalent to the 20-year return period event and represents a major Srainfall event.

Rainfall events for durations of more than one day are most relevant for assessments of fluvial flooding in larger catchments, and for cases where antecedent rainfall may lead to saturation and subsequent flooding from less extreme rainfall events.   

Projections from climate models of changes in wind speed, whether looking at average speeds or extreme speeds, are less reliable than projections of changes in temperature or precipitation. This is not to say that the projections are wrong, just that there is greater uncertainty in the projections of changes in winds. The climate model projections of changes in temperature have the least uncertainty, and the projections of changes in precipitation are the next least uncertain.

One of the reasons for the large uncertainty in projections of changes in winds is the large natural variability of the wind on all timescales.  In the UKCIP02 report, where Met Office climate model projections of changes in mean wind speeds are shown, large parts of the UK are projected to see changes which are within the range of “natural” variability, defined as one standard deviation of the model-simulated 30-year average climate. Mean wind speeds are highest and have the greatest variability in winter, especially in the north of the UK, which is partly why the changes projections by the climate model are mostly less than the range of natural variability in the north of the UK in winter.  EARWIG produces outputs values for wind, although advice from the author of the software is to handle the output in a conservative manner.

In the present study, the sole model used for the projection of extremes of temperature is EARWIG, as UKCIP only provides monthly averaged results which cannot be used to determine severe temperature.  Since EARWIG produces daily results it allows for the study of changes in extreme temperature. 

Changes in severe temperatures are more difficult to project than changes in daily average values, due to the small sample size of such events and the random nature of their occurrence.  In accordance with the recommendations from the European Union-funded research project called STARDEX (Statistical and Regional Dynamical Downscaling of Extremes) undertaken by the Climate Research Unit of the University of East Anglian, a percentile approach has been adopted to represent low and high severe values.

Results are presented for the for the 5th and 10th percentile (cold) day per season and the 90th and 95th percentile (hot) day per season, in accordance with the STARDEX indices for extreme events.  For example, the 90th percentile temperature is the 10th hottest day that could be expected per season, and similarly the 10th percentile temperature is the 10th coldest day per season.

In light of the 2003 heatwave, the Met Office launched the Heat Health Watch system which operates in England and Wales from the 1st June to the 15th September each year.  It comprises four levels of response based upon threshold maximum daytime and minimum night-time temperatures, with the thresholds varying from region to region (Table 1). 

Table 1: Regional Threshold Temperatures from the Met Office’s Heat Health Watch Warning System

The different response levels in relation to the thresholds are:

• Level 1 – Awareness – the minimum state of vigilance during the summer.  The majority of the time the risk of a heatwave will be less than 50%.  However, when the risk exceeds 50% this will be indicated by ‘Level 1 – Awareness – Increased Risk’.

• Level 2 – Alert – triggered as soon as the risk is 80% or above for threshold temperatures being reached in one or more regions on at least two consecutive days and the intervening night.

• Level 3 – Heatwave – triggered as soon as the Met Office confirms threshold temperatures will be reached in one or more regions.

• Level 4 – Emergency – reached when a heatwave is so severe and/or prolonged that its effects extend outside the health and social care system. 

Using the thresholds from the Heat Health Watch system, the output from EARWIG was analysed for each representative location across the region to asses any changes in the frequency of heatwaves between the baseline period and the 2050s during the simulation of 100 typical years.  The Heat Health Watch system warns that the threshold temperatures could have significant effect on health if reached on at least two consecutive days and the intervening night.  With this in mind, the data was analysed for the occurrence of temperatures above the threshold for 2 or more consecutive days.