Analysis of Mediterranean Climate Data: Measured
and Modelled
1. Palutikof
1 Introduction
There are four principal sources of meteorological information:
1. Surface observations of present-day conditions from 'traditional' instrument
arrays:
• raw station values
• gridded observations
2. Remotely-sensed observations, principally from satellites,
3. Proxy data sets from tree rings, sediments etc., important for the study of past
climates,
4. Modelled data:
• climate model output, to provide information on potential future conditions,
and
• reanalyses, from forecasting models and observations, to provide information
on the recent past.
This paper concentrates on just two of these: instrumental observations (1) and
output from models (4).
Climate data is normally used to try and answer four fundamental questions:
• To understand trends in observations
• To understand the underlying causes of the observed trends
• To understand what may happen in the future
• To understand the relationship between what is happening now, and what may
happen in the future.
2 Understanding Trends in Observations
By compositing data from a number of stations, it is often possible to obtain a clear
signal of trend in the data, which would not be apparent by examining time series
from individual stations. Because the mean and standard deviations of individual
records differ, it is necessary to standardize prior to compositing. The formula for
and Modelled
1. Palutikof
1 Introduction
There are four principal sources of meteorological information:
1. Surface observations of present-day conditions from 'traditional' instrument
arrays:
• raw station values
• gridded observations
2. Remotely-sensed observations, principally from satellites,
3. Proxy data sets from tree rings, sediments etc., important for the study of past
climates,
4. Modelled data:
• climate model output, to provide information on potential future conditions,
and
• reanalyses, from forecasting models and observations, to provide information
on the recent past.
This paper concentrates on just two of these: instrumental observations (1) and
output from models (4).
Climate data is normally used to try and answer four fundamental questions:
• To understand trends in observations
• To understand the underlying causes of the observed trends
• To understand what may happen in the future
• To understand the relationship between what is happening now, and what may
happen in the future.
2 Understanding Trends in Observations
By compositing data from a number of stations, it is often possible to obtain a clear
signal of trend in the data, which would not be apparent by examining time series
from individual stations. Because the mean and standard deviations of individual
records differ, it is necessary to standardize prior to compositing. The formula for