Evidence and Data

Proxy Data

We can find a lot of data in historical written records, such as diaries, ships' logs and personal letters. In addition agricultural events, such as the dates of sowing and harvesting give a good idea of weather conditions during that time. Periods of exceptionally good and, particularly, disastrous harvests are usually known. These can provide time frames and a rough idea as to climatic conditions, but such information is qualitative and, therefore, does not give actual figures.

Tree-ring data has been widely used, but in reality is even more difficult to assess, since variations in tree rings are as much due to changes in rainfall and the immediate surroundings as to changes in temperature.

More recently pictures showing how far glaciers have retreated over the past 50-100 years can be very compelling. Although as first glance it seems clear there must have been a rise in temperature at their level, is that the only reason for the retreat? The snows of Kilimanjaro are disappearing rapidly, but that could be due to a lack of snowfall rather than rising temperatures.

Despite the somewhat subjective nature of this type of data, when these types of information are gathered together and related with each other, a better idea of the past climate can be made. As an example, the study of historical records and volcanology has shown that there was a particularly violent eruption in Iceland during 1783. This eruption released both dust and poisonous gases into the atmosphere that not only caused a lot of deaths in Iceland, but are believed to be linked to poor harvests in Europe for some years afterwards. Measurements of temperature were also being made in a few places and these show that summer temperatures were lower than in previous years. Hence, there are interconnections between volcanic activity, temperature and harvests.

Instrumental Data

Over the past few hundred years there have been climatic records from various types of instruments. Until about 300 years ago, these records tended to be short and incomplete, probably because few people had the time or inclination to keep writing them down and some will have been lost or destroyed. Additionally, the instruments were probably too expensive for most people, they would have been cumbersome and not particularly accurate. All the same, such records are useful as a form of calibration for proxy data that can be dated to the same time.

As scientific enquiry grew, observatories were set up that were intended mainly to observe the heavens, but included an increasing number of instruments for measuring and studying the weather. Reasonably accurate measurements of temperature, rainfall and atmospheric pressure have actually been possible for a long time in many areas. A variety of different scales have been used for each, but it is relatively easy to convert these individual scales to standard units.

In more recent years, techniques for looking at meteorological information have been developed that make it possible to study changes on a global scale. It is believed that this can reasonably be done with data back to the middle of the 19th century, since, by then, there have been large number of observing stations available.

What Does the Data Show?

The instrumental climatic record shows an uneven trend towards higher temperatures that seems to have accelerated during the past 20 years. It might be reasonable ot assume that, since this information comes from scientific sources, there should be little argument about the results. However, a number of questions should be asked before we decide whether this "global warming" is real.

1. Where are the observations made?
2. How are the observations made?
3. When are the observations made?
4. What instruments are being used?
5. How are the global averages calculated?

• The environment around a site is likely to change. One of the most obvious is land-use change such as housing development, but there can be more subtle changes, such as the growth or removal of vegetation. Another change may be the site itself, with small changes in the positioning of the instruments.
• Over time, observers will change and some will be better than others. The exposure of the instruments may change; either different screens are used or are not properly maintained. It should be noted that human error should decrease as an increasing number of sites are becoming fully automated.
• The times at which observations are made may change and, over time, the climate averages for a station may be calculated differently.
• Liquid-in-glass thermometers are being replaced by platinum resistance thermometers, but these have been shown to have different characteristics.
• Attempts are made to correct observations or make them more compatible with each other. The methods used involve complex statistics. Also, designated climate stations may close and/or be replaced by others.

All these factors can pose significant questions about the accuracy of figures and conclusions drawn from them.

This is not the only manifestation of global warming and it is difficult to deny that the international expansion and greater rigour of scientific enquiry over recent years means that there is less reason for doubt. As well as the retreat of glaciers, Arctic ice has become thinner over the last 30 years, currently (2006) being barely half its earlier thickness. Its expansion southwards during the winter months is also much less.

In the more temperate northern latitudes, winters are less severe than 30 years ago, with cold snaps generally being short-lived. This has been especially noticeable in places with a more maritime climate. The first signs of spring (bulbs, flowers, birds nesting) also appear earlier, there are more frequent hot spells during summer and the first frosts of autumn tend to be later.