The Observed Climatology - Climate Dynamics - page 4

THE OBSERVED CLIMATOLOGY
This chapter forms a concise atlas of the climate system. An overview of the
system is presented using the variables and terminology commonly used to
characterize climate. These terms are referenced in subsequent chapters as a
deeper understanding of climate processes is developed.
Some features of the climate system are known accurately, while others
are known only approximately. Climate observations can be limited by insuf-
ficient spatial and temporal resolution, inadequate global coverage, or a lack
of long-term records. Precipitation observations are a good example. Because
of the high variability of precipitation over a wide range of space and time
scales, the observing requirements for establishing a precipitation climatol-
ogy are demanding. Global measurements of precipitation or, more accurately,
measurements of radiative fluxes that can be translated into rainfall rates have
been available only since the beginning of the satellite era in the early 1970s.
Pre-satellite coverage over vast regions of the oceans was particularly sparse,
especially in regions where ships rarely traveled. Establishing a climatol-
ogy for other variables, such as evaporation and soil moisture, is even more
challenging.
Many of the figures in this text were drawn using reanalysis products,
which combine simulations using state-of-the-art numerical models with ob-
servations. To produce a reanalysis climatology, computer models are run to
stimulate many decades, with observed fields incorporated into the model at
the time they were observed. This process is called four-dimensional data as-
similation , for the three spatial dimensions plus time. (Data assimilation is also
used routinely in generating weather forecasts.) Thus, the reanalysis product is
not pure observations but a blend of observations and computer model output.
Reanalysis values of variables that are assimilated—for example, winds and
temperatures—are accurate. Other variables, however, are model-dependent
output and may not be as reliable. Sometimes, as in the case of evaporation,
the reanalysis product is the best information available with global coverage.
For other variables, ground-based and satellite observations, if available, are
preferred to the reanalysis product.
Maps of climate variables use latitude and longitude as coordinates with an
equidistant cylindrical projection. Keep in mind that the area of middle and
high latitudes is falsely large in this projection. In reality, half the surface area
of the globe lies between 30°N and 30ºS latitude, whereas in the figures this
region occupies only one-third of the area. Vertical profiles of climate variables
are also shown, averaged globally or over certain regions using area weight-
ing to correctly account for the decreasing distance between meridians (lines
of constant longitude) away from the equator. Another useful way to display