Error message

  • Notice: Undefined offset: 1 in drupal_settings_initialize() (line 789 of /var/www/html/includes/bootstrap.inc).
  • User warning: The following module is missing from the file system: ckeditor. For information about how to fix this, see the documentation page. in _drupal_trigger_error_with_delayed_logging() (line 1143 of /var/www/html/includes/bootstrap.inc).
  • User warning: The following module is missing from the file system: custom_breadcrumbs. For information about how to fix this, see the documentation page. in _drupal_trigger_error_with_delayed_logging() (line 1143 of /var/www/html/includes/bootstrap.inc).
  • User warning: The following module is missing from the file system: custom_breadcrumbs_identifiers. For information about how to fix this, see the documentation page. in _drupal_trigger_error_with_delayed_logging() (line 1143 of /var/www/html/includes/bootstrap.inc).
  • User warning: The following module is missing from the file system: fancybox. For information about how to fix this, see the documentation page. in _drupal_trigger_error_with_delayed_logging() (line 1143 of /var/www/html/includes/bootstrap.inc).
  • User warning: The following module is missing from the file system: jquery_update. For information about how to fix this, see the documentation page. in _drupal_trigger_error_with_delayed_logging() (line 1143 of /var/www/html/includes/bootstrap.inc).
  • User warning: The following module is missing from the file system: lexicon. For information about how to fix this, see the documentation page. in _drupal_trigger_error_with_delayed_logging() (line 1143 of /var/www/html/includes/bootstrap.inc).
  • User warning: The following module is missing from the file system: libraries. For information about how to fix this, see the documentation page. in _drupal_trigger_error_with_delayed_logging() (line 1143 of /var/www/html/includes/bootstrap.inc).
  • User warning: The following module is missing from the file system: taxonomy_sync. For information about how to fix this, see the documentation page. in _drupal_trigger_error_with_delayed_logging() (line 1143 of /var/www/html/includes/bootstrap.inc).

Spatial and temporal resolution

Data from a global climate model (GCM) usually have a spatial resolution in the range 100–300 km, and the temporal resolution of the output varies. Depending on the variable it usually ranges from 6-hourly data to monthly means (or other statistics). For a regional climate model (RCM) the resolution is typically 10–50 km, and the temporal resolution is often 3-hourly (sometimes even higher resolution is saved) up to monthly values. For some variables also the highest and/or lowest value of any single timestep during a day is stored. Common examples of this is daily maximum and minimum 2-metre temperature, and daily maximum wind speed at 10 metres.

For any given spatial and temporal resolution the numerical schemes used to solve the basic differential equations put some limitations. Therefore it is advisable to not analyse individual gridcells at individual time-steps. Instead it is recommended to average the data, either over several neighbouring gridcells for one time-step or analyse data for one gridcell averaged over several timesteps, or both.

Model data vs. observations

There is a substantial and systematic difference in spatial scale and representativity between gridcell data and meteorological measurements. Climate model data is in the form of gridcell averages (essentially one number per gridcell, or divided per land surface tile), where the area of a GCM gridcell is in the range 10000–90000 km2 (100–300 km resolution) and an RCM gridcell 100–2500 km2 (10–50 km resolution). Meteorological measurements are essentially point measurements that are strictly representative for a small area surrounding the measurement site. Depending on the spatial variability of the variable measured the representative area is more or less extended. For example, atmospheric pressure reduced to sea level is a spatially smooth variable and measurement taken at one point is representative for a relatively large surrounding area. Convective precipitation, on the other hand, is highly variable both in time and space. Thus, measuments using a standard raingauge having an orifice of 200 cm2 is representative only for a very small surrounding area. If a severe local shower hits a raingauge a large amount is recorded, but if the shower just misses the raingauge by a few kilometres nothing is recorded. In a climate model such small-scale variations have to be averaged over the whole gridcell. However, if the precipitation is produced by a large-scale frontal system the measurements are representative for a larger area because frontal precipitation is generally more spatial homogeneous. Translating this kind of spatial variability to the uniform model grid results in that the climate model data will smooth local extreme values more than the less extreme values that are closer to the large scale average conditions.

Some variables are more sensitive to the specific surrounding of the measurement site than others. For example, the 10-metre wind measured at a station is not only influenced by individual obstacles in the immediate surrounding area but also the roughness of the vegetation and landscape many kilometres upwind.

The need to somehow reconcile these differences to the extend possible is the driving force behind downscaling and bias correction methods. It is also important to keep in mind whether the meteorological observation data are station data (i.e. point data as discussed above) or whether they have been transformed to gridded observations data through some interpolation method.

Related links