linked in logo | IS-ENES | Contact

Exploring climate model data

News! Checkout Climate4Impact v2, with new Search interface and Jupyter notebooks and become alpha user! Thanks for your feedback!

Bias correction methods

Climate models exhibit systematic errors (biases) in their output. These errors can be due, among others, to:

  • Limited spatial resolution (horizontal and vertical)
  • Simplified physics and thermodynamic processes
  • Numerical schemes
  • Incomplete knowledge of climate system processes

Such errors can and generally should be corrected for, before using climate model data in impact studies. The main assumptions of bias correction methods are:

  • Quality of the observations database limits the quality of the correction.
  • It is assumed that the bias behaviour of the model does not change with time.
  • Limitation: Temporal errors of major circulation systems can not be corrected.

To correct these biases, several methods exist, such as:

  • Delta change approach
  • Multiple linear regression
  • Analogue methods
  • Local intensity scaling
  • Quantile mapping

Currently, for the correction of climate data that are to serve as input data for impact models (e.g. hydrological or crop models) the Quantile mapping approach or some variant of it is perhaps the most widely accepted. All these methods can also be applied to seasonal forecasts, with the provision that biases are not only a function of time-of-year, but also a function of leadtime.

As such, downscaling methods also act to bias correct, since they increase the resolution and thus reduce the bias due to the spatial resolution:

Beware that the concept of biascorrection is not undisputed (see the Ehret (2012) paper below). Apart from the contentious issues surrounding bias correction, some researchers would argue that bias correction is distinct from using change factor ("delta change") methods (e.g. Hawkins et al., 2013, Agric. For. Meteorol., 170, 19-31). Even though they may produce identical results in certain idealized situations, bias correction would not normally be expected to replicate the baseline climate perfectly, so cannot offer a faithful substitute (for all variables and time resolutions) for real observations to represent the present climate. That faithful representation is sometimes required in impact applications. on the other hand, change factors apply the baseline climate but cannot normally be expected to provide information on future changes in high frequency variability. that information can be critical for some other impact applications (e.g. estimates of peak discharge in hydrological catchments).


Ehret, U., E. Zehe, V. Wulfmeyer, K. Warrach-Sagi, and J. Liebert, 2012. Should we apply bias correction to global and regional climate model data?. Hydrol. Earth Syst. Sci. Discuss., 9, 5355–5387, 2012

Haerter, J. O., S. Hagemann, C. Moseley, and C. Piani, 2011. Climate model bias correction and the role of timescales. Hydrol. Earth Syst. Sci., 15, 1065–1079, 2011

Hagemann, S., Chen, C., Haerter, J. O., Heinke, J., Gerten, D., and Piani, C.: Impact of a statistical bias correction on the projected hydrological changes obtained from three GCMs and two hydrology models, J. Hydrometeorol., 12, 556–578, doi:10.1175/2011jhm1336.1, 2011.

Hawkins E, Osborne T M, Ho C K and Challinor A J 2013 Calibration and bias correction of climate projections for crop modelling: an idealised case study over Europe Agric. For. Meteorol. 170 19–31



The ENES3 project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 824084.