Differences

This shows you the differences between two versions of the page.

--- clipc:evaluation [2015-07-14 12:41:57]
andreasd
+++ clipc:evaluation [2022-05-31 09:29:31] (current)
@@ Line 11: / Line 11: @@
 {{:clipc:taylor_tasmin.png?direct&600|}}
-Annual and seasonal biases of the raw and bias-corrected for the individual RCM runs are shown here:\\
+Overall and seasonal biases of the raw and bias-corrected for the individual RCM runs are shown here:\\
 {{:clipc:bias_pr_EUR-44.pdf|}}\\
 {{:clipc:bias_tas_EUR-44.pdf|}}\\
@@ Line 28: / Line 28: @@
 {{:clipc:taylor_tasmin_EUR11.png?direct&600|}}
-Annual and seasonal biases of the raw and bias-corrected for the individual RCM runs are shown here:\\
+Overall and seasonal biases of the raw and bias-corrected for the individual RCM runs are shown here:\\
 {{:clipc:bias_pr_EUR-11.pdf|}}\\
 {{:clipc:bias_tas_EUR-11.pdf|}}\\
@@ Line 35: / Line 35: @@
 ===== Summary =====
-As expected, the bias-correction method reduces (but does not remove) the seasonal biases and removes the annual biases (almost) completely for all variables.
+As expected, the bias-correction method reduces (but does not remove) the seasonal biases and removes the overall biases almost completely for all variables.
+The remaining annual biases are due to the use of "only" 101 quantiles to estimate the empirical cumulative distribution functions within qmap.
+The additionally remaining seasonal biases are due to the calibration process which is carried out without splitting into seasons or months.
 ----
- //[[andreas.dobler@met.no|Andreas Dobler]] 2015/07/10 15:03//
+===== Impact on climate change signal =====
+Within the CHASE-PL project, the same bias-correction method was used but with a different observational data set (for Poland only) and splitting into seasons for calibration.\\
+{{:clipc:cc_pr_POL-05.pdf|Here}} you can see the impact of the method on the mean climate change signal of precipitation for the far (2071-2100) and near (2051-2070) future over Poland for the single models and the ensemble mean.
+----
+//[[andreas.dobler@met.no|Andreas Dobler]] 2016/04/28 15:00//