Dear ISIMIP community,
In the future, we are going to update you once a week on issues regarding ISIMIP input and output data (if there are any). They can also be accessed on our website in the Input Data changelog and Output Data changelog, respectively.
Please note the following caveats for output data:
- ISIMIP2b: all CLM4.5 runs except picontrol retracted (part 1)
Due to a major bug in the simulation configuration all runs except driven by piControl climate were retracted and tagged as do not use.
When branching the historical runs from the piControl ones in 1861, the model was told to switch from constant to transient CO2. But instead the model assumes no CO2, leading to a sudden and complete shutdown of the carbon cycle in 1861 (everything dies). piControl runs are not affected, while historical and rcpXX runs are.
All variables in all sectors will be replaced in the upcoming months. Please watch out here for updates on this issue.
- ISIMIP2a Biomes: CARAIB varsoc runs using Princeton forcing replaced
In addition to this entry in October 2016, the CARAIB runs using the Princeton forcing data for the varsoc scenario were also replaced.
- ISIMIP2b Biomes: VISIT runs replaced
All simulations from VISIT model have been rerun. This activity mainly included a new pre-processing of land use data. The responsible modeler considers the differences to be very small, yet we keep the old version in a hidden folder at DKRZ. We will make the replaced data available to you on request.
In addition, results for experiments IV to VII became available.
Additionally, there are minor updates to the input data:
- ISIMIP2a/b: EWEMBI dataset extended to 2016
The EWEMBI observational climate dataset, which previously ran until the year 2013, has been extended to the year 2016. It will be used to drive the new "histclim" runs in the ISIMIP2a protocol, as agreed at the Strategy Group Meeting in September 2018.
- ISIMIP2b extended: RCP8.5 climate forcing data moved to InputData
In accordance with the agreed ISIMIP2b protocol extension (will be incorporated into the protocol shortly), RCP8.5 runs are included as part of the protocol and have therefore been moved from "SecondaryInputData" to "InputData".
We are currently still working on implementing the protocol updates agreed at the Strategy Group Meeting and will communicate them soon.
Best wishes from Potsdam,