Basic LUCID experiments
Seven modelling groups participated to the first set of LUCID simulations. They all conducted two series of two experiments using prescribed interannually and seasonally varying SST and sea ice extent using data from the C20C project (HadISST1.1, ftp://www.iges.org/pub/kinter/c20c/HadISST/):
- Present-day simulations, with all greenhouse gases, land cover and sea surface temperatures (SSTs) prescribed at their present-day values. The land cover is prescribed using a map reflecting 1992 and the period 1972-2002 is simulated. Five independent realizations were run by each group (experiment PD).
- As PD but with a land cover map reflecting 1870 conditions (experiment PDv).
- Pre-industrial simulations, with all greenhouse gases, land cover and sea surface temperatures (SSTs) prescribed at their pre-industrial values (~1870). The land cover is prescribed using a map reflecting 1870 and the period 1870-1900 is simulated. Five independent realizations were run by each group (experiment PI).
- As PI but with a land cover map reflecting 1992 conditions (experiment PIv).
Main findings to date
A series of papers analysing this first set of experiments concluded:
- There is a clear and statistically significant impact of historical land-use induced land-cover changes (LULCC) on the simulated summer latent heat flux and near-surface temperature over the regions of LULCC (Pitman et al. 2009). However, the direction of regional change differs, thereby highlighting a lack of consistency among the models. Moreover, LULCC in those simulations and at that time period do not trigger remote changes to climate that are common to multiple models. We suggested various reasons for the observed discrepancies, without being able to attribute the changes to one or the other at that stage.
- LULCC indeed leads to a systematic increase in surface albedo at all seasons, which induces a decrease in a) available energy (QT, computed as the sum of absorbed incident solar energy and incident atmospheric infra-red radiation), and b) turbulent fluxes (QA ; computed as the sum of latent and sensible heat fluxes). Moreover, all models show, for pre-industrial conditions, quite similar amount of available energy being used for turbulent fluxes (i.e. similar QT/QA ratio) at all seasons, and this ratio is always decreased in response to LULCC (i.e. in response to deforestation) at all seasons. All those changes are proportional to the amount of deforestation each individual models have undergone (de Noblet-Ducoudré et al. 2012). This proportionality is quite interesting because it means that even though the land-surface and vegetation models have been developed independently, there is a large part of their response to LULCC that is qualitatively and quantitatively comparable.
- However, although a) the amount of available energy used for turbulent fluxes is consistent from one model to the other, and b) its changes in response to LULCC almost linearly depend on the amount of trees removed, there is no consistency among the various models regarding how a) energy is further partitioned into latent (LE) and sensible (H) heat fluxes at a specific time period, b) H and LE change from one season to another in response to the land-use induced land-cover change. In some models, deforestation leads to a decrease in both fluxes in response to the decrease in QA and QT, while in others H increases and LE decreases (or reversely). This is dependent on how these processes are represented in a land surface model (Boisier et al. 2012). We are not able, at this stage, to classify the various models but we strongly suggest that there is a need to revisit the way land-surface models are evaluated before being used for cost-effective global climate simulations.
- The dispersion among the models’ response to LULCC is quite larger than the dispersion that results from the change in CO2SST. We therefore expect more dispersion among the regional climate changes that will be simulated by models participating to the CMIP5 exercise since they will all account for LULCC changes, without having time to have their model go through the validation exercise we have suggested above.
If any researchers seek access to the simulations conducted as part of the basic LUCID experiments, please contact Nathalie de Noblet at email@example.com. These data are freely available for research purposes.