Climate Modelling Group
School of Earth and Ocean Sciences

CFCAS Research Proposal

March 2001

1. Science:

1.1 Introduction

Coupled atmosphere-ocean general circulation general circulation models (GCMs) are frequently used to understand both past, present and future climate and climate variability. The computational expense associated with these models, however, precludes their use for undertaking extensive parameter sensitivity studies. While they must ultimately be used as the primary tool for undertaking climate projections on which policy will be based, it is important to conduct sensitivity studies, in parallel with the coupled GCM studies, using simpler models. Simple models, or models of intermediate complexity, allow one to explore the climate sensitivity associated with a particular process or component of the climate system over a wide range of parameters. In addition, they allow one to streamline the experiments that are performed using more complicated GCMs. These more idealised coupled models vary in complexity from simple one dimensional energy balance/upwelling diffusion models (Wigley and Raper 1987, 1992; Raper et al. 1996; Wigley, 1998), to zonally-averaged ocean/energy balance atmosphere models (Stocker et al. 1992; Stocker and Schmittner, 1997), to models with more sophisticated subcomponents (Fanning and Weaver, 1996; Petoukhov et al. 2000; Weaver et al., 2001).

Simple and intermediate complexity climate models are designed with a particular class of scientific questions in mind. In the development of the model, only those processes and parametrisations are included which are deemed important (or are possible to incorporate) in the quest to address the scientific questions of concern. For example Wigley (1998) used an upwelling diffusion-energy balance climate model (see Kattenberg et al. 1996) to evaluate Kyoto Protocol implications for increases in global mean temperature and sea level. While such a simple climate model relies on climate sensitivity and ice-melt parameters obtained from coupled atmosphere-ocean GCMs, it nevertheless allows for a first-order analysis. Stocker and Schmittner (1997) used a three-basin zonally-averaged ocean circulation model coupled to a simple energy-balance atmospheric model (described in Stocker et al. 1992), to undertake a systematic parameter sensitivity study of the response of the North Atlantic thermohaline circulation to both the rate of increase and equilibrium concentration of atmospheric CO2. While the actual critical thresholds that arose from this study would need verification by more complicated models, their work clearly illustrated the importance of the rate of CO2 increase on the North Atlantic thermohaline circulation, a result difficult, if not impossible, to achieve with the computationally expensive present GCMs.

The CLIMBER group at Potsdam Institute have taken the approach of building a climate model of intermediate complexity to examine climate change and variability with a sophisticated, albeit highly parametrised, atmospheric component. Their atmospheric model is based on the statistical-dynamical approach without resolving synoptic variability (Petoukhov et al. 2000; Ganopolski et al. 1999). Their three-basin, zonally-averaged ocean component is very similar to the ocean component of (Stocker et al. 1992) and they also incorporate a simple dynamic/thermodynamic ice model. The CLIMBER-2 model sacrifices resolution (10° latitudinal by 51° zonal) and complexity for computational efficiency. This model has been used to investigate both the climate of the last glacial maximum (LGM – Ganopolski et al. 1998) as well as the cause for the collapse of the conveyor in global warming experiments (Rahmstorf and Ganopolski, 1999). Fanning and Weaver (1996) developed an energy-moisture balance model coupled to an ocean GCM and a thermodynamic sea-ice model. This model has been used to undertake a number of sensitivity studies including the role of sub-grid-scale ocean mixing and flux adjustments in global warming experiments (Wiebe and Weaver, 1999; Fanning and Weaver, 1997) and steric sea level rise (Weaver and Wiebe, 1999), and the ocean's role in the LGM and Ordovician climates (Weaver et al. 1998, 2000; Poussart et al. 1999).

Figure 1: Annual mean precipitation (m/yr) top-left) model; top-right) NCEP reanalysis; bottom) zonally-averaged model (red) and NCEP(green). Taken from Weaver et al. (2000)

1.2 The UVic ESCM:

The UVic Earth System Climate Model (ESCM) consists of a 3-D ocean GCM coupled to a thermodynamic/dynamic sea ice model, an energy-moisture balance atmospheric model with dynamical feedbacks, and a thermomechanical land ice model (Weaver et al. 2000). A reduced complexity atmosphere model is used to keep the model computationally efficient. Atmospheric heat transport is parametrised through Fickian diffusion. Moisture transport is accomplished through advection (Fig. 1), and precipitation is assumed to occur when the relative humidity reaches greater than 85%. Precipitation over land is instantaneously returned to the ocean via one of 33 observed river drainage basins. Ice and snow albedo feedbacks are included in the coupled model by locally reducing the latitudinal profile of the coalbedo. The atmospheric model includes a parameterisation of water vapour/planetary longwave feedbacks, although the radiative forcing associated with changes in atmospheric CO2 is externally imposed as a reduction of the planetary long wave radiative flux. A specified lapse rate is used to reduce the surface temperature over land where there is topography. The model uses prescribed winds to obtain its present-day climatology, and a dynamical wind feedback is included that exploits a latitudinally-varying empirical relationship between atmospheric surface temperature and density. The ocean component of the coupled model is a fully nonlinear 3-D ocean GCM with a global resolution of a 3.6° (zonal) by 1.8° (meridional) and 19 vertical levels, that includes an option for a new brine-rejection parameterisation. The coupled model incorporates a dynamic/thermodynamic sea ice model (Bitz et al. 2000, Holland et al., 2000). An elastic-viscous-plastic rheology is used to represent dynamics and various options are included for the representation of sea ice thermodynamics and thickness distribution. The systematic comparison of the coupled model with observations reveals a good agreement (Weaver et al. 2001).

One of the virtues of the coupled model is that we do not need to employ explicit flux adjustments to keep the simulation of the present climate stable. It also allows us to conduct many long timescale integrations of the coupled climate system, in order to investigate climate processes through a wide range of parameter space. Thus, it is well suited for both climate and paleoclimate modelling. In Chapter 8 of the IPCC Third Assessment report the simulated LGM SSTs and SATs from the UVic ESCM and several AGCM-mixed-layer ocean models were compared with paleo proxy data. As can be seen from Fig. 2, the UVic model compares very favourably with the paleo reconstructions.

Figure 2: Annual mean tropical cooling at the last glacial maximum: Model – paleodata comparison. (Center) simulated land surface air temperature changes displayed as a function of SST changes, both averaged in the 30° latitudinal band, for all the PMIP simulations: models with prescribed CLIMAP SSTs (circles) and coupled AGCM-mixed layer ocean models (squares). Numbers refer to different models: circles, 1: LMD4, 2-5: MRI2, ECHAM3, UGAMP, LMD5, 6-7: CCSR/NIES1, LMD5, 8: GEN2. Squares : 1: LMD4,2: UGAMP, 3: GEN2, 4: GFDL, 5: HADAM2, 6: MRI2, 7: CCM1, 8: CCC2. Results from two EMIC models including a dynamical ocean model have also been displayed (diamonds): 1-UVic, 2- CLIMBER-2. The comparison with paleodata: (upper) over land is with estimates from various pollen data for altitudes below 1500m (the label "nb data" refers to the number of data points in three different regions corresponding to the temperature change estimate plotted in the abscissa); (right) the distribution of SST changes estimated from alkenones in the tropics. In this figure, model results are averaged over the whole tropical domain and not over proxy-data locations, which may bias the comparison. Taken from McAvaney et al (2001).

The UVic ESCM has not only been used to investigate various scientific questions in contemporary and paleo climate, but also as a laboratory with which to build new subcomponent models, test new parametrisations, and develop intuition and coupling experience/technology for use by the CCCma. The efficiency of the simpler atmospheric component of the ESCM allows for a wide range of parameter space to be explored through long timescale integrations. For example, a new dynamic/thermodynamic sea ice model, developed by researchers in my group (Bitz et al. 2000), has recently been incorporated into the CCCma coupled atmosphere-ocean general circulation model (as well as the NCAR Climate System Model in the US). A new parametrisation for brine rejection is now tested and available for use in the CCCma coupled model. We have also recently developed a bottom boundary layer parametrisation for flow over sills.

This proposal will build upon our demonstrated past success of addressing fundamental questions in climate and paleoclimate while at the same time having an end user for the knowledge, tools and experience we generate. Here the theme of scientific inquiry concerns the role of carbon and vegetation feedbacks in both past and present climate change.

1.3 Scientific Questions to be addressed:

The fundamental new scientific questions I wish to address are:

  1. Do multiple equilibria of the climate system exist as a consequence of ocean carbon cycle/atmosphere interactions?
  2. Do dynamic vegetation feedbacks contribute the glacial inception at 116 kyr BP?
  3. What is the relative role of changes in ocean SST, land surface moisture, and vegetation feedbacks for the changing African climate and biome at 6 kyr BP?
  4. What is the role of inorganic carbon uptake in glacial to interglacial transitions?

The scientific and technological deliverables of this project are further detailed on page 3 of form 101.

1.4 Technology needed to address the scientific questions

To address these questions, two methodological approaches will be used. In the first, my locally developed ESCM will be extended to include a dynamic global vegetation model (DGVM), a simple land surface model, as well as an inorganic ocean carbon cycle model. We choose a top-down DGVM approach, in which the relevant land-surface characteristics (vegetation fraction, leaf area index, albedo, etc.) are modelled directly (Foley et al 1996).

The Hadley Center vegetation model TRIFFID (Top-down Representation of Interactive Foliage and Flora Including Dynamics – Cox et al. 2000, 2001) defines the state of the terrestrial biosphere in terms of the soil carbon, and the structure and coverage of five plant functional types (PFT; broadleaf trees needleleaf trees, C3 grass, C4 grass, and shrub) within each model gridbox. Vegetation dynamics (areal coverage, leaf area index and canopy height of each PFT) are driven by net primary productivity, which is calculated interactively as a function of climate and CO2. Carbon fluxes for each of the vegetation types are derived using the coupled photosynthesis-stomatal conductance model developed by Cox et al (1998), which utilises existing models of leaf-level photosynthesis in C3 and C4 plants (Collatz et al 1991; Collatz et al 1992). The land-surface scheme used will be a simple, one layer model, computing the energy, moisture and snow balance at the surface as recently developed by P. Cox and K. Meisner.

As can be seen in Fig. 1, our precipitation field over the ocean bears fine agreement with the NCEP reanalysis data. Over land, differences are bigger, although we do not suffer from spectral truncation phenomena seen in the NCEP reanalysis field. We believe that the inclusion of a land surface scheme (simplified version of the Hadley Centre MOSES scheme) will dramatically improve our precipitation over land as currently any rain that falls on land instantaneously runs off to the ocean. The inclusion of a land surface scheme will allow for local recycling of water (through evaporation and subsequent precipitation) which should increase the precipitation over land, bringing it more in line with the NCEP reanalysis (e.g. see Fig. 1c).

In the case of the carbon cycle, we will closely follow the guidelines provided by the Ocean Carbon-Cycle Model Intercomparison Project Dissolved inorganic carbon (DIC) will be modelled initially as a passive tracer (i.e. there will initially be no interactive biology), with fluxes of CO2 at the air-sea interface parametrised through temperature and wind dependent bulk formulae.

The second approach uses SST, sea ice and vegetation fields from the first approach to drive the Canadian Centre for Climate Modelling and Analysis (CCCma) atmospheric general circulation model. This will allow for a more detailed resolution of the atmospheric consequences of past climate change/variability. We have had a good deal of success using SSTs from the UVic ESCM in driving the CCCma atmosphere model in the past (e.g., Yoshimori et al 2001; see section 1.5.2).

1.5 Methodology used to address the scientific questions

1.5.1 Do multiple equilibria of the climate system exist as a consequence of ocean carbon cycle/atmosphere interactions?

There is no reason why one would expect the climate system to have one equilibrium under preindustrial levels of atmospheric CO2. In fact, Manabe and Stouffer (1988) showed how the GFDL coupled model possessed two states, with and without active deep water formation in the North Atlantic, under the same external forcing. Coupled models traditionally fix the atmospheric level of carbon dioxide or allow it to change by also specifying its temporal evolution. In reality, there is a continual flux of carbon between the atmosphere and the ocean. Oceanic dissolved inorganic carbon (DIC) changes through solubility, organic matter and carbonate pumps. Initially, we will focus on the solubility pump, following the Ocean Carbon Cycle Model Intercomparison Project (

In this approach, dissolved inorganic carbon in the ocean will evolve solely as a function of temperature-dependent air sea exchange of CO2. Initially the coupled system will be spun up with a prescribed atmospheric CO2 level. Once an equilibrium is reached, this constraint will be removed and atmospheric CO2 will be allowed to freely interact with the ocean. It is unclear to what extent the resulting system will be stable to small climatic perturbations. As such we will add freshwater perturbations to high latitude convective regions, in order to perturb the system away from equilibrium. CO2 is more soluble in colder waters so the reducing North Atlantic overturning should cool the region and thereby draw down CO2. Atmospheric CO2, on the other hand, provides a powerful radiative forcing acting on the global scale so that small reductions will act to cool the surface ocean everywhere. This in turn would lead to further draw down of atmospheric CO2. It is unclear whether the resulting equilibrium climate will be the same as the climate of the initial state and it is conceivable that transitions to new equilibria will be induced through the positive solubility feedback. Such an analysis would provide substantial insight into the coupled climate system which would be useful to CCCma researchers as they move towards the inclusion of an ocean carbon cycle into their coupled GCM.

1.5.2 Do dynamic vegetation feedbacks contribute the glacial inception at 116 kyr BP?

In recent work funded under the NSERC CSHD project, Yoshimori et al (2001) examined the issue of glacial inception at 116 kyr BP in both the UVic ESCM as well as the CCCma AGCM. Initially, we integrated the UVic ESCM under both present-day and 116 kyr BP orbital forcing and atmospheric levels of CO2. We then integrated the CCCma AGCM with prescribed SSTs and sea ice mask taken from the UVic ESCM. In addition, we examined the sensitivity to specified vegetation changes in the land surface component of the CCCma AGCM, based on climate changes induced at 116kyr BP. In the CCCma model, perennial snow cover occurred over northern Canada under 116 kyr BP orbital and CO2 forcing with present-day warm sea surface conditions, and further expanded when 116 kyr BP cool sea surface conditions were applied. Modifying vegetation based on cooling during the summer induced by 116 kyr BP sea surface conditions, lead to much larger areas of perennial snow cover. Our results suggested that the capturing of glacial inception at 116 kyr BP requires the use of cooler sea surface conditions than the present. In addition, we showed how feedbacks induced through changes in vegetation type were important in capturing a more realistic representation of glacial inception.

The results from these experiments clearly highlighted the importance of both SST and vegetation feedbacks on glacial inception. These results have spawned this new area of research that I hope to pursue independently from CSHD funding (which is currently supporting researchers looking at ice sheet/climate interactions).

To accomplish this task we will couple the Hadley Centre TRIFFID/reduced MOSES code into our model with the goals of initially capturing the present day vegetation distribution. We will then integrate the model under 116 kyr radiative forcing to examine whether or not the vegetation feedback will allow ice sheets to begin to grow (where land ice will be modelled using the Marshall/Clarke thermomechanical model already coupled into the UVic ESCM). We will the take the surface SST, sea ice, and vegetation anomalies at 116 kBP relative to the present to change the lower boundary condition of the CCCma AGCM in order to further explore atmospheric changes which arise.

1.5.3 What is the relative role of changes in ocean SST, land surface moisture, and vegetation feedbacks for the changing African climate and biome at 6 kyr BP?

As noted in Chapter 8 of the IPCC Third Assessment Report: "Through PMIP [Paleoclimate Modelling Intercomparison Project] experiments, it is now well established that all atmospheric models are able to simulate several robust large-scale features of the Holocene climate but also that they all underestimate these changes. Several complementary simulations have shown that ocean and vegetation processes introduce important feedbacks which are necessary to explain the observed monsoon changes. These results urge for a systematic evaluation of coupled atmosphere-ocean-vegetation models for the mid-Holocene and for an investigation of the impact of vegetation changes, such as climate-induced density and land-use cover changes, on future climate change projections."

In this subproject I hope to shed insight into the role of vegetation changes and their feedback on climate with respect to the changing Aftrican climate and biome at 6 kyr BP, with the ultimate goal of being useful for future climate projections.

Joussame et al. (2000) have reported upon the results from 18 AGCMs run under 6 kyr BP CO2 and orbital conditions, albeit using specified present-day SSTs over the oceans. As noted by the IPCC TAR, the AGCMs all simulated warmer summer conditions in high northern latitudes, and drier summer conditions in the interior of the northern continents than today, although the latter is in conflict with paleo reconstructions (references can be found in McAvaney et al. 2001). Over the North African subcontinent, both AGCMs and paleo reconstructions suggest an increase and northward expansion of the African monsoon (see McAvaney et al. 2001), although model results substantially underestimated the extent of the northward displacement of the desert-steppe transition (see Fig. 3). Joussame et al. 2000) argued that an additional 100 mm/year of precipitation would be required for most AGCMs to sustain grassland at 23° N (Fig. 3).

Initial experiments with the UVic ESCM revealed precipitation changes in line with the AGCM results shown in Fig. 3. Unlike the AGCMs, the SSTs were allowed to evolve in the UVic model. Our exploratory results were also in line with those from the coupled A/OGCM experiments of Hewitt and Mitchell (1998) and Braconnot et al. (2000) in that SST feedbacks enhanced precipitation over North Africa at 6Kyr BP. We anticipate that our new land surface scheme will allow us to improve the model/paleo reconstruction comparison, and that the inclusion of TRIFFID will allow vegetation to further evolve and so provide additional feedbacks to the climate system (as in Texier et al., 2000, de Noblet et al., 2000). The recent work by Braconnot, et al. (1999), summarised in the caption and lower panel of Fig. 3 suggests that the combined effects of vegetation and ocean SST feedbacks will improve the model-paleo reconstruction comparison. Claussen et al. (1999), using CLIMBER-2, also emphasize the importance of vegetation feedbacks in explaining 6Kyr BP changes in subcontinental North African vegetation.

To round out this particular subproject, we will also drive the CCCma AGCM with SST, sea ice and vegetation changes as derived from the UVic ESCM. This last experiment should be viewed as a sensitivity analysis extension of the work of Vettoretti et al. (1998).

1.5.4 What is the role of the solubility pump in glacial to interglacial transitions.

We have recently coupled a 3-D, thermomechanical ice sheet model which employs continuum mixture theory to incorporate ice streams (Marshall & Clarke 1997a,b), into the UVic ESCM (Yoshimori et al. 2000) as part of the NSERC CSHD project. As a final subproject, which will be used as the link between the present proposed research and that being conducted under the NSERC CSHD project, we will investigate deglaciation from 21kyr BP in a series of long integrations.

Figure 3:Annual mean precipitation changes (mm/yr) over Africa (20°W-30°E) for the mid-Holocene climate: (upper) Biome distributions (desert, steppe, xerophytic and dry tropical forest/savannah (DTF/S)) as a function of latitude for present (red circles) and 6000 kyr BP (green triangles), showing that steppe vegetation replaces desert at 6000 kyr BP as far north as 23°N (vertical blue dashed line). (middle) 6000 kyr BP minus present changes as simulated by the PMIP models. The black hatched lines are estimated upper and lower bounds for the excess precipitation required to support grassland, based on present climatic limits of desert and grassland taxa in palaeoecological records, the intersection with the blue vertical line indicates that an increase of 200 to 300 mm/year is required to sustain steppe vegetation at 23N at 6000 kyr BP. (lower panel) same changes for the IPSL atmosphere-alone (A), i.e. PMIP simulation, the coupled atmosphere-ocean (OA), the atmosphere-alone with vegetation changes from OA (AV) and the coupled atmosphere-ocean-vegetation (OAV) simulations performed with the IPSL coupled climate model. The comparison between AV and OAV emphasises the synergism between ocean and land feedbacks. Taken from McAvaney et al (2001).

In the first, the UVic ESCM, including an inorganic ocean carbon cycle, will be spun up with a prescribed atmospheric CO2 level of 200 ppm and orbital forcing appropriate for 21 kyrBP. In this initial experiment, land ice will be incorporated as accumulated snow as in Weaver et al. (2001). Once the model has reached equilibrium, atmospheric CO2 will be allowed to evolve. It is not a priori known whether this equilibrium will be stable, and perhaps the system will evolve to a new state. A working hypothesis is that the preindustrial climate obtained in section 1.5.2 will be unstable, eventually evolving towards a colder mean state, whereas the glacial climate will be stable. This would be in qualitative agreement with late quaternary proxy records which suggest interglacial and glacial modes act as attractors of the climate system with the glacial mode being the preferred state.

The UVic ESCM will then be integrated for 15,000 years until 6 kyr BP in three modes: 1) with only an evolving inorganic carbon cycle; 2) as in 1) but with dynamic vegetation included; 3) as in 2) but including the interactive thermomechanical land ice model. These experiments will quantify the inorganic carbon cycle feedback in glacial-interglacial transitions.

2. Expertise:

While I cannot claim to be a leader in the area of carbon cycle and vegetation modelling I believe my track record and history of delivering on proposed research, even in areas where I did not previously have expertise, bears testament to my ability to deliver on this project. In my opinion, the most exciting breakthroughs in science come when one crosses traditional disciplinary boundaries and takes risks. What is necessary is to have the ability to learn, to be enthusiastic, to have a proven track record, to be able to surround oneself with the best postdocs and students available, and to collaborate with those who have more expertise in the area. I believe my recent Arctic research bears testament to this. In the course of two years, a sea ice model was developed within my research group for coupling to the CCCma model. Progress towards this goal was expedited by hiring two outstanding postdocs (C. Bitz and M. Holland). A version of this model has now been implemented in the CCCma model. In addition, I was asked to sit on the NSF Steering Committee on Arctic Ocean-Atmosphere Ice interactions (which oversaw the SHEBA, SBI and other initiatives), and I was asked to co-convene the Second International ACSYS scientific meeting in Orcas Island. My Form 100 will attest to my contributions in this area as well as in many other disciplines.

I also believe I have a track record of delivering technology to the CCCma. The ocean component of the first version of the CCCma coupled model was developed and tested in my lab. Many sensitivity studies were undertaken with my ESCM as the CCCma moved towards the coupling of their AGCM to the OGCM. Similarly, the sea ice model (mentioned above) is also now coupled into the CCCma model. There have also been other examples involving various subgrid scale ocean/ sea ice parametrisations that are being used in the context of the CCCma coupled model. My main collaborators in this project (Peter Cox at the Hadley Centre, Ken Denman in the CCCma) are outstanding leaders in their respective fields and so will ensure only highest calibre research will be conducted. I also believe I have a strong track record of being able to attract outstanding students and postdocs to work with me. 3. Targetted: The present proposal directly addresses three of the five targeted areas and indirectly addresses the other two. It is directly relevant to developing an understanding of key climate system processes (land surface, ocean carbon cycle, terrestrial vegetation feedbacks) including greenhouse gas sources and sinks (the ocean). This proposal should also be viewed as central to the both the 4th and 5th targetted criteria. It is aimed at developing a unique Earth System Climate Model for climate and paleoclimate studies, which should lead to better climate and marine climate prediction as the technology and knowledge which is gained is shared with the CCCma.

This proposal should not be viewed as ‘safe’ or ‘conservative’ science, but rather as high risk, cutting-edge science. If successful, the results would make fundamental contributions to our understanding of contemporary and past climates. They would be high profile and hence reflect well on CFCAS which would have funded the work. The ultimate goal of this project is to develop an enhanced capability in ocean carbon cycle and dynamic vegetation modelling within Canada. Through the training of highly qualified personnel, development of coupling technology, and exhaustive exploration of parameter space I will be able to assist those working in GC3M as they move towards the development of modules for the CCCma coupled model. At the very least, our analysis will allow the GC3M group to expedite their efforts by avoiding the pitfalls and blind alleys that we would encounter. This proposal has been specifically designed to address a number of highly focussed scientific questions in order to provide milestones and scientific deliverables throughout the three year research project. The questions are all very relevant to GC3M’s long term goals of developing a carbon cycle modelling capacity in the CCCma model. For example, it will be difficult for the CCCma model to explore parameter space to determine whether or not their model possesses multiple equilibria as a consequence of the inclusion of an interactive ocean carbon cycle. Our experiments (question #1) would provide guidance in this respect. Questions #2 and #3 are aimed at specific periods in Earth history for which there exists many paleo data, and for which no coupled model has been able to reproduce past climatic conditions inferred by these data. Our experiments will be aimed at trying to understand why this is the case. The final project (question #4) is aimed at an unresolved issue in paleoclimate modelling. It will be able to provide insight into the strength of the inorganic carbon cycle feedback in coupled models.

4. National: This proposal clearly addresses issues of national interest through improved understanding of terrestrial vegetation and ocean carbon cycle feedbacks on climate change/variability. These areas remain largely unexplored on the international arena and virtually unexplored on the national arena. It will also complement the work being done in the national CFCAS Global Coupled Carbon Climate Model (GC3M) project headed by Nigel Roulet and Ken Denman. 5. Funding As of October 1, 2001 my total direct funding will include a $58,000 NSERC Operating Grant and $90,000 as part of the NSERC CSHD Project. The former grant covers general operations of my laboratory and the latter covers the salary (and travel/publication charges) for a postdoc (A. Schmittner) and two PhD students (M. Cottet-Puinel, M. Yoshimori). The collaborative CRN Arctic project will award $200,000 to L. Mysak, E. Carmack, G. Flato and me. Of this amount, $40,000 is transferred to McGill, $50,000 pays the salary and operating costs of a research associate (Oleg Saenko) working with the CCCma model under the supervision of G. Flato. $40,000 pays the salary of two students (J. Dumas and L. Waterman) working with Greg Flato, $40,000 will pay the salary of two students (G. Arfeuille and J. Bacle) working with E, Carmack. The remaining $30,000 pays part of the salary of my research assistant W. Lewis and operating costs /travel/ publication charges for the Arctic students and postdocs. I have recently received CLIVAR funding to support the research of graduate students D. Stone, M. Roth and K. Hill and postdoctoral fellow H. Simmons. As the CICS Arctic funds wind down, CLIVAR funds will also be used to redirect Oleg Saenko into CLIVAR and for new co-supervised students within the CCCma. Here I am applying for funds to support my postdoctoral fellow K. Meisner, Research Associate M. Eby and two graduate students (D. Mathews and one other). There is no duplication of funding in this proposal.

6. Collaboration: This project will involve ongoing collaboration with Dr. K. Denman at the CCCma and P. Cox at the Hadley Centre in the UK. In January 2001, Katrin Meisner spent a week at the Hadley Centre with Dr. P. Cox during which time she acquired and became familiar with TRIFFID. In addition, Dr. Cox developed a reduced complexity version of his land surface scheme (MOSES) for use within our EMIC. The TRIFFID/reduced MOSES code is now at UVic and currently being tested and implemented (see letter of support)

Drs. K. Denman and G. Boer are not formally listed as a co-PIs on this project as I wanted to emphasise the different, yet complementary approaches of this project and the GC3M project. In addition, as outlined in the schematic below, interactions with GC3M will be via local collaborations. I do not want to take on more ‘network’ travel commitments although I would attend local GC3M workshops and send K. Meisner to workshops in other regions. The goal of GC3M is the development of a carbon cycle capability for the CCCma coupled model. As I develop and understand the issues involved in coupling ocean carbon cycle and terrestrial vegetation subcomponent models to our EMIC, I will be able to assist the GC3M group with the implementation of their modules into the CCCma coupled model. V. Arora in the CCCma will collaborate with K. Meissner to explore the sensitivity of TRIFFID to its internal parameters. He will be doing this while the terrestrial component of the GC3M project is developed (see letters from K. Denman and F. Zwiers attesting to the fact that this work will indeed be collaborative and of strategic importance to the CCCma). K. Zahariev and M. Eby will work together to understand the sensitivity of inorganic carbon chemistry in the CCCma and UVic models, respectively. I have also attached a letter for N. Roulet indicating how this project is complementary to and not competing with the GC3M project.

7. References not in Form 100