On April 21st, CBI and The Nature Conservancy hosted a meeting in Duluth, MN to discuss climate change in northeastern MN and to begin developing scenarios that could be used with a forest simulation model to estimate the effects of climate change on the extensive forests of northeastern MN.

Our objective at the meeting was to formulate 2-4 scenarios that can reduce uncertainty regarding the effects of climate change on northeastern Minnesota forests. The model inputs and outputs will be shared via the Data Basin on-line data sharing tool in order to solicit feedback from the group regarding desired additional scenarios. These additional 15-20 scenarios will be developed iteratively over the next 18 months.

Below we have outlined our proposed initial scenarios based on feedback from the 22 meeting participants, below. Participants from the meeting can comment directly on the proposed scenarios via this blog posting. If you wish for your comments privately, you can directly access the same information as PDF or Word documents via the Project Web Site. and email your comments to us directly. You can also access other background information as well as examples from previous scenarios at the project web site.

Deadline: Please submit all comments by June 5, 2009. Thank you!

Background
We will use a simulation modeling approach to assessing potential landscape-scale disruptions due to climate change and potential management solutions. Specifically, we will use the LANDIS-II landscape simulation model developed by the University of Wisconsin, the Conservation Biology Institute, and the U.S. Forest Service. We will develop a new succession extension that directly incorporate a forest growth model in order to better incorporate inter-annual variability, drought and CO2 fertilization effects.

There are limits to what our scenarios can include due to our model, our budget, and our time line. Where possible, we welcome collaborations to link our model with other models; however that cannot be our primary objective. Specifically, we will not be able to consider:

• economic factors
• budgets or markets
• land use change or fragmentation
• lowland forest dynamics driven by hydrologic changes

We are only able to simulate management actions. If we can define actions as a consequence of economic forces or markets, then they can be implicitly modeled. Land use change and fragmentation can be modeled within our framework but we lack the resources to do so. This is an area where an interested collaborator could make significant and quick contributions simply by applying rates of land use change to our model inputs.

Initial Scenarios
At the meeting, there were three common themes discussed as potential nuclei for forming scenarios: precipitation, pro-active management, and invasive species.

Precipitation
A consensus emerged that we should focus on two potential futures: ‘hotter and drier’ and ‘warmer and wetter’. These are relative terms that need careful parsing. Hotter is assumed to mean towards the high end of temperature changes predicted by the recently available Global Circulation Models (GCMs). Warmer is assumed to mean towards the low end of temperature changes predicted by the same. Drier does not assume less precipitation. Rather, drier will be the largest decline or smallest increase in available water for plants (roughly precipitation – evapotranspiration). Likewise, wetter will be the largest increase in available water.

We will discuss with TNC climate scientists and others which GCM projections best represent ‘hotter and drier’ and ‘warmer and wetter’. Although an ensemble approach was suggested, this will not be our initial approach. Ensemble approaches require considerable additional data synthesis and they may also average out inter-annual variation in precipitation. Therefore, we will use ‘off-the-shelf’ GCM projections.

The group also emphasized the need to capture annual and monthly changes in precipitation. In particular, droughts may reduce regeneration, reduce growth, and cause mortality. In order to accommodate these effects, we will develop and add a new succession extension that incorporates the PnET forest growth model in order to better incorporate inter-annual variability and drought effects.

In addition to drought effects, we also plan to incorporate a broader suite of disturbance and disturbance interactions. These may include fire with variation across ecoregions and insect outbreaks and mortality (e.g., spruce budworm). The combination of drought effects and disturbance interactions will allow us to assess landscape response to more severe disturbances under different management regimes.

Pro-active Management
We didn’t have as much time in the meeting to discuss potential management scenarios. However, we heard from numerous people that a range of activities should be explored. Previously, we have modeled ‘current rates and patterns’ and ‘restoration focused rates and patterns’. The logical next activity to model may be a ‘climate mitigation’ scenario. This could include significant facilitated migration and other activities (TBD) that would maximize forest productivity and forest resilience. Again, we cannot assess the economics of these actions; only the actions themselves. We must assume that productive, resilient forests are economically viable.

Invasive Species
Although this theme was consistently highlighted by the groups, we may need to delay this scenario. The principle difficulty is that we lack the necessary ecophysiological data about potential invaders and information about their current distribution across northeastern Minnesota. We have very good data about common tree species that are used as model inputs. However, the required data for potential invaders (e.g., leaf nitrogen, lignin content, potential growth rates, reproductive limits due to climate, etc.) are much less common. We therefore propose to delay this scenario until sufficient data become available.

Potential Metrics and Output

The model produces a wide range of outputs. Producing all possible outputs can be time consuming, therefore stakeholder inputs are critical. Potential outputs include:

• Species composition and diversity
• Age structure and diversity
• Productivity
• Harvested biomass
• Aboveground and belowground Carbon storage
• Measures of spatial pattern and landscape structure
• Actual evapotranspiration, runoff