All models are wrong, but some are useful

In light of the ever increasing human population and its growing demands for space and natural resources, it is clear to me that conservation of biodiversity is faced with serious challenges.

The overarching question we face is “how do we accommodate human development and yet ensure the protection of the natural world?”

Kevin McGarigal

Today you’re hearing from Dr. Kevin McGarigal, a professor in the Department of Environmental Conservation at the University of Massachusetts, Amherst, and director of the Landscape Ecology Lab.

One of the lessons learned over the past century of conservation is that acting locally without thinking regionally is not going to ensure the protection of biodiversity in a heterogeneous and changing landscape.

Instead, we need to conserve functioning landscapes that facilitate ecological processes such as dispersal and gene flow that allow ecosystems and species to persist and adapt in the face of change – we need a “landscape approach” to biodiversity conservation.


Probability of persistence of the blackburnian warbler from 2010-2080 in the middle Connecticut River watershed pilot study area averaged across several climate change and urban growth scenarios. The non-zero values are restricted to where the species is most likely to occur in 2010. The areas with the highest probability of persistence (dark green) are the areas where the species most likely occurs today and where it is most likely to occur in the future given our uncertainty in climate and habitat changes. The areas shown in black are where the species is least likely to occur in the future if the species contracts its range in response to the expect climate changes. Based on our landscape change projections, the blackburnian warbler is most likely to persist only in the northwestern portion of the watershed by 2080 if it contracts its range in response to the expected climate changes.


Loss of ecological integrity (i.e., impact) due to climate change and urban growth from 2010-2080 in the middle Connecticut River watershed pilot study area averaged across several landscape change scenarios. The larger the negative index, the greater the effective loss in ecological integrity; in other words, the loss in ecological integrity from cells that currently have high ecological integrity — where it matters the most. The greatest impacts on ecological integrity are patchily distributed throughout the watershed owing to the dispersed development in the future, but also associated with major urban growth areas in the Connecticut River valley and along the major east-west transportation corridors where the development is close to areas of high IEI.

Moreover, in light of the inevitable landscape changes, driven by processes such as global climate change and urban growth, it is clear to me that we need to anticipate the likely changes in order to make strategic conservation decisions today.

For example, protecting a particular patch of critical habitat for a species of conservation concern is unlikely to be effective if the climate will no longer be suitable for the species at that location in 50 years – we need to embrace the notion that landscapes are dynamic, and plan accordingly.

Lastly, given the complexity of biodiversity, it is obvious to me that effective conservation solutions will require the integration of complex, multivariate ecological and socio-economic data. Fortunately, thanks in part to the recent explosion in remote-sensing technology, there are now vast amounts of relevant spatial data available that can be brought to bear on the conservation challenge.

Given the need to integrate lots of ecological and socio-economic data across multiple scales in space and time, I contend that a computer modeling approach to biodiversity conservation is urgently needed. Accordingly, in my role as principal investigator for the Designing Sustainable Landscapes project of the North Atlantic Landscape Conservation Cooperative, I have focused the project on developing a modeling approach to simulate changes to the landscape under a variety of alternative future scenarios (e.g., climate change, urban growth), assess affects of those changes to the integrity of ecological systems and populations of priority fish and wildlife species, and inform the design of conservation strategies (e.g., land protection, management and restoration) to meet biodiversity conservation objectives.

The project will allow us to forecast changes to the landscape across the Northeast and assess the consequences of those changes to biodiversity at multiple scales in space and time. This information will help inform the tradeoffs associated with alternative conservation actions, so that limited conservation resources can be directed to where they will do the greatest good.

And while this model does not include all of the factors that must go into real-world conservation decisions, I am confident and excited about the prospects that the model will provide useful information. My confidence is boosted by knowing that the North Atlantic LCC, through its extensive conservation partnership among Northeast agencies and organizations, will deliver this information to conservation agencies and organizations — the practitioners that get things done on the ground.

To learn more about the North Atlantic LCC and the DSL project, please click here to download the project factsheet. (PDF)

Read more from this series!

One Comment on “All models are wrong, but some are useful

  1. Pingback: The right science in the right places | U.S. Fish and Wildlife Service Northeast Region

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: