## Climate Change Impacts on Ecology and the Environment

• Start Date: 05/04/2008
• End Date: 05/10/2008
Lecturer(s):

Alex Cannon (Meteorological Service of Canada)
Loveday Conquest (University of Washington)
Richard A. Fleming (Canadian Forest Service)
Aquila Flower (Pacific Climate Impacts Consortium)
Tilmann Gneiting (University of Washington)
David Higdon (Los Alamos National Laboratory)
Ulrich Horst (University of British Columbia)
Sylvia Esterby (University of British Columbia – Okanagan)
Zuzana Hrdlickova (University of British Columbia – Okanagan)
Jim Zidek (University of British Columbia)
Jim Ramsay (McGill University)
Charmaine Dean (Simon Fraser University)
Giles Hooker (Cornell University)
Rick Routledge (Simon Fraser University)
and other speakers as identified in the schedule

Location:

University of Alberta

Topic:

This workshop will bring together various experts in order to discuss the relevant issue of climate change.

Climate change is a serious force on the world today. Potential impacts include greater intensity in rainfall in some regions, leading to greater flood and landslide damage, and in other regions, there is likely to be increased droughts, resulting in smaller crop yields, more forest fires and diminished water resources. Climate shifts will change the lives and livelihoods of people around the globe, with a greater impact on the most vulnerable.

The development of methods to assess and quantify climate change and its impacts is a critical area of research. For rapid advancement in this field there is a need to work in a collaborative environment with a merging of tools in information science and ecosystem expertise because of the strong reliance on quantifying the changes using data analytic methods. This workshop will bring together a variety of environmental scientists, statisticians, and epidemiologists, in order to bring to bear modern methods of space-time modeling to assess and model the impact of climate change on agriculture, forestry and human health conditions

Description:

The workshop will bring together a variety of environmetricians in order to bring to bear modern methods of space-time modeling to assess and model the impact of climate change on forest and human health conditions. There has been only limited work by statisticians and stochasticians on the effect of climate change, and this workshop will enable us to set up a rich and politically important research agenda.

A primary aim of the workshop will be to discuss models and methods for the analysis of climate events and the assessment of the sensitivity of environmental outcomes to changes. More precise methods for assessment of evidence for existing impacts will be considered.

Topics include:

• Agroclimate risk analysis:
Climate change has led to the need for agricultural risk management. For example, because drought has assumed increasing importance, Agriculture Canada has now allocated funding for managing associated risk. Such management entails the development of appropriate models for precipitation and methods for finding optimal management strategies. In particular, new methods are needed for spatial precipitation analysis that admit covariates such as elevation, wind direction and velocity, along with other climate covariates to improve the accuracy of precipitation. Moreover, recent advances in statistical theory make it possible to estimate aspects of the entire distribution of rainfall at any given location and time point. Such advances include work on quantile regression and the development of functional analysis methods for estimating the distribution of a response variable at any point over the continua over which the data are distributed. Alternatively, advances in hierarchical Bayesian modeling enable continuous covariate fields (like temperature) to be input as latent variables into discrete (or mixed) process models, such as the binary process of the presence or absence of precipitation, using the clipped Gaussian (or multivariate t) approach. This leads to a Bayesian version of disjunctive kriging appropriate for use in climate change impact modeling of precipitation fields.

• Statistical and deterministic models in georisk analysis:
Neither statistical nor physical models like those in climate modeling suffice for modeling large-scale processes like those involved in climate change. A new modeling paradigm is spawning that seeks to combine the best elements of both. One promising approach uses Bayesian melding to integrate them. Central to this approach is the "truth” represented by a latent random Gaussian process. Deterministic models are assumed to be affine transformations of integrals of the truth over the grid cells that derived from the difference equation methods used to solve these deterministic models; monitoring site measurements, such as those involved in climate modeling, are similarly biased point measurements of the truth. A Bayesian hierarchical framework enables the integration of these two very different outputs. This framework enables uncertainty estimates to be added to deterministic outputs and those outputs to be dynamically tuned over time as data flows in. At the same time, the deterministic models may well provide statistical models with the backbones” they lack when confronting large - scale climatic processes. However, a number of research questions arise in connection with the use of deterministic models in a statistical domain. How should environmental processes best be monitored taking account of the need to protect human populations from the deleterious societal impacts of such change? How can deterministic models constructed at different scales of resolution be statistically integrated by the use of wavelets, for example. How can deterministic models be turned into practical forecasting tools, say to predict drought or exceptionally hot weather? Most environmental fields are not Gaussian processes but rather (after suitable transformation) multivariate-t processes. How can melding be adapted for use with such processes? How can the spatial melding approach best be turned into a space-time melding approach?

• Impact of Climate Change in Animal Populations and Human Health:
In order to assess the impacts of climate change on animal populations, it is necessary to know basic biological parameters about animal populations such as survival rates, population numbers, population movement rates etc. These are often obtained using methods such a capture-capture, point-count, or distance-sampling. The analysis of such studies requires sophisticated analysis tools not usually available, nor accessible to most biological researchers. Discussions will focus on the development of suitable methodologies in capture-recapture, distance-sampling, and point-count methodologies for these large scale problems. In addition, tools for both modeling and analysis of current ecosystems and inferring past conditions from natural recorders will be investigated. Identified needs for improved statistical methodology include relating deterministic, fluid-dynamic modelling to observations containing considerable stochastic components. In the area of human health, methods for quantifying risk for certain diseases (asthma, cardiovascular disease) associated with climate change and specific strategies to assess health impacts will be considered. In a particular and important application, methods for forest fire smoke modeling and linkages with adverse events related to asthma will be discussed.

• Aims and Scope:
The purpose of the proposed workshop is to engage climate change researchers in the scientific enterprise of developing novel methods for addressing these problems. It is envisioned that the proposed workshop would build upon collaborative initiatives already conceived through the environmetrics collaborative research group. Gaps in methodological developments will be identified, for example, methods for isolating the species and ecosystems most vulnerable to climate change. In addition, some techniques discussed will cross several of the themes, for example, detection of changes when observations are available at several spatial and temporal scales. This workshop will also play the important role of providing opportunity for discussion of timelines and progress toward the goals identified in the collaborative research group application on Georisk and climate change" by these organizers and will provide a forum for interim reporting on research objectives of that collaborative research group. It is envisaged that this workshop will bring together researcher in diverse scientific fields. It will also have a strong focus on student participation, bringing together students working in this area in a unique networking opportunity including both environmental and statistical sciences.

Programme08w5054[1].pdf
Schedule:

See .pdf attachment

Organizers:

Charmaine Dean - Simon Fraser University
Sylvia Esterby - University of British Columbia Okanagan
Peter Guttorp - University of Washington
Jim Zidek - University of British Columbia

Other Information:

### Participants

Name Affiliation

Bingham, Derek Simon Fraser University
Braun, John University of Western Ontario
Cannon, Alex Meteorological Service of Canada
Cao, Jiguo Simon Fraser University
Chen, Louis National University of Singapore
Chiu, Grace University of Waterloo
Conquest, Loveday University of Washington
Dou, Yiping University of British Columbia
Esterby, Sylvia University of British Columbia Okanagan
Feng, Cindy Simon Fraser University
Flower, Aquila Pacific Climate Impacts Consortium
Fu, Frances University of Western Ontario
Gneiting, Tilmann University of Washington
Higdon, Dave Los Alamos National Laboratory
Horst, Ulrich Humboldt University Berlin
Hosseini, Reza University of British Columbia
Hrdlicková, Zuzana University of British Columbia Okanagan
Liu, Zhong University of British Columbia
Loeppky, Jason University of British Columbia Okanagan
Martell, David University of Toronto
Newlands, Nathaniel Agriculture and Agri-Food Canada
Petkau, A. John University of British Columbia
Picka, Jeffrey University of New Brunswick
Podur, Justin York University
Ramsay, Jim McGill University
Reese, Shane Brigham Young University
Routledge, Rick Simon Fraser University
Sampson, Paul D University of Washington
Scott, Marian University of Glasgow
Sheppard, Lianne University of Washington
Smith, Ron Centre for Ecology and Hydrology
Stocks, Brian B.J. Stocks Wildfire Investigations Ltd.
Welch, Will University of British Columbia
Woolford, Douglas Simon Fraser University
Wotton, Mike University of Toronto
Zidek, Jim University of British Columbia