Fundamental Research: ESIG Annual Scientific Report 2003

Fundamental Research

Assessing Socioeconomic Value of Weather and Climate Forecasts

The goal of this research is to develop methodology for evaluating the societal and economic value of imperfect weather and climate forecasts, as well as to incorporate this information into decision-making. ESIG continued its contribution during FY03 to the planning of an international research program (i.e., design of appropriate measures of value of potential improvements in weather forecasts as a consequence of The Hemispheric Observing System Research and Predictability Experiment [THORPEX]). In FY03, Rebecca Morss (ESIG/MMM) collaborated with Lenny Smith (U Oxford), Melvyn Shapiro (NOAA), Alan Thorpe (U Reading) and a number of other societal impacts and THORPEX researchers to develop the Societal and Economic Applications (SEA) subprogram for the International THORPEX Science Plan. With F. Martin Ralph and Zoltan Toth (both of NOAA), she also developed the Applications subprogram for the NOAA THORPEX plan, including an FY04 NOAA Announcement of Opportunity for THORPEX. She took the lead in developing a similar subprogram for the US and North American THORPEX Science Plans.

Also during FY03, Morss collaborated with Bill Mahoney (RAP), Robert Harriss, and Bob Gall (MMM) and others to initiate a collaborative program on the societal and economic benefits of weather information. An NCAR proposal to develop the program was submitted to the USWRP/NOAA and was recently funded. The program is an important step toward developing the capacity to conduct the societal impacts research and data gathering necessary to meet USWRP, NOAA, NCAR, and THORPEX goals to improve weather forecasts for the benefit of society.

In collaboration with Martin Ehrendorfer (U Innsbruck, Austria), Richard Katz developed a Bayesian approach to decision making using ensemble weather forecasts. Whether in terms of reliability, skill, or economic value, the apparent effects of ensemble size on forecasting performance are smaller than those previously obtained.

Designing "optimal" observing systems requires balancing the costs and benefits of different observations. Although the meteorological community has discussed the question of optimal investment in observations for more than three decades, it still lacks a practical, systematic framework to analyze the issue. In FY03, Rebecca Morss collaborated with two economists, Kathleen Miller and Maxine Vasil (U Colorado-Boulder) to develop such a framework using an economic approach. To demonstrate the framework, they are analyzing the appropriate level of investment in observations for an idealized example, based on previous research results. This work will continue in FY04.

Carbon Cycle Science

As governments and corporations evaluate their options to respond to climate change, carbon sequestration and active management of the carbon cycle are now being considered. Since the carbon cycle is an integrated system, the need to make integrated data and knowledge available to decision-makers and scientists has become more acute. Lisa Dilling and colleagues are developing supporting projects and human dimensions research for the North American Carbon Program (NACP). Dilling is providing input into developing the implementation plan for the NACP as a member of the drafting committee. During FY03, in preparation for possible participation in a community-based "State of the Carbon Cycle" report, Dilling is collaborating with the NCAR GIS Initiative, the USGS National Atlas, and the Consensus Building Institute to explore developing stakeholder-centered communication of carbon cycle information, using GIS as one of the central tools. Another area of exploration is the study of the role of different scales of decision-making in determining land use patterns (and hence carbon storage and release patterns over land). The ultimate goal of this research is to compare social and physical scales to determine how scientific information might be appropriately focused to support decision-making at different scales. Proposals have been submitted for possible funding in FY04. During FY03, Dilling and colleagues completed a review paper highlighting the potential role of carbon cycle science in carbon management and calling for an expanded dialogue between scientists and decision-maker (Dilling et al., 2003).

Circle of Knowledge: Climate, Weather and Environmental Change

Shannon McNeeley attended a workshop in May 2003 in Huslia, Alaska, that began an extensive collaboration with Alaska Natives along the Koyukuk River in Alaska and scientists who study Arctic and global climate change. A paucity of data and observations in this region have increased interest in the integration of western science with indigenous knowledge and Native observations of climate change. Native elders and hunters possess a legacy of traditional knowledge and observations of weather and environmental change, and they are concerned about the dramatic changes they are witnessing. McNeeley proposes to formulate a model for documenting Native observations of climate change, while exchanging information on climate, weather, and environmental changes between Natives and scientists. This project will continue through FY04. (Click on thumbnail at left for larger image showing Alaska Native concerns about weather and climate change 1998-2000.)

Effects of Climate on Simulated Crops

While most studies of the impacts of climate change on ecosystems have examined the effect of mean change in climate, it is widely believed that climate variability, in addition to the mean, has substantial effects on ecosystems. This issue is becoming more important as we learn more about how climate may vary in coming decades. During FY03, Linda Mearns and colleagues continued examination of the possible additional effects of changes in variability on simulated crops. Mearns, with Cynthia Rosenzweig and Richard Goldberg (NASA Goddard, New York), continued research on the effect of changes in variability of climate on simulated crop yields at other locations in the Great Plains and Midwest. They applied time series of temperature and precipitation with changed variances to CERES-corn and CROPGRO-soybean models. Results so far indicate that increased variance of temperature and precipitation cause substantial decreases in yield, while decreases in variability cause only slight increases in yield. They have begun applying changes in variance from two major atmosphere-ocean general circulation models (AOGCMs), the NCAR Parallel Climate Model (PCM), and that of the NASA Goddard Institute for Space Studies, for the region of the Midwest and Great Plains to these crop models for the end of the twenty-first century. With Marta Vinocur (National University of Redo, Cordoba, Argentina), Mearns has investigated simulated peanut crop responses to climate variability in Cordoba, Argentina. Using PeanutGRO, they explored the effects of different combinations of mean and variance changes of temperature. They found that the crop model was sensitive to both mean and variance changes, but that increases in temperature variance substantially exacerbated decreases in yield and greatly increased the likelihood of crop failures. They are currently exploring the causes for these crop model responses.

Ethics of Generation and Use of Weather and Short-Term Climate Forecasts

This research effort systematically analyzes the assumptions, methods, limitations, and uses/abuses of weather and climate forecasts from an ethical perspective. Criteria and methods are drawn from modern applied ethics; in particular the "Georgetown School" of analytical criteria and the method of iterative "specification of reflection" developed by John Rawls. In collaboration with Rebecca Morss, Eugene Wahl is using these tools to develop a framework specifically oriented to quandaries that arise in the preparation, provision, and use of weather and short-term climate forecasts. Specific case situations are being analyzed with this set of tools, such as the allocation of scarce observational resources in the face of competing demands for the additional accuracy and length of the forecast period that these resources can provide. The initial design phase of this research was completed in FY03. Completion of case analyses and submission of research findings to Science and Engineering Ethics is planned during FY04. Dale Jamieson (City U of New York) is acting as adviser on systematic and applied ethics for this project.

Exploration of GIS Interoperability Technologies for International H2O Project

A research project funded by NSF and conducted by Terri Betancourt (RAP), Olga Wilhelmi and Jennifer Boehnert (ESIG/RAP) explored interoperability technologies currently available in the realm of GIS (Geographic Information Systems). In particular this project was designed as a demonstration of technology options from the OpenGIS standards community, as well as from ESRI, the leading commercial vendor of general-purpose GIS software. The atmospheric context of the GIS demonstration project was tied to the International H2O Project (IHOP) conducted during 2002. The extensive array of heterogeneous atmospheric data collected during IHOP provided an exceptional opportunity to address the question of how GIS and related interoperability technologies can address the question of how GIS and related interoperability technologies address the need to integrate a wide range of existing atmospheric data into a common geospatial framework. The spatial data interoperability was addressed in this research project from the perspectives of different user communities. More information can be found on the new designed GIS website at www.gis.ucar.edu.

Hindcasting Little Ice Age and Last Millennium

The Hindcasting component of the Weather and Climate Impact Assessment Strategic Initiative focuses on the uncertainty characteristics of external forcing and their influences on climate, including variability on different spatial and temporal scales. This project was initiated in FY03 in cooperation with CGD to involve simulations of multiple centuries, including an initial period of the Little Ice Age. This multiyear project will "fingerprint" volcanic and solar impacts on climate (forcings). During FY03, Caspar Ammann (CGD), Philippe Naveau (U Colorado) and Hee-Seok Oh (U Alberta, Canada) generated and applied new statistical tools to extract external forcing fingerprints from climate time series. The ESIG component for FY03 involved using millennium-length climate simulations to compare model-derived proxy measures of El Nino-Southern Oscillation (ENSO) activity with the model's actual ENSO activity over a very long time scale. Linda Mearns and Eugene Wahl, along with Caspar Ammann, Doug Nychka, and Bette Otto-Bleisner of CGD, have completed initial processing phases of this research. Research findings will be submitted to the appropriate journals in FY04. Additional information is available on the following website: www.cgd.ucar.edu/ccr/asr03

Land Cover Forcing from SRES Scenarios in Climate Models

The most recent Intergovernmental Panel on Climate Change (IPCC) SRES (Special Report on Emissions Scenarios) was released in 1998 to assist in the development of a new set of emissions scenarios in order to provide input to the IPCC Third Assessment Report. This current project was begun in FY03 as part of the Weather and Climate Impact Assessment Strategic Initiative in collaboration with CGD to extend future climate change scenarios to include human impacts on land cover and soils. Linda Mearns and colleagues in ESIG and Gordon Bonan in CGD, with involvement from Johannes Feddema (U Kansas) have created two alternative present-day land cover representations and run them in the Parallel Climate Model (PCM). Initial results indicate that the model is very sensitive to land cover change. Additionally, they found that when changing from potential vegetation to an equivalent present-day land cover, the PCM predicts significant cooling, particularly in the Northern Hemisphere midlatitude regions. Continuing work in FY04 includes the creation of a time series of land cover change for the PCM from 1870 to the present, to simulate land-cover change in the twentieth century, as well as continued development of an Urban Canyon Model to be incorporated into the Community Land Model.

Problem Definitions for Designing Observing Networks

Because resources for meteorological observations are limited, choices must be made among proposed enhancements to the weather prediction observing system. In FY03, Rebecca Morss wrote a manuscript (submitted to BAMS) analyzing choices among observing systems from a public policy research perspective. The paper explains the role and importance of problem definition in policy research to a meteorological audience, using five alternate definitions of the observing system design problem in order to demonstrate how different problem definitions can lead to different results. The problem definitions presented build toward an appropriate problem definition for observing system design, an important prerequisite for designing a more cost-effective, integrated global observing system.

Reducing Uncertainty in Paleoenvironmental Reconstruction

Eugene Wahl's research activity on this project centers on technical issues in the calibration of pollen records of fossils for environmental reconstruction. The work covers a wide range of topics focused on the use of the Modern Analog Technique (MAT) that examines how modern pollen assemblages can be used to reconstruct modern and paleoenvironmental conditions and vegetation. Wahl is concentrating on (1) designing mathematical methods for optimizing numerical thresholds used to determine analogs; (2) examining algorithms and rules for converting chosen analogs into climate and vegetation reconstructions, and (3) developing Monte Carlo simulation methods to examine how thresholds and reconstruction rules interact with count sizes of pollen samples in order to optimize scientific efforts toward the development of greater spatial and temporal resolution of pollen-based records. The results of this work will help to establish new methods and calibration benchmarks in this field, and are applicable to other micro-fossil paleoenvironmental archives.

Superstorm '93

On 12-15 March 1993, a major winter storm affected 26 US states, Cuba, and eastern Canada, causing high winds, severe weather, storm surges, heavy rain, record snowfalls, and record low temperatures. Several weather prediction models and forecasters at the National Meteorological Center recognized the threat of a significant storm and were able to predict the storm several days in advance. However, both models and forecasters failed to predict the extent of the rapid deepening of the storm. During FY03, Michael Glantz gathered together researchers to begin a project that will explore the potential for improving different types of forecasts of similar winter storms, as well as the potential value of such improvements to society. This two-year project involves NCAR scientists across three divisions: MMM, ESIG, and CGD. They are studying the event as being representative of a class of events and are approaching the project as a unique and exploratory effort to integrate meteorological and societal impacts knowledge, research, and research applications. The project involves external collaborators from Cuba, Canada, and the University of Colorado. This project will continue through FY04.

Uncertainty in Climate Model Simulations

This continuing project of the Weather and Climate Impacts Assessment Strategic Initiative developed new techniques during FY03 for quantifying uncertainty in climate model projections and began to apply these techniques to recent transient runs of AOGCMs (atmosphere-ocean general circulation models). Linda Mearns joined colleagues from CGD (Doug Nychka, Jerry Meehl, and Tom Wigley), along with Claudia Tebaldi (ESIG/RAP) to analyze "super ensembles" of climate change runs in order to characterize the uncertainty in the AOGCMs. The same team, with the addition of Bruno Sanso (U California) and C. Forest from MIT, examined statistical models for interpolating between scenarios in order to infer climate change responses under patterns of anthropogenic forcings that have not been directly simulated by general circulation models. These human-caused effects can be approximated by combining results from patterns of emissions. This work will continue development during FY04.