WMO/CAS/WWW
FIFTH INTERNATIONAL WORKSHOP on TROPICAL CYCLONES
Special Focus Topic 4: Global Climate Change and Tropical Cyclones
Prepared By: T. R. Knutson
Geophysical Fluid Dynamics Laboratory/NOAA
Forrestal Campus, U. S. Rt. 1
Princeton NJ 08542, USA
E-mail: tk@gfdl.noaa.gov
Fax: 609.452.6509
Session Leaders: T. Knutson, G. Holland, W. Gray
Abstract:
A special focus session will be held at IWTC-V on Global Climate Change and Tropical Cyclones. Since the assessment on this topic by Henderson-Sellers et al. (1998) a number of studies have provided relevant new information. This research will be reviewed in the session, and the question of whether a new assessment should be undertaken will be discussed. A final goal of the session is to provide recommendations for future research efforts in this topic area – particularly to identify those approaches that hold the most promise for providing improved information for future assessments.
SF 4.1 Introduction
This document provides background material for the Special Focus Session on Global Climate Change and Tropical Cyclones. The objectives for session are outlined, new results since the previous assessment are summarized, and research questions for possible future studies are posed. The document is intended as a starting point to stimulate discussion at the meeting.
The assessment by Henderson-Sellers et al. (1998) is taken as the point of departure when discussing new results. This document is accessible on the Web at http://www.aoml.noaa.gov/hrd/Landsea/IPCC/. Some assessment of possible future changes in tropical cyclones was also contained in the IPCC Third Assessment Report (2001; http://www.grida.no/climate/ipcc_tar/wg1/index.htm). However, the IPCC assessment was more brief and not as quantitative as the Henderson-Sellers report as the tropical cyclone/climate change issue was only a relatively minor part of the overall IPCC Report.
SF 4.2 Objectives of the Special Session
The primary objectives of the special session are:
- Have a series of presentations (with time for questions and discussion) on research relevant to the topic. (30-45 minutes)
- Discuss whether developments on the topic since Henderson-Sellers et al (1998) was written merit a new "post IPCC" assessment at this time. If so, decide how to put together the new assessment. (30 minutes)
- Attempt to formulate a list (prioritized) of recommended research activities that have a good probability of increasing credible scientific information on the topic. (30-45 minutes)
Note that since the special session will only last about 1.5-2 hours, the presentations will need to be restricted to material that was not available for the last assessment. The Henderson-Sellers et al. (1998) assessment can be regarded as the “point of departure” for the topic, with emphasis on what new information is now available and what are the implications of the new information.
SF 4.3 Summary of Relevant Research since the Previous Assessment
A question to ask before developing a new assessment is whether enough new information has been published since the Henderson-Sellers et al. (1998) assessment to justify an updated TC/global climate change assessment. The following is a list of some relevant new findings since Henderson-Sellers et al.:
1) More support for the Emanuel/Holland view of increased potential intensities in a CO2-warmed climate (based on the GFDL hurricane model runs, even including ocean coupling (Knutson et al. 2001)).
2) A new statistical analysis (Emanuel 2000) suggests that Potential Intensity theory has implications for the entire intensity distribution -- not just the upper bound.
- GFDL hurricane model experiments indicate potentially large increases (15-30% over 100 years) in hurricane-related precipitation in a warmer climate (Knutson et al. 2001).
- Observed multi-decadal variability in Atlantic hurricane activity appears correlated to a multi-decadal SST "mode" in the Atlantic and to local vertical wind shear, which implies that the present high level of hurricane activity in the basin may persist for several decades (Goldenberg et al. 2001). A GCM simulation study (Vitart and Anderson 2001) provides additional support for the link between Atlantic hurricane activity and these factors.
5) Can hurricanes affect climate? This possibility is raised in a recent study (Emanuel 2001) estimating hurricane effects on poleward ocean heat transport. (Unfortunately Kerry Emanuel will not be coming to the meeting. It would be useful to have someone familiar with the ocean heat transport issues briefly discuss this issue -- perhaps Isaac Ginis?)
6) Further work on paleotempestology – the reconstruction of pre-historic tropical cyclone activity from proxy geologic records – has been completed. (We should try to have someone familiar with developments in that field at the meeting.)
7) A few more recent general circulation model (GCM) studies have been published showing either a decrease or little change in tropical cyclone frequency with climate warming.
8) Further progress has been made in the re-analysis of historical tropical cyclone data.
These new findings may justify a new assessment of tropical cyclones and climate change, although this will be a topic for discussion at the special session. If a new assessment is to be undertaken, some procedural aspects should be discussed at the session (e.g., ways to improve process from previous experience?).
SF 4.4 Recommendations for future research activities
As there are many remaining uncertainties in the area of climate change and tropical cyclones, there are many opportunities for future research. A goal of the session is to provide useful guidance to researchers in the field by attempting to provide recommendations for research activities that we believe will lead to improved credible scientific information on the topic. Since the list of potential recommendations could be quite lengthy, we should attempt to prioritize the activities, perhaps by emphasizing those that we believe hold the most promise for reducing uncertainties and that are most directly relevant to the inter-disciplinary problem of climate change and tropical cyclones.
As a start, a brief status report and some potential research questions are given below for different sub-topic areas related to this problem.
- Observations of TCs and TC-related fields
There is currently little evidence in the historical record for significant long-term trends in tropical cyclone frequency and intensity. Some relevant research questions include:
- What are the prospects for improved historical coverage in terms of time period and different basins?
- Can data homogeneity problems be in TC historical records be reduced?
- Historical records of several TC-related variables – SST, atmospheric temperature, moisture, and wind shear — are also useful for climate studies. Can these be improved in terms of homogeneity and coverage? For example, will either radiosondes or reanalysis be able to provide useful information on multi-decadal variability or trends?
- Given historical changes in various factors (e.g., SST, atmospheric temperature) how much change in various TC measures would be expected? Can the environmental data and tropical cyclone historical data be reconciled?
- Can Gray’s genesis parameters (infer TC frequency from large-scale characteristics of the tropical climate) or Potential Intensity theories be further evaluated using interannual variability in the present observational record?
- Can paleo-reconstruction provide information relevant to the climate change/TC behavior issue?
- Is there any evidence for land-use change effects on TCs?? (a modeling issue as well)
- Are there critical gaps with regard to observing systems that need to be addressed to ensure adequate data for future TC/climate change studies?
- Future Intensity of Tropical Cyclones
Potential Intensity (PI) calculations and nested model simulations indicate greater upper limit intensities in a CO2 warming scenario. For nested simulations the intensity change is of the order of 5-10%/100yr. This change is only marginally significant in one set of case studies (Knutson and Tuleya 1999), and mostly not significant in a second set of case studies with lower resolution nested model (Walsh and Ryan 2000). However, the simulated intensity increase is quite robust in more idealized simulations with the GFDL hurricane model. Using global models, TC intensity results vary from essentially no change to some intensification at the high-intensity end of the distribution. Higher resolution models (~20 km or less) appear to be necessary for simulating fairly realistic tropical cyclone intensities. The simulated changes are small relative to interannual variability measures, particularly in the Atlantic basin, which implies that they are unlikely to be detectable for many decades.
- Do projected PI changes have applicability to the entire frequency distribution of storm intensities, i.e., mean vs. maximum achievable intensity (Emanuel 2000)?
- What is the status of PI research/issues (e.g., ocean spray effects, comparisons of PI vs observations)?
- How robust are nested simulations or PI results to the use of different models (both host GCM and the PI method or nested model)? How dependent are the results on resolution or physical parameterizations?
- Climate model projections have large regional uncertainties (in additional to global mean uncertainties) that can affect tropical cyclone behavior. For example, will the mean climate become more El Nino-like? How will El Nino change with global warming? What are the prospects for reducing these uncertainties in climate models?
- Atmospheric temperature and moisture are important for TC intensities, not just SSTs (e.g., Shen et al. 2000, PI theories). How model-dependent are profiles of atmospheric temperature and moisture changes in global climate change scenarios? Do different radiative forcing scenarios have important impacts on the profiles of temperature/moisture change?
- How important are dynamical effects vs thermodynamical effects on TC intensity? Do present models adequately simulate the effects of wind shear on TCs?
- Can the detectability of projected changes be estimated? Can prospects for detection be enhanced through improved observations?
- Do tropical cyclones have an important impact on poleward heat transport by the ocean? If so, what are the implications of projected TC intensity changes for net poleward heat transport?
- Future Frequency of Tropical Cyclones
Global climate models have been used to simulate the frequency of occurrence of tropical storms based on counting occurrences of tropical storm-like features. The present-day tropical storm frequency simulations of these models are realistic in many respects although they typically also show some unrealistic features. In 2xCO2 climate change experiments, there is little consistency so far between models with regard to future changes in tropical cyclone frequency -- even the sign of change varies among models. However, more simulations, particularly higher resolution global ones, show either little change or a decrease in frequency as opposed to an increase. There are many potential sources of the differences among models, including differences in SST changes (magnitude and regional structure), physical parameterizations, methods of counting storms, experimental design (specified vs computed SSTs; mixed layer vs coupled models), model resolution, radiative forcing, sample size, etc.
- Is simulated cyclogenesis in GCMs realistic i.e., does it occur for the right reasons? Tropical cyclogenesis is still very much a research topic, for example the impact of vertical wind shear.
- Are all model simulations for climate change equally plausible or are some studies more reliable than others?
- Are there other opportunities for evaluating models that can increase confidence in their predictive utility (e.g., seasonal cycle and spatial pattern of TC occurrence by basin; TC spatial structure, intensity, and tracks; Interannual or interdecadal variability of TC frequencies in specified sea surface temperature experiments following Vitart and Anderson 2001).
- Should Gray’s genesis parameter be used more routinely to evaluate tropical part of GCM control runs? How can it be modified for use in climate change context?
- Should other methods, such as surface wind speed extremes, be used as an alternative to counting tropical storms in GCMs?
- Is there a preferred method of assessing TC frequency changes in current generation models?
- Can the relation between the multi-decadal SST mode and Atlantic hurricane activity be exploited for predictive capability?
- Future Tropical Cyclone-related Precipitation
Precipitation associated with tropical cyclones increases by a larger percentage than surface wind intensity in nested hurricane model simulations using the GFDL model (Knutson et al. 2001). Precipitation from tropical cyclones has potentially large societal impact, for example from flood damage. Tropical cyclone precipitation changes have received much less attention than tropical cyclone frequency in most TC/climate change simulation studies.
- How can global model or regional model TC precipitation simulations best be evaluated for the present climate?
- How robust are simulated precipitation changes to different models (either host GCM or nested model)?
- Precipitation change may potentially be more detectable with larger signal, but are the data adequate? How large would the change need to be in order to be detectable using the present observing system and historical data?
- Tropical Cyclones Tracks and Location
Changes in tracks or locations of occurrence of tropical cyclones could have important societal impacts, but currently information is very limited. Tropical cyclogenesis is not expected to necessarily expand along with the area covered by the 26.5 degree SST isotherm. One study for the Southwest Pacific region (Walsh and Katzfey 2000) found that tropical storms tended to travel slightly further poleward under 2xCO2 conditions.
- As the finding of more poleward occurrence of TCs is still limited to the Walsh and Katzfey study for the Southwest Pacific region, how robust is this result?
- What impact does model resolution have on TC track simulations?
- Future changes in storm surge
Storm surges associated with tropical cyclones could change due to changes in storm characteristics or due to other climate-related influences such as sea-level rise.
- Can projected changes in storm intensity be used to infer changes in storm surge? What types of models are required (e.g., coupling to storm surge models) for this type of integrated assessment?
SF 4.5 Summary
Since the tropical cyclones and climate change assessment of Henderson-Sellers et al. (1998) significant additional research on this topic has become available. However, many areas of uncertainty need to be addressed. This background document has attempted to summarize these areas of recent progress and unresolved questions in order to stimulate further discussions at the IWTC-V special session.
Bibliography
Emanuel, K., 2000: A statistical analysis of tropical cyclone intensity. Mon. Wea. Rev., 128, 1139-1152.
Emanuel, K., 2001: Contribution of tropical cyclones to meridional heat transport by the oceans. J. Geophys. Res., 106: D114, 14,771-14,781.
Goldenberg, S. B., C. W. Landsea, A. M. Mestas-Nunez, and W. M. Gray, 2001: The recent increase in Atlantic hurricane activity: causes and implications. Science, 293, 474-479.
Henderson-Sellers, A., H. Zhang, G. Berz, K. Emanuel, W. Gray, C. Landsea, G. Holland, J. Lighthill, S.-L. Shieh, P. Webster, and K. McGuffie, 1998: Tropical cyclones and global climate change: A post-IPCC assessment. Bull. Amer. Meteor. Soc. 79, 19-38.
IPCC, 2001: Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change [Houghton, J.T., Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, K. Maskell, and C. A. Johnson (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 881 pp.
Knutson, T. R., and R. E. Tuleya, 1999: Increased hurricane intensities with CO2-induced warming as simulated using the GFDL hurricane prediction system. Clim. Dyn. 15: 503-519.
Knutson, T. R., R. E. Tuleya, W. Shen, and I. Ginis, 2001: Impact of CO2-induced warming on hurricane intensities as simulated in a hurricane model with ocean coupling. J. Clim. 14, 2458-2468.
Shen, W., R. E. Tuleya, and I. Ginis, 2000: A sensitivity study of the thermodynamic environment on GFDL model hurricane intensity: implications for global warming. J. Clim. 13, 109-121.
Vitart, F., and J. L. Anderson, 2001: Sensitivity of Atlantic tropical storm frequency to ENSO and interdecadal variability of SSTs in and ensemble of AGCM integrations. J. Clim. 14, 533-545.
Walsh, K. J. E., and B. F. Ryan, 2000: Tropical cyclone intensity increase near Australia as a result of climate change. J. Clim. 13, 3029-3036.
Walsh, K.J.E., and J.J. Katzfey, 2000: The impact of climate change on the poleward movement of tropical cyclone-like vortices in a regional climate model. J. Clim., 13, 1116-1132.