Progress 10/01/09 to 09/30/10
Outputs
OUTPUTS: Progress reported by PSW 4402 under this problem area includes the development of a climate/fire danger forecasting system. The fire climate/fire danger forecasting system gave fire weather specialists at the Southern California GACC a new tool for formulating fire severity forecasts. The forecasting system performance continues to be monitored and plans have been developed to couple the system with a statistical model that predicts probability of ignitions and number of fire occurrences. This statistical model to predict ignition probability and large fire occurrences was adapted for an online application by the EROS Data Center, for use with the Fire Potential Index. Work is currently underway to integrate the statistical model with the fire climate forecasting system. PARTICIPANTS: S. Chen, Research Meteorologist. PSW Riverside Fire Laboratory. Lead scientist for this problem area, conducting and coordinating multidisciplinary research in long-range fire danger forecasting. F. Fujioka, Research Meteorologist. PSW Riverside Fire Laboratory. Lead scientist for guiding research direction and coordinating research activities with potential user agencies. J. Ritchie, Scripps Institution of Oceanography, Experimental Climate Prediction Center, Systems Manager. Maintained hardware and software systems and databases used for fire research at ECPC. H. Preisler, Statistician, PSW Albany. Provided lead for development of probability model for fire occurrence projection. A. Westerling, Collaborator, School of Engineering, UC Merced. Provided research assistance and data in development of probability model for fire occurrence prediction. H.-M. Henry Juang, Collaborator, National Centers for Environmental Prediction. Provided modeling and forecast support. J. Benoit, Computer Analyst. PSW Riverside Fire Laboratory. Provided data analysis and archival. Managed cluster computer which was used for model simulations. TARGET AUDIENCES: Target audiences for this project are fire planners at regional to national levels, where long-term fire danger assessments are needed to manage resources in a weekly to seasonal time frame. Research also conducted for the benefit of scientists developing applications that use numerical weather prediction for long-term planning. PROJECT MODIFICATIONS: None.
Impacts
We completed and published an evaluation of the monthly-to-seasonal fire climate/fire danger forecasting system which we developed jointly with the Scripps Institution of Oceanography (SIO) Experimental Climate Prediction Center and the NOAA National Center for Environmental Prediction in Washington, DC. The target application for this system is national and regional prediction of fire severity on a weekly to seasonal time scale, which Predictive Services units perform for Geographical Area Coordination Centers (GACCs) across the country and the National Interagency Fire Center in Boise. The study showed some skill in the prediction of monthly to seasonal fire index means, particularly in the western US. An important complementary study reported this fiscal year described a statistical method to forecast the probability and number of large fires for areas of interest from gridded fire environment variables such as weather and climate elements, and fire danger indices. The combination of this statistical methodology and the preceding fire climate forecasting system translates fire climate forecasts into fire ignition and large fire occurrence forecasts, along with forecast uncertainty information useful for decisionmaking.
Publications
- Mallet, V.; Keyes, D.E; Fendell, F.E. 2009. Modeling wildland fire propagation with level set methods. Computers and mathematics with applications 57(7):1089-1101
- Preisler, H.K.; Burgan, R.E.; Eidenshink, J.C.; Klaver, J.M.; Klaver, R.W. 2009. Forecasting distributions of large federal-lands fires utilizing satellite and gridded weather information. International Journal of Wildland Fire 18:508-516.
- Roads, J.; Tripp, P.; Juang, H.; Wang, J.; Fujioka, F.; Chen, S. 2010. NCEP-ECPC monthly to seasonal US fire danger forecasts. International Journal of Wildland Fire 19:399-414
|
Progress 10/01/08 to 09/30/09
Outputs
OUTPUTS: We co-sponsored a workshop with the Scripps Institution of Oceanography/Experimental Climate Prediction Center on the spectral family of weather/climate models. An international community of atmospheric scientists described various studies of spectral models to simulate weather and climate from regional to global scales. We explained how the Forest Service was using the weather models in tactical and strategic fire planning, in applications from fire danger rating to fire behavior prediction. We also published research on characteristic variations of the Keetch-Byram Drought Index (KBDI) in Hawaii, which has previously been shown to be a potential long-lead indicator of an active fire season in the state. This research correlated high KBDI at leeward stations with an El Nino index and with anomalous surface anticyclonic circulation, surface divergence and subsidence in winter, fall and spring. PARTICIPANTS: S. Chen, Research Meteorologist. PSW Riverside Fire Laboratory. Lead scientist for this problem area, conducting and coordinating multidisciplinary research in long-range fire danger forecasting. F. Fujioka, Research Meteorologist. PSW Riverside Fire Laboratory. Lead scientist for guiding research direction and coordinating research activities with potential user agencies. J. Roads, PI, ECPC Director, Scripps Institution of Oceanography, UC San Diego. Lead cooperator for weekly to seasonal fire climate prediction R&D. Supervised atmospheric research activities at ECPC that related to fire management applications designed by PSW. Leveraged funding from NOAA to develop applications that use National Center for Environmental Prediction models. J. Ritchie, ECPC Systems Manager. Maintained hardware and software systems and databases used for fire research at ECPC. H. Preisler, Statistician, PSW Albany. Provided lead for development of probability model for fire occurrence projection. A. Westerling, Collaborator, School of Engineering, UC Merced. Provided research assistance and data in development of probability model for fire occurrence prediction. H.-M. Henry Juang, Collaborator, National Centers for Environmental Prediction. Provided modeling and forecast supports. J. Benoit, Computer Analyst. PSW Riverside Fire Laboratory. Provided data analysis and archival. Managed cluster computer which was used for model simulations. TARGET AUDIENCES: Target audiences for this project are fire planners at regional to national levels, where long-term fire danger assessments are needed to manage resources in a weekly to seasonal time frame. Research also conducted for the benefit of scientists developing applications that use numerical weather prediction for long-term planning. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.
Impacts
We transferred fire danger rating research at Scripps to the National Weather Service/National Center for Environmental Prediction in Washington DC, with whom we are developing a prototype forecasting system to generate NFDRS index predictions on a national scale. We are working with the NWS on an experimental 5-member ensemble coupled atmosphere-ocean global modeling system launched at the beginning of each month for 7-month forecasts. We continue working with NCEP in validating these predictions for fire management. Downscaling NCEP's operational weather forecast to 1-km grid space is underway.
Publications
- Dolling, K.; Chu, P.-S.; Fujioka, F. 2009. Natural variability of the Keetch-Byram Drought Index in the Hawaiian Islands. Intl. J. Wildland Fire. 18:459-475.
- Nunes, Ana M.B.; Roads, John O. 2005. Improving regional model simulations with precipitation assimilation. Earth Interactions, 9:1-44.
|
Progress 10/01/07 to 09/30/08
Outputs
OUTPUTS: In partnership with the Experimental Climate Prediction Center (ECPC) at the Scripps Institution of Oceanography, UC San Diego, we provided prototype weekly to seasonal fire weather/fire danger predictions for the US and regional predictions for the Southwest and California. Users view the products on a web site maintained by ECPC. The PI provided input to seasonal fire danger outlooks prepared jointly by Predictive Service units at the geographic area coordination centers and the National Interagency Coordination Center at Boise. PARTICIPANTS: S. Chen, Research Meteorologist. PSW Riverside Fire Laboratory. Lead scientist for this problem area, conducting and coordinating multidisciplinary research in long-range fire danger forecasting. F. Fujioka, Research Meteorologist. PSW Riverside Fire Laboratory. Lead scientist for guiding research direction and coordinating research activities with potential user agencies. J. Roads, PI, ECPC Director, Scripps Institution of Oceanography, UC San Diego. Lead cooperator for weekly to seasonal fire climate prediction R&D. Supervised atmospheric research activities at ECPC that related to fire management applications designed by PSW. Leveraged funding from NOAA to develop applications that use National Center for Environmental Prediction models. J. Ritchie, ECPC Systems Manager. Maintained hardware and software systems and databases used for fire research at ECPC. H. Preisler, Statistician, PSW Albany. Provided lead for development of probability model for fire occurrence projection. A. Westerling, Collaborator, School of Engineering, UC Merced. Provided research assistance and data in development of probability model for fire occurrence prediction. H.-M. Henry Juang, Collaborator, National Centers for Environmental Prediction. Provided modeling and forecast supports. J. Benoit, Computer Analyst. PSW Riverside Fire Laboratory. Provided data analysis and archival. Managed cluster computer which was used for model simulations. TARGET AUDIENCES: Target audiences for this project are fire planners at regional to national levels, where long-term fire danger assessments are needed to manage resources in a weekly to seasonal time frame. Research also conducted for the benefit of scientists developing applications that use numerical weather prediction for long-term planning. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.
Impacts
As a result of our long standing cooperation with ECPC in long-range experimental prediction for fire applications, the National Weather Service is adapting the NFDRS forecast application developed at Scripps as part of their standard output for their operational fire weather and climate predictions. This action responds partially to user requests in recent surveys and a resolution to Congress from the Western Governors Association.
Publications
- Chen, S.C.; Preisler, H.; Fujioka, F.; Benoit, J.; Roads, J. 2008. Seasonal prediction for wildland fire severity. Third International Symposium on Fire Economics, Planning, and Policy: Common Problems and Approaches. April 29-May 2, 2008, Carolina, Puerto Rico.
- Preisler H.; Chen, S.C.; .Benoit, J.; Fujioka, F.; .Westerling, A. 2008: Wildland fire probabilities estimated from weather model-deduced monthly mean fire danger indices. International Journal of Wildland Fire 17: 305-316
|
Progress 10/01/06 to 09/30/07
Outputs
We continued the development of statistical models relating the number and location of large fire events in the western US to climate, drought, and fire index variables. A model to predict large fire occurrence from monthly mean temperature and the Palmer drought severity index showed significant promise in its ability to discern areas of high probability of large fire occurrence from areas with low to moderate probability, particularly in comparison with predictions based on historical fire frequency. The statistical methodology developed in this research can be applied in similar analyses at different spatial and temporal scales.
Impacts
The fire occurrence forecast methods derived from this research will improve the strategic fire planning process used to determine how firefighting resources are allocated across the country prior to the start of the nation's collective fire season.
Publications
- Chen, S.-C.; Preisler, H.; Benoit, J.; Fujioka, F. 2006. Projecting wildland fire severity using RSM simulations with a probability model. In: Proceedings, conference on weather analysis and forecasting, 2006 October 18-20, Taipei, Taiwan. Publisher and number of pages unknown.
- Preisler, H.K.; Westerling, A.L. 2007. Statistical model for forecasting monthly large wildfire events in Western United States. J. Applied Meteorology and Climatology 46:1020-1030.
- Roads, J.; Tripp, P.; Juang, H.; Wang, J.; Chen, S.; Fujioka, F. 2007. ECPC/NCEP March 2007 seasonal fire danger forecasts. Experimental long-lead forecasts bulletin 16:1-7.
- Roads, J.; Tripp, P.; Juang, H.; Wang, J.; Chen, S.-C.; Fujioka, F. 2006. Experimental monthly to seasonal fire danger forecasts. VAMOS Newsletter.
|
Progress 10/01/05 to 09/30/06
Outputs
We developed a statistical model to produce spatially and temporally explicit maps of probability of occurrence of large fire events. With an external cooperator, we developed a computer algorithm to forecast monthly fire danger on the basis of estimated probability models. The maps are distributed to users through the Internet. Also on a climate time-scale, we generated statistical relationships between a drought index and fire activity in Hawaii. Relationships were strong between the monthly mean Keetch-Byram Drought Index and total acres burned by month on the islands of Oahu, Maui and Hawaii. Further work continues to determine if fire managers can use these results in a system to predict monthly fire potential.
Impacts
Approximately two percent of the agency's largest fires have accounted for up to 70 percent of its total fire suppression expenditures. Management can use improved climate scale forecast models to formulate more effective and efficient presuppression strategies to reduce firefighting costs.
Publications
- Preisler, H.K.; Westerling, A.L. 2005. Estimating risk probabilities for wildfires. In: Proceedings, Joint Statistical Meeting, 2005 August 7-11, Minneapolis, MN.
- Brillinger, D.R.; Preisler, H.K.; Benoit, J.W. 2006. Probabilistic risk assessment for wildfires. Environmetrics 17:623-633.
- Dolling, K.; Chu, P.-S., Fujioka, F. 2005. A climatological study of the Keetch/Byram drought index and fire activity in the Hawaiian Islands. Ag. and For. Meteorol. 133:28-39.
|
Progress 10/01/04 to 09/30/05
Outputs
We studied the effects of horizontal resolution and land surface process representation in a weather model on the accuracy of surface weather simulations. Estimates of surface temperature, dew point, wind direction and speed were significantly improved with a horizontal grid spacing of 1.5 km vs. 10 km. Coupling the weather model with a land surface model also resulted in better simulations. This research is ongoing. We used global and regional spectral models to generate forecasts ranging from 0-16 weeks for fire management applications. Even in a seasonal timeframe, we found some forecast skill in predicting long-term means of National Fire Danger Rating System indices. We continued to work on methodology to translate the forecast weather information to expected impacts on fire activity.
Impacts
Progress in the high resolution weather modeling research will enhance our ability to accurately predict fire growth. The long-range forecasting research will benefit operational planning in regional area coordination centers and the National Interagency Coordination Center in Boise, ID.
Publications
- Juang, H.-M.; Lee, C.-T.; Zhang, F.Y.; Song, Y.; Wu, M.-C.; Chen, Y.-L.; Kodama, K.; Chen, S.C. 2005. Applying horizontal diffusion on pressure surface to mesoscale models on terrain-following coordinates. Monthly Weather Review 133:1384-1402.
- Reinbold, H.; Roads, J.O.; Brown, T. 2005. Evaluation of the Experimental Climate Prediction Center's fire danger forecasts with remote automated weather station observations. International Journal of Wildland Fire 14:19-36.
|
Progress 10/01/03 to 09/30/04
Outputs
We studied the effects of horizontal resolution and land surface process representation in a weather model on the accuracy of surface weather simulations. Estimates of surface temperature, dew point, wind direction and speed were significantly improved with a horizontal grid spacing of 1.5 km vs. 10 km. Coupling the weather model with a land surface model also resulted in better simulations. In related studies, we used the high resolution weather model output to describe weather to a fire spread model. One of the cases was the Old Fire, which burned the northern portions of San Bernardino, CA in 2003. This research is ongoing. In another study, we found differences in drought characteristics between windward and leeward locations in Hawaii, using the Keetch-Byram Drought Index. Long-period harmonics (36-60 months) of the KBDI at leeward stations showed a strong statistical relationship with El Nino. Long-range forecasts were also a part of our research agenda. We used
global and regional spectral models to generate forecasts ranging from 0-16 weeks for fire management applications. Even in a seasonal timeframe, we found some forecast skill in predicting long-term means of National Fire Danger Rating System indices. We continued to work on methodology to translate the forecast weather information to expected impacts on fire activity.
Impacts
Progress in the high resolution weather modeling research will enhance our ability to accurately predict fire growth. In October 2003, we demonstrated that we could prepare timely forecasts of weather and fire spread for operational use on the Old Fire, in the San Bernardino National Forest. The long-range forecasting research will benefit operational planning in regional area coordination centers and the National Interagency Coordination Center in Boise, ID.
Publications
- Brown, T.J.; Fujioka, F.M.; Fontana, C. 2003. The California and Nevada smoke and air committee (CANSAC): An interagency partnership to meet decision-making needs. In: Proceedings of the fifth conference on fire and forest meteorology; 2003 November; Boston, MA. American Meteorological Society; 6 p.
- Chen, P.S.; Dolling,K.P.; Fujioka, F.M. 2003. A climatological study of the Keetch/Byram drought index in the Hawaiian Islands. In: Proceedings of the fifth conference on fire and forest meteorology; 2003 November; Boston, MA. American Meteorological Society; 9 p.
- Chen, Y.L.; Zhang, Y.X.; Hong, S.Y.; Kodama, K.; Juang, H.M.H. 2003. Validations of the NCEP MSM coupled with the NOAH LSM over the Hawaiian Islands. In: Proceedings of the fifth conference on fire and forest meteorology; 2003 November; Boston, MA. American Meteorological Society; 15 p.
- Dolling, K.P.; Chu, P.S.; Fujioka, F.M. 2003. The validity of the Keetch/Byram drought index in the Hawaiian Islands. In: Proceedings of the fifth conference on fire and forest meteorology; 2003 November; Boston, MA. American Meteorological Society; 6 p.
- Jones, C.; Dennison, P.E.; Fujioka, F.M.; Weise, D.R.; Benoit, J.W. 2003. Analysis of space/time characteristics of errors in an integrated weather/fire spread simulation. In: Proceedings of the fifth conference on fire and forest meteorology; 2003 November; Boston, MA. American Meteorological Society; 7 p.
- Preisler, H.K.; Brillinger, D.R.; Burgan, R.E.; Benoit, J.W. 2004. Probability based models for estimation of wildfire risk. International Journal of Wildland Fire. 13:133-142.
- Roads, J.; Chen, S.; Fujioka, F.M.; Burgan, R. 2003. Fire danger forecasts. In: Proceedings of the fifth conference on fire and forest meteorology; 2003 November; Boston, MA. American Meteorological Society; 8 p.
- Yang, Y.; Chen, Y.L. 2003. High resolution simulations of the island-induced circulations for the island of Hawaii during HaRP. In: Proceedings of the fifth conference on fire and forest meteorology; 2003 November; Boston, MA. American Meteorological Society; 8 p.
|
Progress 10/01/02 to 09/30/03
Outputs
In FY 2003, PSW-4401 was absorbed by PSW-4402, with the Fire Meteorology research agenda unchanged. Problem 1 focused on the development of weather models for higher spatial and temporal resolution than is currently available to fire managers. One application integrated a high resolution weather model and the FARSITE fire behavior modeling system. Unit scientists developed a new method to analyze errors in the integrated modeling system in FY 2002, which was peer-reviewed and published in FY 2003. Additional cases were identified for study, and the Troy Fire of June 2002 is currently being simulated and analyzed as a result. This fire provides the opportunity to use high resolution fire spread data from the Riverside Fire Lab's Firemapper imaging system. The funded research and development of high resolution weather models are running on personal computer clusters at the University of California, Santa Barbara, and the University of Hawaii. The UCSB system will
provide improved predictions for fire applications in California, and the University of Hawaii system will do the same for fire management in Hawaii. The weather models will support both fire behavior and fire danger rating forecasting.
Impacts
An integrated weather/fire behavior modeling system developed by the former Fire Meteorology Research Work Unit (PSW-4401) passed peer review and is being used in additional case studies of the Williams Fire and Troy Fire of 2002. The system will provide improved modeling capabilities for suppression and prescribed fire planning. It includes an error analysis methodology to construct probability-weighted error bounds on model-based fire spread predictions. This information will be useful for risk assessment and decision support in fire operations planning.
Publications
- Fujioka, F. M., 2002. A new method for the analysis of fire spread modeling errors. Intl. J. Wildland Fire, 11:193-203.
|
|