Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

MAPP Webinar Series: Fire: Modeling and Prediction Issues (Part 2)

unnamed2

Idealized smoke simulation from a linear stationary fire.
Image Credit: Adam Kochanski, University of Utah. Image Source.

The NOAA CPO Modeling, Analysis, Predictions, and Projections (MAPP) program is hosting a webinar on the topic Fire: Modeling and Prediction Issues (Part 2) on Tuesday, April 24, 2018, 1- 2:30 p.m. ET

Date/Time Title & Presenters
April 24, 2018
1:00 PM – 2:30 PM ET

 

Fire: Modeling and Prediction Issues (Part 2)
Speakers and Topics Uma Bhatt (University of Alaska, Fairbanks) 

Towards Co-Production of Seasonal Forecast for Fire Managers in Alaska

U. S. Bhatt1, A Sampath1, PA Bieniek1, A York1, R Ziel1, B Brettschneider1, R Thoman2, H Strader3, S Alden3, R Jandt1, G Branson3, P Peng4  
(1U. Alaska Fairbanks, 2NOAA Alaska, 3Predictive Services, 4NOAA CPC)

The record fire year of 2004 resulted in 6.5 million acres burned and high costs from property loss (> $35M) and emergency personnel (> $17M). Fires in Alaska result from fuel availability and lightning strikes coupled with persistent dry warm conditions in remote areas with limited fire management. Seasonal climate determines the extent of the fire season in Alaska. The need for seasonal outlooks is increasing as budgets tighten and efficient allocation of resources/staff requires planning a season in advance. Though currently few tested products are available at the seasonal scale, probabilistic forecasts of the expected seasonal climate/weather is highly needed.

For fuel conditions, the Canadian Forest Fire Weather Index System (CFFWIS) has been used in Alaska since 1992 as it was developed for the boreal forest. The CFFWIS is based on 2-m air temperature, RH, 10-m winds and daily precipitation. The Buildup Index (BUI) is an integrated index of fuel conditions and has been constructed for the seasonal forecasts of June-August that are initialized in March and May for the NOAA CFSv2 model. Fire managers use divisions of the state called Predictive Service Areas (PSA) of which there are 21 in Alaska. Seasonal forecasts of BUI were prepared for the PSAs over which the observational data is of good quality. Forecast BUI is lower than observed for all the PSAs, particularly in spring.  May-June forecast temperatures are too low and precipitation is too high, resulting in low BUIs. A bias correction for precipitation and temperature is being developed for the CFSv2-based using quantile mapping.

This type of analysis helps to identify shortcomings in forecasts of parameters that are needed for decision support. This activity also encourages us to think more creatively about how current forecasts can be made into skillful products. Developing seasonal forecasts and their products is a challenge but they are critical for decision support.

Edward Delgado (Bureau of Land Management)

Challenges for Wildland Fire Forecasters

Wildland fire forecasting is a critical part of decision support for managing and suppressing fire on the landscape. Land managers need to this support to: know where fire is likely to occur; identify when and where to position limited fire fighting resources to maximize effectiveness and efficiency; and ultimately, protect life and property. There are challenges to wildland fire forecasting. Wildland fire varies across the country both seasonally and geographically. It is drive not just by the natural elements of weather and vegetation but also by the day-to-day activities of people. Number of ignitions, spread or growth rates, and final size of fires provide some usefulness in accounting but their value in forecasting decreases when factors such as management strategies, values at risk, and resource capabilities enter the equation. Meteorologists must rethink how wildland fire forecasts are made. This will require better definitions of what is being forecast, approaching the problem from the perspective of fire business, and providing scalable products to address spatial and temporal needs. 

Adam Kochanski (University of Utah) 

Recent advancements in smoke modeling using coupled fire-atmosphere model WRF-SFIRE

The United States has entered a new era of increasing wildfire frequency & intensity and worsening fire impacts. The landscape has become more fire-prone as a result of recent climatic change and urban development resulting in steeply rising fire-suppression costs. Yet, fire is a part of the natural environment and fire prevention practices can at times lead to excessive fuel accumulation and catastrophic fires that are difficult to manage. The need for management decisions based on multifaceted analyses of benefits and risks associated with both wildfires and prescribed burns, including smoke impacts calls for new advanced decision support tools that interactively integrate satellite/aerial remote sensing with coupled high-resolution fire-weather modeling.

As the resolution of operational weather prediction products increase, coupled forecasting of fire progression, smoke generation, as well as plume rise, and dispersion becomes feasible. This integrated approach, based on coupled fire-atmosphere models, facilitates simulations in which not only the weather conditions drive fire propagation, but the fire itself also impacts local weather conditions through the fire heat and moisture fluxes released into the atmosphere. However, fire impacts on the weather conditions are not limited to local warming and generation of pyro-convective updrafts inducing inflows into to base of the convective column. The smoke itself proves to be an important factor significantly altering local weather conditions by its impact on the radiative heat budget.

In this presentation, we illustrate general capabilities of WRFX (the integrated forecasting system based on WRF-SFIRE), in terms of simulating plume rise and dispersion. We also present new model developments that extend model capabilities in terms of rendering fire-atmosphere interactions. The existing coupling mechanisms, through the wind field (modified by fire heat and moisture fluxes) and through the fuel moisture (controlled by local the weather conditions), are now extended by radiative smoke impacts. We present selected test cases and compare modeled plume rise to MISR observations. We also show how the new coupling mechanism improves estimates of the incoming solar radiation and the surface temperatures in smoked valleys.

James Randerson (University of California, Irvine) 

Global Fire Prediction on Daily and Seasonal Timescales

 

Watch Webcast

 

Scroll to Top