Compounding Extreme Events

Compounding extreme climate events are events that are spatially or temporally connected on scales such that their risks to humans and the environment are compounded and amplified. Anthropogenic climate change is expected to modify the spatial and temporal characteristics of individual extreme climate events, possibly leading the changes in compounding extremes. My research investigates these changes in recently observed and future climates.

What is the risk of a heavy precipitation event on a recently burned area? An intense storm on a burn scar can lead to flash floods and debris flows, and the compounded risks of the wildfire and extreme precipitation can amplify the impact to a region. Using the CESM1 Large Ensemble simulations, we find a strong increase in extreme fire weather events being followed by extreme rainfall events over the Western US by mid- and late-21st century. The largest increases in these compounding events are seen over the Pacific Northwest; by the end of the 21st century, there is a 700% increase in the number of extreme rainfall events following extreme fire weather events within one year. Here, over 90% of extreme fire weather events will be followed by extreme rainfall events by 2100. Our paper in Science Advances details the expected changes in post-fire hydrologic risks.

Number of extreme fire weather events followed by extreme rainfall events

Disentangling the roles of anthropogenic activity on extreme fire weather

Extreme fire weather, which is warm, dry, and windy conditions, leads to large wildfires to occur and spread. Given that wildfires are still too complex to simulate in global scale climate model simulations, we assess the meteorological conditions that could lead to wildfire. Read more about how we can quantify extreme fire weather using daily climate variables and explore the research topics that I tackle below.

Is adapting to the changes in the long term mean of our climate enough to overcome changes in extreme climate events? In this study, we explore the role of long term changes in the mean climate and changes in the variability of the climate in driving increases in the frequency, duration and area of extreme fire weather events using the CESM2 Large Ensemble. Using a fixed or historic threshold of extremeness, we find expanded and more persistent extreme fire weather events in the 21st century. However, when we use a continuously moving window threshold, changes in extreme fire weather events are less robust but more nuanced spatially. This in progress research will allow us to understand how extreme fire weather events respond to changes in the mean and variability of different climate variables, and what that could mean for future wildfire risk.

Can Solar Radiation Management (SRM) limit projected increases in extreme fire weather? One form of SRM, is stratospheric aerosol injection (SAI). SAI effectively reduces surface temperatures at global scales by SAI, but leads to an uneven emergence of regional patterns of temperature and other climate variables. So how do these varied outcomes of SAI impact extreme fire weather - a phenomenon that accounts for temperature, precipitation, relative humidity, and wind on daily to seasonal scales? We simulations of SAI in CESM2-WACCM (ARISE-SAI-1.5) and compare them to SSP2-4.5 to understand how extreme fire weather is impacted by SAI. We also disentangle the impacts of SAI on the daily and seasonal-scale variability of precipitation, temperature, wind, and relative humidity. This research, published in Earth's Future, allows insight into how SAI goals and implementation can modify wildfire risk over the globe, and how to overcome averse outcomes of SAI.

How do different anthropogenic activities change the risk of extreme fire weather -- i.e., warm, dry, and windy conditions that are suitable for fire ignition and spread?  By using a suite of  large ensemble single forcing climate model experiments (CESM-LE-SF), we are disentangling the impacts of greenhouse gases, aerosols, biomass burning, and land use change on extreme fire weather risk. By using the large ensemble experiments, we can quantify the uncertainties in our findings that stem from internal variability of the climate system and the robust signals that emerge. We found that aerosols have historically dampened the greenhouse-gas driven increases in extreme fire weather, but these effects are projected to decrease or reverse in future climates. The findings from this research expose how future climate change mitigation scenarios could change fire weather risk over time and in different regions. You can read more about it in our 2021 paper in Nature Communications, and the associated news article in The Current (UCSB)!

Extreme precipitation in the 20th and 21st centuries

Extreme precipitation can lead to high socioeconomic and environmental damages through flooding, nutrient transport, and crop damage. We look at the different characteristics of extreme precipitation and tropical cyclone precipitation throughout the 20th century using observations. In a paper led by Dr. Darren Ficklin, we are also now exploring the role the land in modulating the runoff response during extreme precipitation events using the CESM2 Large Ensemble (more information to come).

How do the spatial scales of extreme precipitation vary in historic and future climates over the US? When an extreme precipitation event has a large footprint it can cause associated damages to be more severe and long-lasting. In our 2018 Journal of Climate paper, we found that the spatial extent of extreme precipitation connectivity varied highly over different regions and seasons. By developing a novel geostatistical framework allowing us to use station data, we were able to achieve the longest climatological analysis of the spatial scales of extreme precipitation to date. We also looked at future trends of the spatial scales of extreme precipitation using high-resolution climate simulations. In this paper led by Dr. Deeksha Rastogi, we found that precipitation events are projected to become more intense and cover larger areas as the climate continues to warm in the 21st century

How do the characteristics of precipitation vary for different strengths of tropical cyclones? We used the same framework to analyze Atlantic tropical cyclone precipitation. Similar to our previous study, the use of station data as well as tropical cyclone track data, allowed us to achieve the most comprehensive analysis to date, from 1900-2018. In our 2019 GRL paper, we found that tropical cyclone precipitation was more intense and covered larger spatial areas of major hurricanes after they weakened to tropical storms. Additionally, we found that for these weakened major hurricanes, tropical cyclone precipitation intensity has increased in the second half of our study period, compared to the first half.

Uncertainties in drought projections

What are the uncertainties in the characteristics of drought projected for the 21st century? Projections of extreme climate events can suffer from model uncertainties. Additionally, how we define a drought event (i.e., index and threshold) can also cause discrepancies in 21st century projections. In my 2015 Journal of Hydrology paper, I analyzed drought characteristics for several CMIP5 models, and found that index uncertainties can be just as large as model uncertainties. When using a precipitation- or runoff-based index (SPI and SRI), there are only small and uncertain changes in drought characteristics. By accounting for changes in evapotranspiration (SPEI and SDDI), increases in the frequency, duration, and spatial extent of drought in the 21st century are large and robust among models.