How does Risk Factor account for uncertain future climates?

Have more questions? Submit a request

By providing a range of possible flood depths and likelihoods, Risk Factor acknowledges the inherent uncertainty of climate projections and probabilities associated with flood risk estimation.

Data used to calculate a property's Flood Factor comes from the First Street Foundation Flood Model, which is based on decades of peer-reviewed research and developed in collaboration with more than 80 world-renowned experts. The Flood Model's methodology was reviewed by experts in the field, who confirmed its scientific rigor and deemed it worthy of scientific publication.

Risk Factor simplifies flooding so every American can determine their risk, understand the science, and make informed decisions to prepare for the future.

Climate modeling and changes in the environment

The First Street Foundation Flood Model is unique in that it presents risk of flooding with respect to a changing climate. The model explicitly accounts for climate uncertainty in a way that attempts to reflect current and future changing environmental factors associated with the mechanisms by which flooding occurs naturally. The factors taken into consideration as non-static climate inputs include: sea level rise, changing hurricane intensity and landfall locations, changing hurricane precipitation patterns and impact to river discharge landfall locations, and changing non-hurricane precipitation patterns and impact to river discharge.

Current climate conditions are estimated using the carbon emissions associated with the Intergovernmental Panel on Climate Change’s (IPCC) Representative Concentration Pathway (RCP) 4.5. This is a middle of the road scenario with respect to carbon concentrations in the atmosphere, representing some interventions but not the most aggressive scenario. This concentration pathway then informs the climate models that yield the changing environmental factors. The Flood Model used the output of an ensemble of 21 global climate models to identify three different potential values for each changing environmental factor - a high, a median, and a low. Flood Factor presents the depths of flooding and likelihood of flooding for the median scenario, while also displaying a range of values defined by the low and high scenarios. 

About RCP curves

The RCP curves represent an agreed upon set of scenarios that can be used in the development of research focused on future environmental changes. The standard set of comparative curves are based on carbon concentrations into the future, with trajectories related to scenarios in which emissions are relatively uncontrolled (RCP 8.5) on the high end to a dramatic restriction on emissions (RCP 2.6) on the low end. In contrast, RCP curves 4.5 and 6.0 are thought of as more moderate, and potentially more realistic trajectories into the future.

Screen_Shot_2022-05-16_at_9.52.11_AM.png

The Flood Model uses the RCP 4.5 curve to project future emissions characteristics, and associated environmental change, out into the year 2050. More importantly, the RCP 4.5 and RCP 6.0 curves are relatively indistinguishable from one another through 2050, making our models representative of a moderate projection of environmental change into the future.  

Climate uncertainty

To add climate uncertainty to our models, a series of simulations using the RCP 4.5 expectations and information from the GCM ensemble were conducted to produce a distribution of expected environmental factor outcomes from highly unlikely to highly likely. To round out our presentation of environmental change, we draw risk from the 25th percentile (low), 50th percentile (middle), and 75th percentile (high) of the distribution. These three projections represent the most likely projection using the RCP 4.5 curve (middle) and more unlikely low and high projections out to 2050.  By including the low, middle, and high projections on Flood Factor we are able to provide a more comprehensive set of expectations of flood risk into the future. The inclusion of these global climate models, forward-facing climate considerations, and high-resolution flood risk layers ultimately contribute to the uniqueness of the Flood Model in terms of coverage, precision, and climate adaptability.

Uncertainty and accuracy

By presenting the uncertainty associated with the flood model outputs, the Flood Model acknowledges the inherent uncertainty associated with both climate projection and probability associated with flood frequency estimation. Providing a range in this way shows different possible realizations of the true climate today, which cannot always be known from a limited set of observations over a short period of time, and what the climate could be in the future. This treatment provides for a greater confidence in the results, as is common in statistical and scientific research, but atypical of common flood modeling outputs.

Uncertainty of depth

Flood Factor shows this range of uncertainty first with respect to the depth of flooding in a given year and return period event (ie the 2020 1 in 100 event). A median scenario. For instance, this property shows 7.8’ of flooding in the median scenario 1 in 100 event, but a range of 6.9’ (low) to 8.7’ (high).

climate_uncertainty.png

Uncertainty of probability

This range of depth also enables First Street to quantify a different likelihood of flooding to a certain depth in the low and high climate scenarios. Because the depth of water can greatly affect the damage done by a flood, First Street calculates the likelihood of flooding at all, to 6”, and to 12”. Using a statistical relationship between depth and probability, First Street calculates the probability of flooding 12” or more in the low, median and high climate scenarios. 

 

Learn more

Flood Model Methodology - Calculating property-level risk

What is urban flooding? - Visit Risk Factor to check your risk

Tidal flooding - you could be at risk even on sunny days

Flood Factor probabilities - The likelihood of water reaching a home

 



Articles in this section

Was this article helpful?
0 out of 0 found this helpful
Share