Understanding the Factors that Impact Climate Risk Management Strategies
Planners use all kinds of projections to understand how climate change will affect infrastructure, but all projections are based on assumptions that change their outcomes. Decision makers are tasked with weighing different projections to design cost effective risk management strategies to protect populations from rising seas.
Researchers from Penn State University analyzed a suite of sea-level rise and storm surge projections for Norfolk, Virginia to understand what factors influenced the outcomes of the projections the most. Their work was recently published in Scientific Reports.
The team was aware of how much divergence there can be between different projections of the same area because there are lots of ways to represent uncertainties. They chose to look at Norfolk because the city has good historical data and has been featured in numerous projections.
Similar to how a researcher might conduct a literature review, the team reviewed multiple projections and analyzed underlying assumptions for each one and how they captured uncertainties. The projections were standardized for the same time period so they could be directly compared.
Five factors that differentiate between climate projections
Adaptability is hugely important to planning infrastructure, “but that’s difficult to do if you’re not fully aware of what assumptions are baked into a particular projection,” explained Dr. Vivek Srikrishnan, an assistant research professor in the Earth and Environmental Systems Institute (EESI) at Penn State.
The analysis identified five factors that influence sea-level rise and storm surge projections the most. Assumptions about the behavior of ice sheets led to the greatest differences in the projections. In fact, models that accounted for the potential of rapid melting of ice sheets projected 1.7 to 4.5 more feet of sea-level rise by 2100.
Srikrishnan noted that it’s crucial for decision makers to understand which models include rapid ice sheet melting as a possibility. “If you focus on a projection that doesn’t excludes rapid ice sheet melting, you may not prepare for very high sea-level rise , so it may reduce the robustness of your plan if that [level of rising] does end up occurring,” he explained.
Other factors were projections that look at a time scale past a century, taking into account storm surges that have a low probability of occurring, and understanding that storm surges might deviate from the historical record. Srikrishnan explained that these factors are important because massive storm surge heights may be outside the range of historical data taken into consideration in the projection, so the projection doesn’t allow for preparation for those events.
Finally, their last factor was that projections need to take uncertainty into account. Many decision makers are aware of the importance of uncertainty, but some projections may not make uncertainty estimates, which complicates planning for potential interactions between uncertainties.
Dealing with uncertainties
Their analysis identified the assumptions from each model so that decision makers could manage tradeoffs and expectations. “Assumptions need to be transparent so that your strategy is robust enough to plan for future challenges,” explained Srikrishnan. “You don’t want to rely on a very aggressive projection, or you risk overspending money that could have been used on other municipal needs. A suite of projections with transparent assumptions can help you plan for different eventualities.”
Many coastal projections provide discrete scenarios, like high, moderate, and low sea-level rise scenarios, but researchers often exclude probabilities for those scenarios, so they might appear as if they could be all as likely to occur.
Decision makers tend to prefer discrete scenarios because they are easier to budget for and understand, but Srikrishnan warns that they may not reveal the full picture. “It’s important for decision makers to understand if we believe that higher [storm surge] scenarios are more likely than lower [surge] scenarios. The distribution may not be like a bell curve,” he explained.
Srikrishnan thinks transparency is important in order to provide the best information to decision makers. He explained, “I tend to think the more transparent we are about assumptions and uncertainties, the better off everybody is.”
He sympathizes with the challenges that decision makers face. “Managers have to synthesize these projections and it’s hard to do that with a set of probability distributions. I understand why decision makers say, ‘just give us a number.’ There can be a lot of information overload.”
Understanding assumptions and the future for the Bay
Although their review only covered projections in Norfolk, Srikrishnan noted that their factors apply to anywhere coastal adaptation is required. “Decision makers all across the Bay have analogous problems and complications, and we need to give them the information to make decisions.”
Decision makers should continue to use multiple projections with different assumptions to best plan for the future. “There’s an open discussion that needs to be ongoing on how to help decision makers navigate that process and make it as smooth as possible, given the non-scientific constraints they have to work with,” explained Srikrishnan.
He added that their decisions can only be as strong as the information they’re provided by the projections. “It’s important to make decisions that are robust and to be able to adapt if current assumptions aren’t realized, especially when people’s lives are at play.”