Communities
Improving Community Hurricane Resilience (Salisbury and Koeser)
When a hurricane barrels through a city, it doesn’t just shred shingles and snap power poles. The trees that line our streets and shade our homes often become both casualties and culprits—falling across roads, crushing cars, and knocking out power. Cleanup costs are staggering, but the real loss runs deeper: the ecosystem services those trees provide—shade, cooler air, stormwater absorption, beauty—can vanish overnight.
But here’s the hopeful twist: trees can be part of the solution, not just part of the problem. The key lies in species selection and long-term planning. Thanks to new research led by Allyson Salisbury and Andrew Koeser, we now have a stronger, science-based roadmap to guide which trees can help our communities weather the storm.
Tree Choice Matters
It turns out that not all trees bend—or break—the same way in a hurricane. For years, arborists and urban foresters have known this anecdotally and experienctially: live oaks stand strong while water oaks seem to shatter to pieces – sabal palms ride out storms better than queen palms. I recently read a meta-study from Salisbury & Koeser (2023) who took this intuition / experience that all arborists develop over years and decades, and put real data behind it!
They pulled together 58 studies across 15 countries and 42 hurricanes and tropical cyclones, creating one of the most comprehensive looks at storm-related tree damage to date. By analyzing how different species fared, they found that a handful of traits consistently predicted resilience:
- Wood density: denser wood = stronger tree.
- Maximum height: taller trees catch more wind, but the middle ground often fares best.
- Leaf mass per area (LMA): denser leaves can sometimes translate to better storm survival.
No single factor decides a tree’s fate, but taken together, these traits help paint a predictive picture.
Machine Learning Meets Arboriculture
Here’s where things get exciting (to this tree-geek at least!). The research team used a random forest machine learning model—an ensemble of hundreds of trees—to sort species into wind resistance categories.
The model was trained on older data from Florida and the Caribbean (Duryea et al., 2007) and then applied globally. Its accuracy? An impressive 91%. IYKYK … That’s about as good as it gets in ecological modeling!
With that power, they expanded the hurricane resistance ratings to 281 additional tree species, many of them commonly planted in cities across hurricane-prone regions of North America, Central America, Asia, Australia, and Oceania.
The results weren’t always comforting:
- 42% of species landed in the “Lowest” wind resistance category.
- Only 14% reached the “Highest.”
But even this sobering news is useful. It gives urban foresters a realistic way to assess vulnerabilities and plan smarter for the next storm.
Turning Data into Action
Data’s great, but only if you can use it. That’s why the researchers built the Estimating Tree Community Hurricane Resistance (ETCHR) Tool, a free Excel-based program.
Here’s how it works:
- A community uploads its tree inventory (species + quantities).
- The tool automatically classifies each species as Lowest, Medium Low, Medium High, or Highest resistance.
- The summary tab shows what proportion of the canopy falls into each category.
This gives a bird’s-eye view of how storm-ready—or storm-fragile—an urban forest really is.
Communities can then:
- Set canopy goals that prioritize stronger species near infrastructure.
- Flag at-risk species for pruning, cabling, or phased replacement.
- Refine planting lists so homeowners and developers don’t unknowingly choose the weakest performers.
Think of it as a hurricane stress test for your tree canopy.
Balancing Resilience and Diversity
It would be tempting to say: “Let’s only plant the strongest!” But that’s not the full story. Diversity still matters. An urban forest made up entirely of live oaks, for instance, might handle wind well—but it would be vulnerable to pests, diseases, or changing soil conditions.
As the authors point out, resilience is about balance:
- Plant high-resistance species along roads, near power lines, and next to homes.
- Reserve lower-resistance species for parks, natural areas, or spaces where failure is less of a hazard.
- Keep the overall mix broad enough to guard against future pests and climate uncertainties.
Why This Matters Now
Climate change isn’t making life easier. Models predict that tropical cyclones will grow more intense as the planet warms. At the same time, coastal populations continue to grow, putting more people, property, and infrastructure directly in harm’s way.
In Florida’s 2004–2005 hurricane seasons alone, tree debris cleanup averaged 448 cubic yards per mile of street, at a cost of $28 per cubic yard. Multiply that across multiple cities and you’re looking at billions of dollars.
If communities can shave even 10–20% off those losses through better tree planning, the savings are enormous—not just in dollars, but in lives, health, and ecosystem services preserved.
Final Takeaway: Smarter Forests, Stronger Communities
No city can hurricane-proof its trees entirely. But by blending field data, machine learning, and practical planning tools like ETCHR, communities can tilt the odds in their favor.
The choice of what we plant today—whether a storm-resilient oak or a fragile ornamental—shapes how our neighborhoods will look, feel, and function after the next storm.
Want to see how your own canopy stacks up? The ETCHR tool is free to explore on GitHub: Trees for Hurricanes Project.
Because when the winds roar, the best time to have chosen wisely was twenty years ago. The second-best time? Today.
Need an accurate interpretation of this data and guidance on taking steps to strengthen your community’s urban forest? Just fill out the form to schedule your initial Digital Arborist Consultation today!
<script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/iframe-resizer/3.5.14/iframeResizer.min.js"></script><iframe src="https://hello.dubsado.com/public/form/view/68870cca4b4f14d70e3b4873" frameborder="0" style="width:1px; min-width:100%;"></iframe><script type="text/javascript">setTimeout(function(){iFrameResize({checkOrigin: false, heightCalculationMethod: "taggedElement"});}, 30)</script>
