Nearly half of Michigan’s 83 counties were under a state of emergency by April 20, 2026. Record rainfall stacked on top of a historic snowpack pushed rivers over their banks, overwhelmed storm drains, and forced residents downstream of century-old dams to evacuate. That’s not a slow-moving natural disaster. That’s a crisis.

Cheboygan County, close to the tip of the Lower Peninsula, took some of the worst of it. Michigan State Police documented the damage block by block: stressed infrastructure, surging water, neighborhoods cut off. In the city of Cheboygan itself, emergency crews hauled in industrial pumps to relieve pressure on a dam that’s been standing for more than 100 years, a structure that was never engineered to move water at this volume. It’s still standing. Whether it was designed to handle a 2026 rain-on-snow event is a different question entirely.

Richard B. Rood, professor emeritus of meteorology at the University of Michigan, has spent his career studying how communities adapt when the climate their infrastructure was built for disappears. He wasn’t surprised by what happened across Michigan and Wisconsin. “Disasters like the one Michigan is experiencing are setting higher benchmarks for safety as communities plan future infrastructure,” Rood said.

The flooding had two inputs, and they didn’t arrive together by coincidence. Michigan, Wisconsin, and parts of northern Illinois recorded their wettest March and April on record. A March blizzard dumped three feet of snow across stretches of Michigan. Then the rains came in mid-April.

Rain on frozen ground.

Forecasters call it a rain-on-snow event. The snowpack can’t absorb rainfall once it’s saturated. Water doesn’t soak in. It runs. It carries ice with it, and that ice doesn’t just raise water levels, it batters shorelines, tears into aging infrastructure, and destroys lakefront cabins and homes that were built close to the water back when flooding at this scale wasn’t something anyone was planning around.

The moisture feeding those storms came from the Gulf of Mexico, pulled north by a high-pressure system sitting over the southeastern United States. Michigan Advance’s coverage of the flooding traces how that atmospheric pattern turned two weeks of storms into a compounding, sustained disaster rather than a single bad weekend.

The dams weren’t built for this.

Michigan’s inland lakes are held back by a patchwork of small dams, many of them privately owned, many of them more than 100 years old, and some of them effectively invisible until a flood forces a reckoning. These structures were designed for a climate that doesn’t exist anymore.

Rood’s research documents a measurable shift in the Cheboygan area’s winter temperatures. During the baseline period from 1991 to 2020, the region averaged 10 days at or near freezing in March alone. Go back further, to the climate conditions that existed when most of these dams were engineered, and those numbers look completely different. In 1958, planners weren’t designing for the freeze-thaw cycles that are now the norm. They weren’t designing for snowpack that partially melts, refreezes unpredictably, then catches a major rainstorm on top.

Data tracked by NOAA has documented the long-term shift in Great Lakes regional precipitation patterns. That data doesn’t just show more rain. It shows rain arriving differently, concentrated in fewer, more intense events. Wet Marches. Ice-heavy snowpacks followed by 50-degree April days and then another storm system rolling in before the ground can drain.

Michigan’s state dam safety records show hundreds of structures across the Lower Peninsula, many of them rated as high-hazard, meaning failure would threaten lives downstream. The $200 million question isn’t whether these dams will face more events like April 2026. They will. The question is whether the state can fund inspection, reinforcement, or removal fast enough to matter.

Twenty-three counties under a state of emergency. Evacuation orders near 100-year-old dams. Emergency pumps running around the clock in Cheboygan. That’s where Michigan is right now, and Rood’s point is that the benchmarks communities used to plan their infrastructure, the assumptions baked into every spillway and drainage calculation back to 1958, those assumptions are gone.