Miami-Dade's Real-Time Monitoring Failed to Stop Zika Spread

As climate change pushes mosquitoes into new territory, expanding the range of dengue, Zika. West Nile virus, we assume better monitoring and faster alerts will let us stop outbreaks before they spread.

The unstated assumption runs deep. We can monitor our way out of this. That detection is the bottleneck—and if we just get the surveillance network right, public health officials will have what they need to act.

The 2016 Zika response in Miami-Dade County revealed something different. The county had real-time mosquito monitoring and knew where the virus was circulating, yet transmission continued. It only stopped when they shifted to aggressive source reduction, emptying standing water and spraying habitats.

The monitoring didn't fail because it was imprecise. It failed because it produced a false sense of control—officials watched the mosquitoes multiply while debating how to respond. By the time they moved to habitat intervention, the virus had already moved. This is not a technical problem. It's a governance problem dressed up as a measurement problem.

Observation without action is spectation.

The field has spent years building elaborate surveillance networks while the actual interventions—labor intensive, unglamorous, requiring repeated visits to private property—languish underfunded. Entomologists increasingly argue that vector control and habitat reduction must come first, monitoring second. But the policy world keeps optimizing the measurement because it's cheaper to buy sensors, easier to fund data infrastructure than to pay for the unglamorous work of actually removing breeding sites. Observation without action is spectation. It feels like preparation.

This matters beyond public health. It's the recurring pattern of how institutions respond to displacement and change. When things move faster than our ability to control them, we build better instruments to track the movement. We convince ourselves that precision observation creates optionality. The question becomes this. How do you know when your system for detecting a problem has become a substitute for solving it?