Microwave data such as those from the DoD Special Sensor Microwave Imager Sounder are essential to hurricane forecasts, not a nice-to-have. They’re used in a variety of critical applications, including estimating hurricane intensity through AI-driven neural networks like the Deep Multispectral INtensity of TCs estimator or DMINT. In the absence of hurricane hunters, DMINT has been shown to be one of the most crucial tools in a hurricane forecaster’s arsenal for estimating storm intensity, largely because of the microwave data it utilizes.

Though other microwave data will still be available to forecasters, the DoD weather satellites comprise half of all microwave instruments, which means data availability will be sliced in half, greatly increasing the odds of missing rapid intensification episodes, underestimating intensity, or misplacing the storm and degrading forecast accuracy.