Rainfall Basics: Using Data on Past Precipitation to Plan for the Future

As local governments plan infrastructure to address anticipated changes in flooding from rain, they face a challenge – protecting people and property while keeping costs down.  

Underestimating rainfall will result in flooding, but overestimating rainfall can create an unnecessary burden for taxpayers. Thankfully, scientific models can help turn high-quality, historical data on precipitation into estimates of flooding that are applicable to the future.  

So, where do modelers go to get the relevant information?  

The National Weather Service, a primary office within the National Oceanographic and Atmospheric Administration (NOAA), runs a program that documents accumulation of rain. The current dataset, called Atlas 14, provides data for much of the United States. 

In Florida, 242 stations are incorporated into Atlas 14. The data from these stations are run through a quality assurance and quality control process before being used to calculate depth-duration-frequency (DDF) curves for each station.  

The DDF curves document how much rain can be expected to accumulate at a location over a given period of time for a storm of a given intensity. In essence, they describe rainfall events by answering three questions: how much water will accumulate, how long does the event last, and how intense is the storm as indicated by how rarely it occurs? 

An example of the depth-duration-frequency (DDF) curves for 242 stations around Florida as part of the National Oceanographic and Atmospheric Administration’s (NOAA) Atlas 14 Program. The FLAMINGO-RS Station (above) is located in Homestead, Florida. (Source: https://hdsc.nws.noaa.gov/pfds/pfds_map_cont.html?bkmrk=fl)  

Tabular data and DDF curves can be downloaded from the Atlas 14 website. This information guides planning for infrastructure to mitigate flooding or maintain adequate supplies of water. 

A portion of Atlas 15, the successor to Atlas 14, was released by NOAA in early 2026. This initial release covers the state of Montana, and a second release, covering the contiguous United States, is anticipated in late 2026. The new atlas includes one important modification compared to Atlas 14: it incorporates future non-stationarity. In other words, it considers how future climatic conditions will shift the intensity of rainfall. This improvement will allow users of the data to adjust their estimates of accumulation of rain as they plan out to 2100.  

Currently, users can account for non-stationarity using change factors, like those released by the Flood Hub's Rainfall Workgroup in early 2026. Change factors act as multipliers that are applied to DDF curves to account for heavier rainfall arising from increasing global air temperature.  

The Atlas program has been the gold standard for estimating accumulation of rain since its inception in the early 1970s. Combined with factors to account for non-stationarity in Florida, these estimates yield information that enhances the ability of practitioners to safeguard lives, property, and the environment as they build infrastructure that is resilient to flooding.