RaInDROP – About

Update written by Alyssa Griffin, Reviewed by Jared. H. Bowden

The North Carolina State Climate Office (SCO) Rainfall, Intensity, Duration, and Return for Observations and Projections Tool for North Carolina (RaInDROP) has recently been updated to include any HUC-10 water basins that cross or fall within NC. The inclusion of these water basins will assist partner agencies with watershed modeling that often incorporates precipitation data from neighboring states. In addition, new reference NOAA Atlas 14 Intensity-Duration-Frequency (IDF) precipitation values are scaled. Additional information can be found below.

Why is a new NOAA Atlas 14 IDF precipitation depth used?

The first version of RaInDROP utilized a NOAA Atlas 14 test file that was for a pilot project. The pilot project did some additional smoothing of the precipitation totals, leading to incorrect precipitation accumulations in the Southern Mountains region of NC (see Figure 1) where two different NOAA Atlas Volumes were being merged (Georgia and NC). The updated data more accurately portray what is listed on their point precipitation frequency estimates page.

What are the differences between the old and new NOAA Atlas 14 values?

The main difference is that precipitation depths in the southern part of the Southern Mountains have changed (increasing at some locations and decreasing in others) by 1.00-2.00’’. Elsewhere in the state, most precipitation depths have not changed more than 0.10’’ for a 24-hr event (Figure 2). However, there are some larger differences for the Coastal Plains regions when utilizing the 90th percentile precipitation projections (Figure 3). Lastly, the new NOAA Atlas 14 includes more points in NC, as the older smooth file had a slightly coarser grid resolution.

Why are there differences between the old and new NOAA Atlas 14 values?

The larger differences in the Southern Mountains region may be explained by the CONUS wide smoothing that was done in the test file used for a pilot project. The much smaller precipitation depth differences across the rest of the state may be attributed to the use of the Annual Maximum Series (AMS) in creating the IDF values rather than the Partial Duration Series (PDS). Please refer to the NOAA Atlas technical 14 document for additional details on these two time series.

How does incorporating additional water basins impact the scaled precipitation totals?

The scaled precipitation projections are based on climate division-wide scaling factors. These scaling factors are derived from CMIP5 global climate model output that project future precipitation trends based on different future climate scenarios (RCP4.5 and RCP8.5). Each climate division is assigned a scaling factor by applying L-moments regional frequency analysis. Three calculated moments (the mean, standard deviation, and asymmetry) are applied to a Generalized Extreme Value function, which calculates the average recurrence intervals (ARI; i.e. the 10-yr storm, the 100-yr storm, etc.) for the precipitation durations of interest. The scaling factor is defined as the ratio of the projected future ARI and the historical ARI for that location and are smoothed on a 1×1° grid to prevent arbitrary jumps in precipitation totals along climate division boundaries. Since some HUC-10 basins extend beyond NC, points outside of the state are assigned the scaling factor of the nearest climate division.

Introducing more points in the spatial averaging minimally affects scaling factor magnitude for both the mean scaling factors and 90th percentile scaling factors. Mean scaling factors change by less than 0.01 in the Mountains, Southern Piedmont, and North Coastal Plains, increase by 0.02 – 0.06 in the Northern Piedmont, Central Piedmont, and Central Coastal Plains regions and increase by 0.03 – 0.06 in the Southern Coastal Plains (Figure 4). These small differences in scaling factors have little impact on the precipitation depths associated with the mean scaling factor (Figure 2). Comparatively, the 90th percentile scaling factors are generally greater than the mean scaling factors, and so exhibit changes of greater magnitude – within 0.03 of their original value in the Mountains Region and most of the Piedmont, but increasing by 0.06 in parts of the Southern and Central Coastal Plains and decreasing by 0.03 in the Northern Coastal Plains (Figure 5).

Figure 1. The eight climate divisions of North Carolina.

Image of Atlas Differences — Figure 2. The differences in scaled precipitation totals using the mean scaling factor from the RCP8.5 climate forcing scenario at End of Century (2070-2099). Differences that are between -0.10’’ and 0.10’’ are shaded in white.

Image of Atlas Differences for RCP8.5 Mid century — Figure 3. Same as Figure 1 but for the 90th percentile scaling factors under RCP8.5 at Middle of Century (2040-2069).

Image of Atlas Differences for RCP8.5 End century — Figure 4. The difference in smoothed mean scaling factors for RCP8.5 at End of Century (2070-2099) between version 1 (original) and version 2 (new) of RaInDROP.