USGS North Carolina Water Science Center Publication
Complete report in PDF (64 pages, 2.7 MB)
A study of annual maximum precipitation frequency in Mecklenburg County, North Carolina, was conducted to characterize the frequency of precipitation at sites having at least 10 years of precipitation record. Precipitation-frequency studies provide information about the occurrence of precipitation amounts for given durations (for example, 1 hour or 24 hours) that can be expected to occur within a specified recurrence interval (expressed in years). In this study, annual maximum precipitation totals were determined for durations of 15 and 30 minutes; 1, 2, 3, 6, 12, and 24 hours; and for recurrence intervals of 2, 5, 10, 25, 50, 100, and 500 years.
Precipitation data collected by the U.S. Geological Survey network of raingages in the city of Charlotte and Mecklenburg County were analyzed for this study. In September 2004, more than 70 precipitation sites were in operation; 27 of these sites had at least 10 years of record, which is the minimum record typically required in frequency studies. Missing record at one site, however, resulted in its removal from the dataset. Two datasets—the Charlotte Raingage Network (CRN) initial and CRN modified datasets—were developed from the U.S. Geological Survey data, which represented relatively short periods of record (10 and 11 years). The CRN initial dataset included very high precipitation totals from two storms that caused severe flooding in areas of the city and county in August 1995 and July 1997, which could significantly influence the statistical results. The CRN modified dataset excluded the highest precipitation totals from these two storms but included the second highest totals.
Two additional datasets were developed using National Oceanic and Atmospheric Administration (NOAA) records from eight selected National Weather Service precipitation stations in or near Mecklenburg County. These datasets were included in this investigation in order to analyze long-term records of precipitation data in the vicinity of Mecklenburg County and to provide an overall quality-assurance check of results of the L-moment method that was applied to all datasets. The periods of record for the quarterly data (15-, 30-, and 60-minute durations) ranged from 24 to 33 years at the eight National Weather Service stations, and the periods of record for the hourly data (1-, 2-, 3-, 6-, 12-, and 24-hour durations) generally were about 55 years. One of the NOAA datasets consisted only of precipitation totals from the National Weather Service stations. The other NOAA dataset consisted of the eight National Weather Service stations combined with an aggregated U.S. Geological Survey site represented by using regional (county) L-moment statistics to simulate a set of statistics for one site (as though only one U.S. Geological Survey site had been in operation).
The L-moment method used in this report is the collective group of statistical techniques used in the analyses of annual maximum precipitation totals to compute the regional weighted L-moment statistics for each duration. The L-moment statistical calculations included the mean, L-scale, coefficient of L-variation, L-skew, and L-kurtosis. L-moment ratio diagrams of the site and regional L-skew and L-kurtosis statistics plotted against selected theoretical probability distributions resulted in the use of the generalized logistic and extreme-value distributions as the most appropriate for estimating precipitation frequency. Precipitation-frequency estimates were provided as depth-duration values, in inches, and intensity-duration values, in inches per hour. Values also were presented as precipitation-frequency curves to assess the differences in results from the four datasets. For comparative purposes, tabulated and graphical frequency information presented in this report include values determined from previous and similar studies completed by the National Weather Service.
Precipitation-frequency curves were examined in the order of ascending recurrence intervals, and the spread between the curves for the CRN initial and CRN modified datasets generally increased. Beginning with the 60-minute duration and 5-year recurrence interval, percentage differences between many of the corresponding depths in the two datasets exceeded 15 percent, indicating that the large annual maximum precipitation totals for 1995 and 1997 exerted a high degree of influence on the results. The precipitation depths for the NOAA datasets exhibited a smaller degree of spread than the CRN datasets and generally tended to straddle the midrange between the CRN datasets. Comparisons of precipitation depths between the combined NOAA and aggregated U.S. Geological Survey dataset and the CRN initial dataset indicate that percentage differences generally were less than 15 percent for durations less than 6 hours and for recurrence intervals less than 25 years. The estimated 24-hour, 100-year precipitation depth of 10.99 inches for the CRN initial dataset is about 51 percent higher than the corresponding value of 7.29 inches for the combined NOAA and aggregated U.S. Geological Survey dataset.
The small sample available from the U.S. Geological Survey precipitation network, which was influenced by the occurrence of two large observations (1995 and 1997), indicates that frequency characteristics may be biased higher than actual values. Resolution of this question can only be accomplished with additional data. In the absence of further analyses, results from the combined NOAA and aggregated U.S. Geological Survey dataset currently (2005) can be considered appropriate for use in deriving frequency characteristics for Mecklenburg County. This combined dataset provides consideration of the substantial amount of precipitation data, including some large rainfalls that have occurred in the county, and the NOAA precipitation data representing longer periods of record.
Download the free PDF reader.
| For more information, contact | To order printed copies, contact |
| North Carolina District U.S. Geological Survey 3916 Sunset Ridge Road Raleigh, North Carolina 27607 (919) 571-4000 |
U.S. Geological Survey Information Services Box 25286, Federal Center Denver, CO 80225 1-888-ASK USGS |
Contact Information: North Carolina Web Development Team
U.S. Department of the Interior | U.S. Geological Survey
URL: https://pubs.water.usgs.gov/sir2006-5017
Page Contact Information: Publications Team
Page Last Modified: Thursday, 01-Dec-2016 19:01:32 EST
