Redeveloping the EPA/FHWA Synoptic Statistical Rainfall Analysis Program for use with new National Climactic Data Center hourly rainfall formats and to produce statistics that can be used to evaluate potential effects of climate change on hourly precipitat
Focus Area
Water Quality/Wetlands
Subcommittee
Natural Resources
Status
Archived
Cost
$250k-$499k
Timeframe
2-3 years
Research Idea Scope
The U.S. Environmental Protection Agency (USEPA), the Federal Highway Administration, the U.S. Geological Survey and Several National Cooperative Research Program studies use the USEPA definition of a synoptic storm event to evaluate and simulate potential effects of stormwater runoff-quality on receiving waters (USEPA, 1976; Strecker and others, 1989; Driscoll and others, 1989; Driscoll and others, 1990; National Cooperative Highway Research Program, 2006; National Research Council, 2008; Granato, 2010a, Taoylor and others, 2014). The SYNOP program was written in the mid-1970’s (USEPA, 1976, Driscoll and others, 1989) and updated to address Year-2000 issues in 2001 (Tarig Omer, HydroQual, written comun. 2003). In 1990, the FHWA published a personal-computer version of the program with a preprocessor named “SYNPREP” to reformat newer precipitation file formats (Driscoll and others, 1990). In 2010, FHWA published a batch-processing Windows graphical-user-interface shell named the SYNOP precipitation analysis preprocessor (SPAF) and included a compilation of the previous reports and FORTRAN software on CD-ROM (Granato, 2010a). Granato (2010a) used SPAF to run SYNPREP and SYNOP in batch mode to calculate storm-event precipitation statistics with data from 2,610 National Oceanic and Atmospheric Administration hourly-precipitation data stations in the conterminous United States with at least 25 years of data during the 1965-2006 period. Granato (2010a) compared the previous National study (Driscoll and others, 1989) values to the new statistics and did not find significant differences, but a large portion of the periods of records used may have overlapped for stations that were discontinued in the latter half of the time range. The primary outputs from SYNOP include the average and coefficient of variation of the storm-event duration, intensity, volume, and time between storm-event midpoints for various time periods and provides the annual average number of storms per year (Strecker and others, 1989; Driscoll and others, 1989; Granato, 2010a). Granato (2010b) evaluated prior methods to evaluate the Poisson process, probability distributions for modeling storm-event statistics, and storm-event hyetographs that could be used for characterizing storm-event precipitation statistics as part of the Stochastic Empirical Loading and Dilution Model (SELDM) development process (Granato, 2013). Currently (2019) there are two problems with the available programs for calculation runoff-generating event statistics. The first problem is that the National Climatic Data Center has discontinued the text-file formats that were needed to run SYNOP by using SYNPREP and/or SPAF. It has been more than a decade since the end of the last national analysis (2006) and the scientific consensus indicates that climate change will affect the statistical distribution of precipitation as the frequency, intensity, and variability of precipitation changes with temperature and other variables. The second problem is that current outputs from these programs do not support analysis of precipitation statistics with more sophisticated probability distributions than the one- or two-parameter exponential distributions currently used by stormwater practitioners (Granato, 2010b; 2013). These problems are compounded by the fact that SYNOP and SYNPREP were written in an archaic version of FOTRAN and SPAF was written in Visual Basic 6.0, which is no longer supported by Microsoft. Software compilers for the source code for these programs may not be available for current or future updates.
Urgency and Payoff
The objectives of this research would be to redevelop the SPAF-SYNPREP-SYNOP suite of programs into a single integrated computer program that provides a variety of statistics to facilitate current and future precipitation analyses. To meet these objectives it would be necessary to: • Identify a stable National Climatic Data Center format for hourly precipitation data; • Identify other commonly used hourly-precipitation data formats and/or sub-hourly data formats so that long-term precipitation data collected by other State or National agencies may be used to calculate more site-specific data. • Do research to identify and adopt a commonly used, stable, and easy-to-follow computer programming language, this language should support open-source code-sharing so that the program could be adopted and maintained by multiple organizations in the future; • Reverse engineer the algorithms used by each program by using the original code files most recently published by Granato (2010a); • Research the literature on the potential effects of climate change on storm event characteristics to identify the statistical distributions and therefore statistics that may be needed for current and future analysis. • Redevelop the program(s) with open-source coding and rigorus code-documentation protocols (for example Flynn and others, 1994). • Verify that the new code will reproduce the values calculated by the original SPAF-SYNPREP-SYNOP suite of programs and verify that any new statistics are properly calculated. • Obtain and analyze National Climatic Data Center data for thousands of hourly precipitation data stations • Examine methods such as the PRISM method to create GIS coverages of the SYNOP statistics from point at-station data. • Examine existing U.S. EPA Rain Zones and evaluate potential alternatives for level I regional analysis. • Document the methods, open-source computer code, compiled computer programs, and interpretive results in the public domain.
Submitted
06/02/2020