TER Data Processing

I obtained the following Youghiogheny River water temperature files from Steve Schreiner (sschreiner@versar.com) of Versar as approved by Bruce Michael (bruce.michael@maryland.gov) of DNR on 5/28/2019.

1. Brookfield-SangRunTemps-2015.xlsx
2. Deep_Creek_Sang_Run_River_Temperature_July_2017.xlsx
3. Figure2-1_Station_Map.pdf
4. Sang_Run_2018_July.xlsx
5. Sang_Run_August_2018.xlsx
6. Sang_Run_June_2018.xlsx
7. Sang_Run_River_Temp_June_2016.xls
8. Sang_Run_June_2017.xlsx
9. Sang_Run_Temperature_8-1-2017_through_8-31-2017.xlsx
10. Sang_Run_Temperature_for_August_2016.xlsx
11. Sang_Run_Temperature_July_2016.xlsx

The first objective was to make sure the data set is clean, that I can look at daily temperature profiles, and be able to select the most challenging parts of the problem.

I started with examining the first file. I converted it to a tab-delimited text file and wrote an R-script (actually adapted from an old script I had written several years ago) to extract the data and plotted the temperature profiles for each day. The script is “process_excel_filesV1.R”

I found that the next file was in a slightly different format. I replaced the header to be consistant with the one in the first file and delete some field markers (deleted apostrophies and commas), again to be consistent with the format of the first file. This allowed me to use almost the same R-script as the one used for file 1. The years in this file are expressed with 4 digits, as opposed to the first file which only used 2 digits. The script is “process_excel_filesV2.R” I also had to change the output file name to reflect the file only had data for a month.

All other files had a month’s worth of data. For all I had to slightly adjust the data to make them run in the same format and deleted extraneous data or redid the header fields. All relatively simple.

For file 11 I actually had to add a header to the data. I had to write three versions of the data processing program because one was for the file that had multiple months, one because the year in the date was expressed with two digits, and one to process files that had the year in four digits. The process was smooth. All of the data for the profiles have been stored as individual files for each day.

I also tallied all of the time intervals for which the temperature was 25C or above. The results are summarized in this Excel file.

These results can be summarized as follows:

1. I’ve cleaned all of the Brookfield temperature data as measured at the Sang Run River bridge from 2015-2018 and plotted the profiles for each day.
2. I also kept track as to how often the 25C limit was reached or exceeded. The table of those exceedances is referenced above.
3. Note that each exceedance interval is 10 minutes. From the period 2015-2018, 4 years, there were 35 days with exceedances with a total of 253 intervals.

 
That was a good start. Next is to obtain corresponding weather data. While there are a number of private/amateur weather stations in the Deep Creek area, most have only limited data. I’m looking for temperature trends in the area and have decided to focus on the data from Garrett County airport, which is an official weather site, and on Garrett College which has a station that has been operating for quite some time.