GIS in Water Resources
CEE 6440
(Fall 2010)
Precipitation and Runoff Generation
Chalk Creek Watershed, Utah
(Dec 6, 2008)
Bereket Tesfatsion
Submitted to:
Dr. David Tarboton.
Precipitation is an important element of the hydrologic cycle. It is a process where water is brought back on to the surface of the earth. On the other hand evaporation from the soil and open water bodies as well as evapotranspiration from plants are responsible to taking water into the atmosphere. Precipitation can take different forms such as rain, snow, hail, ice etc but all of these forms have a commonality in that they are all just water in its various forms. Some of the immediate hydrologic events that follow precipitation are runoff and seepage. Depending on the form of precipitation, the amount and timing of runoff could vary. Understanding how precipitation and runoff are related to each other, in a given area, is essential in water resources management. In this project, I attempt to explore the relationship between precipitation and runoff in Chalk Creek Watershed.
For this project I chose to work on the Weber Basin, located in northeast Utah. To conduct such kind of studies, the ideal situation is to find a watershed with sufficient precipitation data covering the entire area and a discharge data suitably located at the out let of the selected watershed. Identification of such a watershed within the Weber Basin was facilitated by mapping the watershed boundaries, precipitation stations and discharge gauging stations. Chalk Creek Watershed was selected for this study. Both precipitation and discharge data will be downloaded for the stations selected. To make a comparison with the discharge at the outlet on a consistent unit, precipitation data will be converted to what I would henceforth call Equivalent Flow in cfs. Using excel, the results will then been displayed graphically and the pattern of the cycles are analyzed and possible relationships between Equivalent Flow and discharge are explored.
For this study Chalk Creek Watershed shall mean all the drainage area draining to USGS10131000 station. Chalk Creek Watershed is relatively small watershed with an area of 159322 Acres. The NRCS website provides two SNOTEL stations situated on the southeast of the watershed. There are two USGS discharge gauging stations in the watershed. Only one of these stations is functional and will hence be used in this study.
Figure 1: Location Map for the Chalk Creek Watershed.
The data used for this project include, watershed boundary, flow lines, precipitation and discharge data. Note that all the data used are monthly averages and therefore the analysis will be based on a monthly time scale unless stated otherwise.
Watershed boundaries and flow line data were downloaded from the NHDPlus website. Chalk Creek Watershed lies within the Weber Basin. The Weber Basin is extracted from the Great Basin (Hydrologic Region 16) which is divided into two production units: 16a and 16 b. Weber Basin lies within the Great Basin (Hydrologic Region 16), unit b.
Monthly accumulated precipitation and SWEQ were acquired from the Natural Resources Conservation Service (NRCS) website. Currently there are only two precipitation stations (SNOTEL sites) within the Chalk Creek Watershed: SITENUMBER392 and SITENUMBER393. Time range for data range from the two sites is 1979 to present.
Monthly discharge is acquired from the USGS website. There are two discharge measurement stations within Chalk Creek Sub watershed: USGS10130700 and USGS10131000. USGS10130700 has apparently been abandoned and has 10 years data only (1964-1974). USGS10131000, which is suitably placed at the outlet of the watershed and has data record starting from 1927 till present, will be used in this study.
Discharge at the USGS station is provided in units of flow, in this case cfs. Therefore, in order for us to do a comparison between the average depth of precipitation that accumulates on the watershed under consideration and the discharge that leaves the watershed, we need to convert the average depth of precipitation to total volume of water. Since there were only two stations in the watershed, it was not necessary to do interpolation to estimate the average depth of precipitation. Therefore, in this case the average precipitation is the average of the two SNOTEL stations.
Total Volume of Water in the watershed = Average Precipitation x Watershed Area
The total volume of water on the watershed must now be divided by the period of time (in this case a month) to obtain an average monthly Equivalent Flow. Table 1 shown below summarizes the calculations which were done on Excel for a single year (1979).
|
Year |
Month |
Days |
Monthly
Precipitation, inches |
Average
Precip (inches) |
Total
Volume of Water in the Watershed(Acre-Feet) |
Equivalent
Flow(cfs) |
LN(Equivalent
Flow) |
Discharge
at USGS10131000 (cfs) |
|
|
SITE_392 |
SITE_393 |
||||||||
|
1979 |
10 |
31 |
0 |
0 |
0 |
0.00 |
1.0 |
0.0 |
10.7 |
|
1979 |
11 |
30 |
1.6 |
1.4 |
1.5 |
19915.25 |
334.7 |
5.8 |
15.2 |
|
1979 |
12 |
31 |
4.5 |
2.8 |
3.65 |
48460.44 |
788.1 |
6.7 |
12.9 |
|
1979 |
1 |
31 |
4.8 |
3.2 |
4 |
53107.33 |
863.7 |
6.8 |
19.3 |
|
1979 |
2 |
28 |
4 |
2.5 |
3.25 |
43149.71 |
777.0 |
6.7 |
20.3 |
|
1979 |
3 |
31 |
3.3 |
2 |
2.65 |
35183.61 |
572.2 |
6.3 |
29.7 |
|
1979 |
4 |
30 |
3.7 |
3.4 |
3.55 |
47132.76 |
792.1 |
6.7 |
59.7 |
|
1979 |
5 |
31 |
2.3 |
1.8 |
2.05 |
27217.51 |
442.7 |
6.1 |
126.2 |
|
1979 |
6 |
30 |
1.2 |
0.5 |
0.85 |
11285.31 |
189.7 |
5.2 |
48.6 |
|
1979 |
7 |
31 |
1.1 |
0.2 |
0.65 |
8629.94 |
140.4 |
4.9 |
13.5 |
|
1979 |
8 |
31 |
2.2 |
1.6 |
1.9 |
25225.98 |
410.3 |
6.0 |
27.1 |
|
1979 |
9 |
30 |
1.7 |
0.7 |
1.2 |
15932.20 |
267.7 |
5.6 |
9.5 |
Table 1: Sample calculation of Equivalent Flow
The Figure 2 below shows the periodic cycles of the Equivalent Flow and Discharge for the period under consideration. It is clear that the Equivalent Flow is much larger than the discharge but the two cycles are related.
By taking a closer look at the data (the first two years only) as shown in Figure 3, we see that the Equivalent Flow peaks between the months of January and February. On the other hand, the Discharge peaks around May.
Therefore, we have a pattern in that the Equivalent Flow peaking earlier by about 3 to 4 months than the Discharge. In the watershed under study, this pattern is consistent since we know that winter precipitation falls in the form of snow and does not melt substantially until around May.
Figure 2: Monthly cycles of Equivalent Flow and Discharge at USGS10131000
Figure 3: Depiction of time lag (about 3to 4 months) between the peaks of Equivalent Flow and Discharge at USGS10131000.
Now
we will try to see if there is any correlation between these two variables.
First a linear relationship is explored. Figure 4 shows that there no perfect
linear relationship between these two variables but that there is apositive
relationship between the two.
Figure 4: Linear relationship between Equivalent Flow and Discharge at USGS10131000
If
we transform the equivaletn flow using the natural logarithm, we find the
following relationship as shown in Fig 5.
Figure 5: Logarithmic Fitting between the Ln(Equivalent Flow, cfs) and Discharge at USGS10131000
The
logarithmic function used to fit the Ln(Equivalent Flow) with Discharge, seems
better in correlating high values of discharge. It perfoms poorly in
peredicting lower discharges, however. I have also tried to correlate these two
time series by shiffting the Discharge time series so that its peak matches the
peak of the Equivalent Flow. The results were not so different from the ones
shown above. Perhaps, these relatively poor correlations may be pointing to the
fact that precipitation and runoff events in this particular watershed may not
be correlated due to the gradual melt of snow and hence most of the discharge coming
from springs. In the following section we will explore how Annual Equivalent
Flow correlates Annual Discharge.
|
Year |
Annual Equivalent Flow(cfs) |
Annual Discharge(cfs) |
Runoff Coefficient |
|
1979 |
5579.45 |
392.70 |
0.07 |
|
1980 |
7565.75 |
977.40 |
0.13 |
|
1981 |
7024.95 |
697.20 |
0.10 |
|
1982 |
9880.00 |
1366.00 |
0.14 |
|
1983 |
8573.87 |
2158.90 |
0.25 |
|
1984 |
8694.50 |
1916.50 |
0.22 |
|
1985 |
7153.56 |
1326.50 |
0.19 |
|
1986 |
10424.93 |
2375.00 |
0.23 |
|
1987 |
6092.66 |
531.00 |
0.09 |
|
1988 |
5090.33 |
364.70 |
0.07 |
|
1989 |
7137.18 |
518.90 |
0.07 |
|
1990 |
6028.24 |
384.90 |
0.06 |
|
1991 |
7413.31 |
598.70 |
0.08 |
|
1992 |
5797.88 |
305.51 |
0.05 |
|
1993 |
9166.10 |
1354.50 |
0.15 |
|
1994 |
5978.86 |
487.80 |
0.08 |
|
1995 |
9154.79 |
1170.70 |
0.13 |
|
1996 |
7427.96 |
996.30 |
0.13 |
|
1997 |
9207.08 |
1297.30 |
0.14 |
|
1998 |
8176.15 |
1369.10 |
0.17 |
|
1999 |
8124.00 |
1267.70 |
0.16 |
|
2000 |
6143.16 |
416.40 |
0.07 |
|
2001 |
5702.45 |
377.55 |
0.07 |
|
2002 |
6198.62 |
395.97 |
0.06 |
|
2003 |
6780.13 |
418.46 |
0.06 |
|
2004 |
6182.39 |
369.24 |
0.06 |
|
2005 |
7977.14 |
990.30 |
0.12 |
|
2006 |
7766.53 |
762.70 |
0.10 |
|
2007 |
6941.23 |
573.70 |
0.08 |
|
2008 |
7839.83 |
827.90 |
0.11 |
Table 2: Annual Equivalent Flow,
Annual Discharge and Runoff Coefficient for Chalk Creek Watershed (1979-2008)
Figure 6: Linear relationship between Annual Equivalent Flow and Annual Discharge at USGS10131000
As shown in Figure 6 above
there is a fairly good linear fit between Annual Equivalent Flow and Annual
Discharge with an R2 of 0.7. This result may not be as useful as a
montly correlation but it establishes the observed annual relationship between
the two variables.
The following conclusion and observation can be made for Chalk Creek Watershed:
· There is a pattern between Discharge at the outlet and Equivalent Flow which is a derivative of precipitation on the watershed. A time lag of about 4 months between the peaks of the two variables has been observed.
· There is generally a positive relationship between monthly discharge and monthly Equivalent Flow but none of the simple mathematical models available in Excel seem to fit.
· Based on the annual time series, the average runoff coefficient for the watershed is about 11%, that is, only about 11% of precipitation in the watershed reaches USGS10131000. The rest of the water is lost in seepage and evapotranspiration.
· Annual Discharge correlates well with annual Equivalent Flow. This is an indication that if not sooner, precipitation has a direct influence on the discharge at a later time.
· Precipitation stations for this watershed have been very sparse. The precipitation data used in this study was compared with the NCDC data. Mostly, the precipitation obtained from the two SNOTEL stations was higher than a comparable data provided by NCDC.
1. Lecture Notes, and other document for GIS in Water Resources (CEE6440.)
2. NHDPlus website (http://www.horizon-systems.com/NHDPlus/data.php)
3. USGS website (http://www.usgs.gov/)
4. NRCS website (http://www.wcc.nrcs.usda.gov/snotel/Utah/utah.html)