Flow Measurement Devices and Water Management for Summit Creek

By Brian M. Crookston

December 1, 2008

Term Project for

CEE 6440 – GIS in Water Resources – Fall 2008

Abstract

Efficient management of water resources rests on the foundation of accurate information. A recent study at the Utah Water Research Laboratory investigated flow measurement devices along Summit Creek, an irrigation canal that passes through the city limits of Smithfield, located in northern Utah. This term project makes use of the findings from that study to take a closer look at the water requirements and water consumption of Birch Creek Golf Course(located in Smithfield, Utah), which takes water directly from Summit Creek for irrigation purposes. This project made use of ArcGIS to compile useful geographic information of Summit Creek, the flow measurement structures found on the system, and the surrounding region. In addition, flow telemetry specific to the golf course was juxtaposed with estimated water requirements of the turf grass to draw conclusions regarding irrigation practices.

Introduction

Flow measurement devices have been used for centuries to quantify and distribute water resources. It is common for water user groups and irrigation companies to have devices to measure and record the quantities of water passing through the various streams, canals and rivers that comprise their conveyance system. Regular inspection and maintenance of each device is required to ensure accurate flow measurements, as a wide variety of influences can result in measurement inaccuracies and discrepancies. Examples include an incorrect installation or construction of a device, a misplaced staff gage or incorrect reference, a structure may have become unlevel, or sedimentation might be a recurring problem.

Recently, the Utah Water Research Laboratory (UWRL) began a study to determine the quality of flow measurement structures in Northern Utah. Summit Creek state distribution system was selected as a case study and five flow measurement devices were inspected for common installation and maintenance problems. A calibration was performed for each structure and any corrective actions were identified. The findings were reported to the water master in an effort to improve flow measurement accuracy and device maintenance.

The results of that study have been utilized by this project, which has investigated the water requirements and irrigation practices of the Birch Creek Golf Course. The area encompassing the hydraulic structures and structure locations is presented in Figure 1. Each device and the corresponding findings of the UWRL case study are presented in Table 1 (listed upstream to downstream), which include the new device accuracy utilizing the custom calibrations performed in the case study.

Figure 1. Satellite image of Summit Creek with flow measurement devices.

Table 1. Summary of inspected Summit Creek flow measurement devices.

Structure Name	Type	GPS Coordinates	Telemetry	Pre/Post Calibrated Device Accuracy	Corrective Actions
300 South	Parshall Flume (3 ft width)	4630992.689 Northing 432995.75 Easting	Yes	2.3% / 19.67%	None – device out of level and incorrect staff gage reference appear accounted for
Black Pipe	Parshall Flume (3 ft width)	4632163.246 Northing 432505.746 Easting	Yes	-6.91% / -3.15%	Yes – flume unlevel, incorrect staff gage elevation.
3 Creeks	Parshall Flume (8 ft width)	4632563.486 Northing 432529.028 Easting	No	-81.53% / -7.55%	Yes – Sedimentation at approach, eroded flume floor, non-uniform approach flow, submergence, incorrect staff gage elevation.
Armory	Sharp Crested Weir (7 ft width)	4631967.911 Northing 431079.478 Easting	No	-65.69% / -47.33%	Yes – flume unlevel, non-uniform approach flow, incorrect staff gage elevation, adjustable gate upstream of weir
Big Ditch	Cutthroat Flume (6 ft width)	4631853 Northing 430636 Easting	No	-25.67% / -21.73%	Yes – flume unlevel, incorrect staff gage elevation, eroded streambed upstream.

As illustrated in Table 1, four of the five measurement devices were found to increase in accuracy with minor corrections; larger corrective measures for Three Creeks, Armory, and Big Ditch should greatly improve flow measurement accuracy. These three devices are planned to receive telemetry after corrective measures are completed. Photographs of all five devices are presented in Figure 2.

(A) (B)

(C) (D) (E)

Figure 2. Photographs of 300 South (A), Black Pipe (B), Three Creeks (C), Armory (D), and Big Ditch (E)

300 South and Black Pipe flow measurement devices have telemetry; realtime data is made available by the Utah Division of Water Rights at www.waterrights.utah.gov. The measurements from these two devices are used to determine the amount of water that the Birch Creek Golf Course takes from the system (directly upstream of Black Pipe) for irrigation. The objectives of this project are to:

· Utilize ArcGIS to compile and process useful geographic information of Summit Creek and the adjacent region

· Estimate water requirements of the turf grass on the Birch Creek Golf Course

· Utilizing flow measurement device telemetry to estimate quantity of water diverted to the golf course

· Compare estimated water requirements and water usage

· Illustrate the importance of flow device accuracy in managing water resources

Procedure and Results

Data Acquisition and Geoprocessing

ArcGIS allows information and datasets from a variety of sources to be compiled into one place. Once assimilated, a wide variety of organizational and analytical tools are available to the user, including different visual tools to allow efficient and often visually appealing presentations of the gathered information and derived relationships. For this project, several data components were collected to create images of Summit Creek and the flow measurement devices; the compilation of information was also used in estimating water requirements of the golf course. The basmap first began by creating a Mircosoft Excel table which included all of the information available for each flow measurement device, excluding flow data. This table was uploaded into ArcGIS, resulting in the five devices defined geospatially. A digital elevation model (1/3 arc second resolution) of surrounding area was selected and downloaded from the USGS server. The GPS coordinates of the flow measurement devices are defined by the Utah Division of Water Rights in the UTM 12 (NAD 27) projection. All subsequent data sets and layers were projected or converted to this projection. The National Hydrolography Dataset Plus (NHD Plus) provides hydrologic datasets for the United States. Northern Utah is found in Region 16b, and catchment grids, catchment shapes, flowlines and attributes, and waterbodies were downloaded and added to the basemap. Furthermore, the Utah Climate Center (UCC) was accessed to download climate data, including temperature data, precipitation data, and ETo data. There was no station available in Smithfield; however, in close proximity and directly south a station at Utah State University was maintained. Data for the year 2008 was not available; therefore, a sampling of years 2005 to 2007 was used. It was important to determine the acreage of the Birch Creek Golf Course, therefore water related land use data was downloaded from the Utah GIS Portal, which was found to contain a fairly accurate area representation of the golf course. However, the entire area is not irrigated, therefore to further increase area estimation accuracy, a jpeg image was downloaded of the golf course from GoogleEarth, scaled and correctly located for comparisons. It should be noted that several of the datasets were very extensive, and therefore the entire basemap was trimmed to the area including and surrounding the five flow measurement stations. One representation of the basemap generated from the compiled data is presented in Figure 3.

Figure 3. Basemap image generated in ArcGIS 9.3

Upon closer examination, the catchments and flowlines downloaded from NHD Plus had several discrepancies. In ArcGIS, the ArcHydro toolbox was used to obtain a better relationship between catchments and flowlines. The DEM was reconditioned by burning in the streams and filling any sinks. Because it is not readily apparent which direction Summit Creek flows (North and then West), the flow directions were processed and represented by arrowheads placed upon each stream. The flow accumulation function was then used to determine the flow accumulation grid and illustrate how flow accumulates to Summit Creek. To determine the length of each stream reach between each flow measurement device, the streams were defined and segmented. This terrain process did not automatically recommend a stream threshold. Therefore several iterations were made to determine an appropriate threshold based upon the surrounding topography and to depict streams that may actually carry water for some period during the year. Coincidentally, the final selected threshold closely corresponded to the NHD Plus stream data. Finally, the catchment grid was delineated to determine which cell would correspond to which catchment and Catchment Polygon Processing was utilized to convert the information from raster to vector form, resulting in nice polygons representing each catchment. Although the catchments, water bodies, and other features obtained or generated were not used in later analyses, this information is important in two ways. Firstly, the images generated convey drainage information in a way that is easily comprehended. Secondly, the information would be useful to the irrigation company if sites for additional flow measurement devices needed to be determined.

Water Requirement Estimations

Good irrigation practices begin with understanding the crop water requirements and optimal times to irrigate. Due to the fact that 300 South and Black Pipe are used to measure water diverted to the Birch Creek Golf Course, this project was able to do analyses regarding estimated water requirements and water consumption of the golf course for irrigation. Estimates of the amount of water required by the turf grass began by determining the area irrigated utilizing ArcGIS and the distance between flow measurement devices to estimate seepage losses. In Figure 4, the water related land use parcel of the golf course is presented next to a satellite image of the same area.

Figure 4. Images used to estimate vegetation cover.

ArcGIS estimated the parcel to be 171.4 acres; after creating a polygon of only the 18 holes, the area was estimated to be approximately 118.3 acres, a 31% reduction. Following Summit Creek, the two flow measurement devices were measured to be exactly 1 mile apart. The climate data previously obtained from the UCC included temperature and reference ET data, which is presented in Figures 5 – 7.

Figure 5. Climate Data for 2005.

Figure 6. Climate Data for 2006

Figure 7. Climate Data for 2007

Due to the fact that climate data was not available for the 2008 irrigation season, data from the previous three years was collected and averaged, the total average is based upon the entire dataset, and is presented in Table 2.

Table 2. Summary of climate data.

Year	Month	Mean Max Temp	Mean Min Temp	Mean Eto	Mean Eto
		(°F)	(°F)	(mm/d)	(in/d)
2005	Sept	73.5	47.6	3.5	0.14
2005	Oct	61.8	40.0	2.0	0.08
2006	Sept	71.3	46.0	3.4	0.13
2006	Oct	57.4	37.7	1.8	0.07
2007	Sept	75.0	49.5	3.6	0.14
2007	Oct	58.9	38.1	1.8	0.07
Data Set	Sept	73.3	47.7	3.5	0.14
Average	Oct	59.4	38.6	1.9	0.07

For comparison to the reference ET obtained, a full Penman-Montieth evapotranspiration model was developed. This model required additional information to the temperature data, such as solar radiation data, which was found at a weather station located in Delta, Utah. Azmet evapotranspiration crop coefficients (http://ag.arizona.edu/azmet/et1.htm) specific to golf courses in warm and cool seasons were obtained (0.76 warm and 0.72 cool). The model is presented in Figure 8, and the resulting water requirement estimates for the golf course is presented in Table 3.

Figure 8. Penman-Montieth model

Table 3. Estimated water requirements for Birch Creek Golf Course

	Measured ET	Penman-Montieth ET	Water Quantity
	(mm/d)	(mm/d)	(cfs)
September	2.66	3.61	0.52-0.71
October	1.37	2.52	0.27-0.49

Actual Diverted Water Estimations

The flow rate data for 200 South and Black Pipe, obtained from the Utah Division of Water Rights for the 2008 irrigation season, was found to periodically contain erroneous entries and periodic errors in measurement. After a thorough cleanup, the data was aligned according to time. The 300 South telemetry logged a flow rate every 15 minutes, whereas the Black Pipe Telemetry had irregular intervals of recording flow measurements. The difference between each flow device was calculated, and the results were fit with a linear trend and a 1-day moving average. A one month period is presented in Figure 9.

Figure 9. Flow device telemetry and corresponding flow rate difference

From the resulting data set, the mean and various useful percentiles were calculated to discern approximate quantities of water that represent the amount removed from the golf course and lost to the system (seepage, etc). These results are presented in Table 4.

Table 4. Flow rate difference statistics

Measured Flow Rate Difference 300 South - Black Pipe
General Statistics (cfs)
Mean	10 Perc	25 Perc	50 perc	75 perc	90 perc
2.21	0.61	1.16	2.14	3.09	4.05

According to Aaron Hunt, who works at the State of Utah Division of Water Rights and is familiar with the flow measurement devices, the difference between 300 South and Black Pipe is used to determine how much water the golf course removes for irrigation. Minor losses are assumed, but according to him the golf course should only be taking ~2 cfs. A closer examination of the time series data gives several insights, presented in Figure 10.

Figure 10. Flow device telemetry and corresponding flow rate difference for 9/11 to 9/18

The data shows the fairly constant periods of flow passing Black Pipe. The effects of changes to the system can be seen in the larger valleys and peaks. Most likely, the golf course irrigates daily and withdraws what they feel is required quantities (understood that water is stored in ponds and then pumped into a sprinkler system) while ensuring that sufficient water is passed downstream to other water users. A lookup table of estimated flow losses based upon recommended losses in earth canals (Hill, 2000) was developed, presented as Table 5.

Table 5. Estimated losses for earthen canals

Estimated Losses Lookup Table
Base Flow	10% Loss	20% Loss	50% loss
(cfs)	(cfs)	(cfs)	(cfs)
2.0	0.2	0.40	1.00
4.0	0.4	0.80	2.00
6.0	0.6	1.20	3.00
8.0	0.8	1.60	4.00
10.0	1	2.00	5.00
12.0	1.2	2.40	6.00
14.0	1.4	2.80	7.00
16.0	1.6	3.20	8.00
18.0	1.8	3.60	9.00
20.0	2	4.00	10.00
22.0	2.2	4.40	11.00
24.0	2.4	4.80	12.00
†Per 1 mile length of canal

From this table and the analyses of the flow rate data, it appears that the 1-mile length of canal experiences losses on the order of 10%. However, this could easily be verified by a field crew. The final range of estimations for water consumption and crop requirements are summarized in Table 6, which are not completely dependent on total flow in Summit Creek.

Table 6. Final approximations

Flow Rate	ΔQ	Losses	ET	Excess	Percent
(cfs)	(cfs)	(10%)		Withdrawn	Excess
12-20	2-4	1.2-2	0.75	0-1.25	0%-31%

Device Accuracy and Water Management

This project has determined that when water is being removed from Summit Creek for irrigating Birch Creek Golf Course, a difference of approximately 2 to 4 cfs is measured. The exact quantity removed for irrigation may be approximately 1 to 2 cfs. However, one must question how accurate this information is and how uncertainty propagates in an irrigation system, before attempting to improve water management. For example, the 300 South measurement device is accurate to within 2.3%. The accuracy of Black Pipe has been improved from 6.9% to 3.2%. By utilizing the root-sum of squares to combine determine the uncertainty associated with water withdrawn between each device, if 4 cfs is measured, it is only accurate to within 3.9%. This is minor for this application, however, what about larger measured errors, as found with the remaining devices, applied to larger irrigated areas of land, such as the farmland surrounding the City of Smithfield. If the combined uncertainty was, for example, 25% (which is very plausible), that uncertainty will propagate to each outtake location between measurement structures. Each outtake may only take several cfs, but the error in measurement could accumulate to a significant discrepancy in actual quantities of water delivered. Unless a data instrument such as a weather station or flow meter is regularly maintained and calibrated, the information recorded may not merit a high level of confidence when improved water resource management practices are required due to increases in water demand or during seasons of water scarcity.

Conclusions

Geographic information of Summit Creek and five flow measurement devices were compiled into a basemap utilizing ArcGIS. Analyses regarding water requirements and water consumption of Birch Creek Golf Course were conducted, using geographic information, climate data, and flow telemetry. The effect of uncertainty propagation in an irrigation system was illustrated. Proper water resource management of irrigated farmlands surrounding Smithfield requires improved accuracy and telemetry for existing flow measurement devices, and possibly additional hydraulic structures. The case study of Birch Creek Golf Course could be used as an example to analyze this larger irrigation network (by means of ArcGIS) when the necessary data becomes available. The findings may indicate areas of efficient irrigating practices or areas where water management improvements are possible.

References and Bibliography

Aisenbrey, A. J. Jr., Hayes, R. B., Warren, H. J., Winsett, D. L., Young, R. B. (1978). Design of small canal structures. Denver, CO: U.S. Government Printing Office, 243-258.

Brown, Paul. Azmet Evapotranspiration Estimates: A Tool for Improving Water Management of Turfgrass. http://ag.arizona.edu/azmet/et1.htm

Climate Data Summary http://www.wrcc.dri.edu/cgi-bin/clilcd.pl?ut24127

Genovez, A., Abt, S., Florentin, B., Garton, A. (1993, November/December). Correction for settlement of parshall flume. Journal of Irrigation and Drainage Engineering, 1081-1091.

Global Warming/Climate Change for Logan Utah State University, Utah, USA. http://www.climate-charts.com/USA-Stations/UT/UT425186.php

Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results http://physics.nist.gov/Pubs/guidelines/sec5.html

Hill, Robert. 2000. How Well Does Your Irrigation Canal Hold Water. http://extension.usu.edu/files/publications/publication/ENGR_BIE_WM_03.pdf

Johnson, M. C. (2000). Discharge coefficient analysis for flat-topped and sharp-crested weirs. Irrigation Science, 133-137.

National Hydrography Dataset Plus. http://www.horizon-systems.com/nhdplus/

Personal Contact. Aaron Hunt, State of Utah Division of Water Rights. Nov. 25, 2008.

Personal Contact. Dr. Blake Tullis, Utah Water Research Laboratory and Utah State University. Nov. 25, 2008.

Personal Contact. Dr. David Tarboton, Utah State University. Nov. 24, 2008.

Utah Climate Center http://climate.usurf.usu.edu/products/data.php

Utah Division of Water Rights. http://www.waterrights.utah.gov/distinfo/realtime_info.asp

Utah GIS Portal: Vector GIS Data Layer Download Indexhttp://gis.utah.gov/sgid-vector-download/utah-sgid-vector-gis-data-layer-download-index

Utah Water Measurement Pocket Reference – Irrigation Water Measurement for Agriculture. (2000). Salt Lake City, Utah: Utah Association of Conservation Districts.

Station Graphs. Weather Station Delta Utah. http://www.wrcc.dri.edu/cgi-bin/wea_graph.pl?nvdelu

Tullis, P., Hunt, A. and Whyte, E. (unpublished). Increasing Data Accuracy, Reliability, Accessibility, and Understandability to Improve Basin-Wide Water Resources Decision Making.

The National Map Seamless Server. http://seamless.usgs.gov/website/seamless/viewer.htm

Water Measurement Manual – A Water Resources Technical Publication. (1997). Denver CO: U.S. Government Printing Office. 7-1 to 10-40.

Western Regional Climate Center http://www.wrcc.dri.edu/SOURCES.html