Hardware Ranch is a wildlife management unit located approximately 15 miles east of Hyrum, Utah in Blacksmith Fork canyon. The primary mission of Hardware Ranch is to enhance big game winter range and other critical habitats (Utah Division of Wildlife Resources 2006). The ranch also serves as a center for wildlife and habitat research and as a location for public wildlife viewing. The ranch is best known for its elk feeding program. During the winter the ranch is home to approximately 600 elk (Marnie Lee, personal communication, December 6, 2006).

Figure 1: Elk at Hardware Ranch

Source: Utah Division of Wildlife Resources, 2006

The elk are fed hay which is grown during the summer in two meadows near the near the ranch visitor’s center. Hardware Ranch offers sleigh rides to the public during the winter months which present an excellent opportunity to view elk up close.

Background of a Ranch Difficulty

Elk need a lot of food during the winter because of the cold temperatures. As an example, the elk herd at Jackson Hole was fed about 400 lbs of hay per elk for the winter of 1997. A quick calculation reveals that if the Hardware Ranch elk were fed at this rate, 240,000 lbs (120 tons) of hay would be needed. Figure 2 is a photograph of the elk being fed at Hardware Ranch.

Figure 2: Feeding of elk at Hardware Ranch

Source: Utah Division of Wildlife Resources, 2006

In past years, the hay grown at Hardware Ranch has generally been sufficient to feed the elk throughout the winter. However, hay yield has been declining for the past three years. This year, hay was bought in order to supplement the hay grown on-site. The decline in hay yield has been primarily attributed to the irrigation practices at Hardware Ranch.

Three specific problems have been identified:

1. High points in the two meadows used for growing hay

2. Poor ditch conditions

3. Lack of irrigation planning

The meadows at Hardware Ranch are flood irrigated. High points in the meadows would make uniform irrigation impossible without special considerations. The irrigation ditches at Hardware Ranch are in poor condition. Previously, there had been ditches throughout the interior of the upper meadow that helped irrigate more uniformly. These ditches have fallen into disrepair and are no longer used. Two primary ditches are currently used for irrigation of the upper meadow (Figure 3).

Figure 3: Map of upper meadow and ditches

Lack of irrigation planning has also adversely affected the hay yield at Hardware Ranch. In past years there were seasonal employees at Hardware Ranch that focused specifically on irrigation. The lack of a formal irrigation plan was overcome by investing a lot of time in irrigating. Implementation of a carefully developed plan would help optimize yield with minimal time spent irrigating.

Objective

The principal focus of this paper is to describe the analysis of the flow of water through the upper meadow. The analysis of flow through the upper meadow will be performed using ArcMap with the TauDEM toolbar plug-in.

The first objective is to identify the areas of Hardware Ranch that presently cannot be irrigated. After determining the areas that are not presently being watered, alternatives will be considered for getting water to the unirrigated land. Possibilities that will be considered include adding more irrigation culverts along the ditches and adding simplified network of ditches in the interior of the upper meadow. An irrigation plan will not be formulated at this time although suggestions will be presented on how to proceed with creating a plan.

Summary of Theory

“Drainage flows downhill” (Maidment, 2002). One of the simplest ideas in hydrology is that water flows downhill. This fundamental principle allows the delineation of watersheds using only elevation information. While irrigation is an artificial substitute for the natural processes that cause water to flow over the surface of the earth, the idea of water flowing downhill is still very applicable. After reviewing the needs of the project, it was determined that the downslope influence function of TauDEM would be used in order to determine the efficiency of the current layout of irrigation water sources in the upper meadow of Hardware Ranch. TauDEM is an acronym for Terrain Analysis Using Digital Elevation Models and was developed by Dr. Tarboton of Utah State University. One of the unique functions of TauDEM is the D_∞ flow direction grid. Standard flow direction grids are set up so that each grid cell has a specific outlet grid cell. The result is that water flows along a defined path with no possibility of dispersing over the landscape. The D_∞ flow direction grid of TauDEM differs from the standard flow direction grid in that TauDEM allows for dispersal of the drainage by proportioning water to the outlet grid cells as shown in Figure 1Figure 4.

Figure 4: TauDEM flow directions

Source: Tarboton, 2005

The downslope influence function is used by combining the D_∞ flow direction grid with a disturbance grid. In this study the disturbance grid represents the sources of irrigation water for the upper meadow of Hardware Ranch. Figure 5 illustrates how the D_∞ flow direction grid and the disturbance grid are combined to determine downslope influence. The grid cell outlined with a red box represents a source in the disturbance grid. It can be seen the value of ‘1’ in the disturbance grid is proportioned according the D_∞ flow direction grid.

Figure 5: TauDEM downslope influence function progression

Source: Tarboton 2005

It is important to note that while irrigation shares many similarities with natural drainage processes there are also some very important differences. One of the most important differences lies in the mode of delivery. The natural process for the delivery of water to the earth’s surface is rain. Compared to the irrigation sources considered in this study, rain differs in that it is delivered over a much larger area and then collects to form streams and rivers. The irrigation point sources supply water at a single point. The water then spreads out over the land before collecting once more. Because of these differences the results of this study should be applied with caution.

Solution Process

Data Collection

The digital elevation model (DEM) was obtained from the Utah Automated Geographic Reference Center (AGRC). This DEM was chosen for its 10-m grid cell resolution (AGRC 2006). There was no GIS data available that contained shapefiles for the upper meadow of Hardware Ranch. Similarly, the ditches and irrigation culverts were also not available from any GIS database. In order to obtain the necessary data, GPS survey equipment was used to obtain coordinates for the points of interest. In all, 130 points were surveyed at Hardware Ranch. Several different types of features were surveyed. First, the boundaries of the upper meadow were surveyed in order to know its extents. The primary irrigation ditches were surveyed in order to determine their location in relation to the upper meadow. Finally, the small culverts used to provide irrigation water sources for the meadow were surveyed in order to use their locations to create a shapefile of irrigation water sources.

Importing the Data

The DEM was downloaded from the AGRC website and converted to raster using the ‘DEM to Raster’ tool. The raster was then added to ArcMap using the ‘add data’ function. There were several steps in importing the GPS survey data. First, the survey data was downloaded from the data collector and inserted into a spreadsheet. Next, it was necessary to convert the degrees-minutes-seconds data to decimal degrees with four digits to the right of the decimal point. After the data preparation, the survey points were imported into ArcMap using the ‘Add X-Y Data’ function. At this point it was necessary to specify the spatial reference of the data, which was WGS 1984. A visual inspection revealed that the points were in the correct location.

Creation of the New Shapefiles

The survey data was used to create shapefiles for the upper meadow boundary, irrigation ditches, and the irrigation water sources. Each of the shapefiles was created by creating a new shapefile and then adding new features to the shapefile using the ‘Editor Toolbar’ in ArcMap. Figure 6 is an illustration of the shapefiles overlaying the survey points.

Figure 6: Illustration of the shapefiles created from survey points

Area and perimeter were calculated for the meadow boundary shapefile and length was calculated for each of the irrigation ditch shapefiles.

Preparation for Using the Downslope Influence Function

The TauDEM downslope influence function requires two inputs: a D_∞ flow direction grid and a disturbance grid. The D_∞ flow direction grid was obtained by performing the TauDEM ‘Basic Grid Analysis’. For this study the disturbance grid was used to indicate the location of irrigation water sources. The grid cells representing irrigation water sources were required to have a value of ‘1’ and the value for all other grid cells were required to be ‘0’. The input disturbance grid was obtained by manipulating the point shapefile of irrigation water sources. The first step was to create a new field in the attribute table of the irrigation water sources shapefile. All values in the new field were set to ‘1’. Next, the shapefile was converted to a grid being careful to set the value field for the new grid to the field of 1’s just created in the irrigation water sources shapefile. It was also important to set the extent of the new grid equal to the extent of the D_∞ flow direction grid. The final step in creating the disturbance grid was to reclassify the ‘no value’ grid cells of the newly created grid to value = ‘0’. The downslope influence was then calculated using the disturbance grid and D_∞ flow direction grid. This process was repeated using the irrigation ditch polyline shapefile to create a new disturbance grid. This modification had the affect of using the two ditches as line sources for irrigation water.

Data Manipulation

It was necessary to crop downslope influence grids in order to use the grid values in analyzing the upper meadow. This was done through the use of the raster calculator. When rasters with different extents are added together the resulting raster has the same extent as the smaller of the two rasters. One of the disadvantages in using this method to crop rasters is that interpolation of the grid cells will occur if the grid cells of each grid are not aligned. In order to perform the raster calculation, the shapefile of the upper meadow boundary was converted to a grid with the value for each grid cell equal to ‘0’. Adding the newly formed grid to the downslope influence grid resulted in a downslope influence map cropped to the shape of the upper meadow. The final step in preparing the data was to discretize the grid values by using the reclassify tool. This enabled the use of the attribute table by reducing the number of entries in the table. The number of cells with zero downslope influence was then compared to the total number of grid cells and the percentage of grid cells with zero downslope influence was calculated. Figure 7 is an illustration of the process of data manipulation.

Figure 7: Process of data manipulation (results shown were using point)

Results

General Meadow Information

Creating shapefiles of the GIS information allowed calculation of several geometric attributes. The area of the upper meadow was found to be 329,000-m² (81.3 acres). The perimeter of the meadow was calculated to be 3.08-km (1.91 miles). The two primary irrigation ditches were found to have lengths of 1.5-km (0.92 miles) and 0.21 km (0.13 miles). A total of 20 points were inputted as irrigation point sources.

Downslope Influence Results

The discretized map of downslope influence using the irrigation culverts as point sources is shown in Figure 8. The purple areas have the highest downslope influence and the orange areas the lowest downstream inflence.

Figure 8: Discretized downslope inflence

Figure 9 is a map of the discretized downslope influence using the irrigation ditches as line sources. As in the previous map, the purple areas have the highest downslope influence and the orange areas have the lowest downslope influence.

Figure 9: Discretized downslope influence using line sources

Table 1 is a summary of the downslope influence results.

Table 1: Summary of downslope influence

Case	Total Grid Cells	Grid Cells with Downslope Influence Equal to '0'	Area with Downslope Influence equal to '0' (m²)	Percentage of Area with Downslope Influence equal to '0'
Point Sources	3290	1081	108,100	32.9%
Line Sources	3290	562	56,200	17.1%

It can be seen that the analysis using the irrigation culverts as point sources predicted 32.9% of the total ranch area would not be irrigated by water flowing from the irrigation culverts. Using the irrigation ditches as lines sources increased the area of irrigated land but still left 17.1% unirrigated.

Discussion

The analysis indicates that approximately 17% of the total ranch area that is not currently irrigated could be irrigated by adding more irrigation points along the two ditches. The area of the ranch with zero downslope influence decreased by 51,900-m² by using the ditches as line sources for irrigation water as opposed to using the culverts as point sources. This represents a decrease of nearly 50% in unirrigated area. Using the irrigation ditches as line sources represents the maximum amount of land that could be irrigated in the upper meadow by adding more irrigation points along the ditches.

Consideration of the DEM indicated that contrary to previous beliefs, high points in the upper meadow did not significantly reduce the area of irrigated land. Slope was found to by the limiting factor rather than elevation. The dry areas did not receive water because the slope of the meadow resulted in water flow over different areas. With that finding, two options were considered for further increasing the area of irrigated land in the upper meadow of Hardware Ranch. The first option was explored by adding an additional irrigation point source as shown in Figure 10.

Figure 10: Altered disturbance grid (shown in red) with downslope influence grid (point source)

Figure 11 is a map of the discretized downslope influence calculated using the new disturbance grid.

Figure 11: Discretized downslope influence using new point source disturbance grid

After adding the new irrigation point it was found that the area of unirrigated land in the upper meadow decreased to 86,500-m², a decrease of 21,600-m² as compared to the original results using point sources for irrigation. The results of this solution could be obtained by adding a ditch or pipe that would transport water to the location where the irrigation point was added. The percentage of the total area of the upper meadow with zero downslope influence for this case is 26%.

A similar analysis was performed using the disturbance grid formed using the ditch polyline shapefile.

Figure 12: Altered disturbance grid (shown in green) with downslope influence grid (line source)

Figure 13 is a map of the discretized downslope influence calculated using the new disturbance grid.

Figure 13: Discretized downslope influence using line source disturbance grid

Adding the new irrigation line source decreased the unirrigated area to 30,500-m², a decrease of 25,700-m² as compared to the original results using line sources for irrigation. This solution could be obtained by adding a ditch or pipe to transport water to the location where the line source was added and adding a device for diffusing the water at that location. The percentage of the total area of the upper meadow with zero downslope influence for this case is 9.3%.

The final large region of the upper meadow without water is shown in Figure 14. No easy solution for irrigating this area of the meadow has been identified. This area is raised in elevation and would require substantial expansion of the irrigation ditch netork. Because digging new ditches was considered cost prohibitive, this option was not analyzed.

Figure 14: Final large region with no water

Conclusion

The largest area of irrigated land was added by considering the irrigation ditches as line sources. Using the line sources decreased the unirrigated land in the upper meadow of Hardware Ranch by nearly 50%. Additional land could be irrigated by using a ditch or pipe to transport water to the interior of the meadow. An additional source of water placed at the head of the large section of unirrigated land in the interior of the meadow could further decrease the area of unirrigated land by over 20,000-m². Finally, no easy solution for irrigating the section of unirrigated land along the western edge of the meadow was identified.

Further Research

There are two primary subjects of particular interest for further research: the accuracy of the downslope influence function in predicting irrigated land and the development of an irrigation plan. The accuracy of the downslope influence function could be determined by performing testing in the meadow. Surveying the areas of land irrigated by specific irrigation culverts and comparing the survey results to the areas predicted by the downslope influence function would give a good basis for determining the accuracy of the downslope influence function. One option for developing an irrigation plan is to consider the respective areas irrigated by each irrigation culvert. A plan could then be developed by weighting the time of irrigation at each culvert based on the area for which each culvert provides water.

References

Arc Hydro: GIS for Water Resources. (2002). D. R. Maidment ed., ESRI Press, Redland, Ca.

Tarboton, D. G. (2005). “Terrain Analysis Using Digital Elevation Models (TauDEM).” <http://hydrology.neng.usu.edu/taudem/> (Dec. 2, 2006)

Utah Division of Wildlife Resources. (2006). “About Hardware Ranch.” <http://www.wildlife.utah.gov/hardwareranch/about.php> (Dec. 2, 2006)