Final Project


Hydrology and GIS in the Sevilleta National Wildlife Refuge


Motivation and Background

Within the semiarid American Southwest, the Rio Grande is well-studied as a regionally important water resource, but the smaller springs that surface within the Rio Grande rift are also a vital resource for irrigation, livestock, and wildlife. Most of theses springs are poorly characterized, both geologically and hydrologically. The Sevilleta National Wildlife Refuge (NWR) is an ideal locality (Figure 1) to study the complex hydrology and hydrochemical mixing of water chemistries in an area free from anthropogenic pressures. Part of my graduate research at UNM is to geochemically characterize the springs and wells of the Sevilleta NWR to develop a better understanding of the hydrologic controls in this complex area. The Sevilleta NWR is located approximately 80 kilometers south of Albuquerque, New Mexico (Figure 2). It is located east of the Colorado Plateau and west of the Great Plains within the actively rifting Rio Grande Valley, at the intersection of 1) the Albuquerque and Socorro basins and 2) four major biotic zones.

Objectives

The objectives of this project are to 1) evaluate the stream networks and watersheds that could potentially influence surface water flow of springs and streams in the Sevilleta, 2) estimate Sevilleta evapotranspiration using the Jensen-Haise Radiation Method and meterological data collected from sites around the NWR, and 3) evaluate the terrain in 3-D to project difficulty in spring sample reconnaissance.

Methods

Data sources included the USGS Seamless website (http://seamless.usgs.gov), the NM Resource Geographic Information System Program (http://rgis.unm.edu), data collected for my thesis (Williams M.S. Thesis data), Sevilleta Meteorological data (http://sev.lternet.edu, Data.Tabular. Meteorological Data, Doug Moore-PI), and Environmental Hydrology (A. Ward and S. Trimble. 2nd ed., Boca Raton: Lewis Publishers, 2004. ) I used the ArcInfo Licensing Level to utilize as many extensions as possible. These included Spatial Analyst, Animation, and 3-D Analyst. I used NAD83 as my datum for all layers in this project. As most of the data I collected from the web was in NAD83, it was easier to use this projection and change those few that were originally in NAD27. I used UTM zone 13N (with the median set at -106.5) as my projection for all data because the Sevilleta is located at about latitude -106.5, and a UTM projection best represents a long vertical area such as the Rio Grande rift with a short horizontal extent, such as the Sevilleta.

Results

Sevilleta Hydrology
Using the hydrology toolset, I delineated several watersheds of different drainage areas (>5 km2, >20 km2, >30 km2, and >40 km2), to determine which area would best group and define each of my spring systems. I decided on the > 30 km2 drainage area, because it separated my springs into 5 groups defined by watersheds (Figure 3). In the process of delineating the watersheds, I had to produce a stream network (Figure 4) that can be used to identify contributing stream sources. This network is useful for analyzing a geochemical mixing trend that may represent both a surficial input and a spring input. The headward extent of these streams, as indicated by the stream network, allow me to analyze where a stream water sample could be interfacing with the chemistry of a deeply sourced spring. Finally, I produced a contour map (Figure 5) with Spatial Analyst to complement the other hydrology analyses and identify areas of high slope and potentially high flow during precipitation.

Evapotranspiration
I used the Jensen-Haise Radiation Method to estimate the evapotranspiration in the Sevilleta NWR using meteorological data collected by Sevilleta personnel at ten Met Stations across the Sevilleta. The controlling equation is , where CT is the temperature coefficient, T is the mean temperature, TX is the intercept of the T axis, RS is the solar radiation, . is the latent heat of vaporization, H is meters above mean sea level, and e2 and e1 are the saturated vapor pressure at mean maximum and minimum temperatures. The following supporting equations were used to calculate several of these variables: , , , , and With an estimated ET value for each met station from the years 2002 through 2007, I was able to create a color contour map for each year. High values of ET (up to 0.396 mm day -1) were represented from light to dark blue, while low values of ET (down to 0.144 mm day -1) were represented with red. Played in chronological order, the changes in ET become apparent. ET on the Sevilleta is typically high (0.3 range), but certain met stations report moderate to low ET values in the six years studied (Figure 6). Notably, met stations 50 and 49 have high variability between low and high ET values, while the other stations report relatively high ET values. It is important to note that ET studies by the Bosque are typically much higher, but they are not represented in this series of contour models because the Bosque study utilizes the Eddy covariance method. Due to the limited meteorological data available on the rest of the Sevilleta, I chose the Jensen-Haise Method, and did not represent the Bosque data, so as to not mix two different ET estimating methods.

3-D Terrain Evaluation
I used the 3-D rendering capabilities of ArcScene to .fly-through. the Sevilleta to evaluate areas where terrain may inhibit my ability to access springs. Comparison between topographic maps of the area and the rendering in ArcScene allowed me to determine that the topography in the southwestern corner near the Lemitar Mountains may limit some access to the San Lorenzo Springs. Springs on Ladron Peak in the northwest corner may also be quite hard to access due to steep slopes. Both topography and saturated sand may limit access to the Rio Salado Springs on the far west side, but springs and wells on the east side near the Los Pinos Mountains may be easier to access due to a good road system.

Conclusions

During the course of this project, I have found that the suite of ArcGIS tools can be useful for a wide range of tasks in the course of my graduate research. Using the hydrology toolset in ArcMap, I was able to identify potential stream networks to locate springs and sources of surface water contamination to spring water chemistry. I was also able to delineate watersheds of different areas to determine which springs in the Sevilleta may be hydrologically linked. Using Spatial Analyst in ArcMap I was able to produce a contour model of estimated evapotranspiration calculated from limited meteorological data collected on the Sevilleta. This data can be useful in hydrological mass balances of precipitation, evaporation, evapotranspiration, and groundwater storage, all vital to my continued graduate level study of the hydrochemistry of surface and groundwater in the Sevilleta. Finally, with ArcScene I was able to .fly through. my field area and determine difficulty in accessing springs in restrictive areas.

Future Work

Portions of this project will continue during my graduate research. The map making tools I learned will be useful for future presentations and publications. I will be able to perform hydrologic mass balances with the ET data to determine how much potential groundwater storage there is at a specific site, and possibly integrate the known hydraulic head at several wells within the NWR to develop a better hydrologic understanding of this complex region. Portions of this project will continue during my graduate research. The map making tools I learned will be useful for future presentations and publications. I will be able to perform hydrologic mass balances with the ET data to determine how much potential groundwater storage there is at a specific site, and possibly integrate the known hydraulic head at several wells within the NWR to develop a better hydrologic understanding of this complex region.

Figure 1. The Sevilleta NWR is outlined in red. Sites labeled with a beaker and water drop are springs sampled by Williams in the Fall/Winter of 2007/08. Those marked with green NWR symbols are meteorological stations operated by Sevilleta LTER personnel. The Rio Grande, Rio Salado, and Arroyo las Alamos are identified as blue streams, and the bounding mountains, Ladron Peak and the Los Pinos, are labeled on the rift flanks. RSB=Rio Salado Springs, RS1=Rio Salado, SLS=San Lorenzo Springs, SanA=San Acacia brine pool, SdC1-1,1-2=Cibola Spring, SdC3=Milagro Spring

Figure 2. Watershed delineations of variable areas over the Sevilleta NWR with spring locations in purple. The 30 km2 watershed is ideal because it grouped RSB, SLS, SdC3, SdC1, BS1, and RS+SanA separately. This can help in understanding whether two springs of close proximity are physically connected.

Figure 3. Potential stream networks in the Sevilleta based on topographic lows.

Figure 4. Contour map of the Sevilleta showing regions of higher flow potential and slope near the Rio Salado, as well as areas in the east near Arroyo Los Alamos.

Figure 5. Contour model of estimated evapotranspiration in the Sevilleta based on meteorological data collected at ten met stations throughout the Sevilleta.

Acknowledgements

UNM Department of Earth & Planetary Sciences
New Mexico Geological Society
Sevilleta LTER Program
L.Crossey, K.Karlsrom, N.Engdahl, D.Gaugler, M.Halick, T.Naibert

Figure 1: Location of Met Stations and Springs in the Sevilleta NWR.
Figure 2: Sevilleta watershed delineations of different areas.
Figure 3: Potential stream networks in the Sevilleta.
Figure 4: Contour map of Sevilleta. Lines close together represent regions of increased surface flow potential.
Figure 5: Contour model of evapotranspiration in the Sevilleta.