John Wilson and a graduate student in Water Canyon, Magdelana Mountains near Socorro New Mexico. This is the site of field work investigating mountain block recharge.

A. Long, C. Eastoe (UA-GEO)

Mountainous regions play a critical role in the hydrology of semi-arid drainage basins. Due to orographic forcing of precipitation, especially snow, they receive much more water than surrounding areas and they provide most of the runoff. They also provide most of the groundwater recharge. We hypothesize that a significant proportion of groundwater recharge for the entire Rio Grande basin, and other semi-arid basins, originates as infiltration through fractured bedrock high in the mountains and, depending on the geology, much of it reaches the alluvial aquifer systems by permeating range-bounding faults. This leads us to four general questions:

• What is the spatial pattern of mountain front recharge: is it diffuse or focused, surface or subsurface, and shallow or deep?



• What is the temporal pattern: can at least part of it be regarded as quasi-steady instead of intermittent?



• How non-linear is it?



• What are the roles of climate, landscape, vegetation and geology?

Activities and Results

Our main activities fall into two categories: 1) mathematical modeling and 2) field studies. To date both efforts have been at best cursory, but we expect to dramatically expand them in the future through collaborations within and outside SAHRA.

Modeling: We obtained a two-dimensional saturated-unsaturated (variably saturated) flow code, HYDRUS 2D, that we have been using in profile (cross-section) to model processes at both the hillslope (~100 m) and mountain block (10-50 km) scales. The code did not include dual permeability, or any other ability to readily model fractures. With the cooperation of SAHRA collaborator Jirka Simunek, a composite property capability was added to HYDRUS to allow a first cut simulation of fractured rock. The code is currently being upgraded to dual permeability capability. We also began preliminary simulations with the dual permeability model Tough 2 and are considering other codes (e.g. HMS-Modflow and FEHM) for future simulations. There are other SAHRA researchers interested in the hillslope scale studies, and we will make our code selection in cooperation with them. HYDRUS includes a rudimentary model of ET and the effects of vegetation, an important consideration in our studies and in most of these other SAHRA studies. Properties for different soils and rocks in our work were taken from a literature search, while various mountain blocks in the Rio Grande basin served as models for various aspects of the models. New Mexico Tech structural geologist Laurel Goodwin assisted us with geological models for the basin and range province, focusing on the Rio Grande rift. Precipitation in the complex terrain of mountain blocks was estimated from gauge records using co-kriging and other geostatistical tools.

Field Studies: Our field efforts have focused on the Sandia Mountains, bordering the Albuquerque Basin, and the Magdelena Mountains just west of Socorro. These sites were picked because of their geographic convenience, availability of data, and the presence of relevant processes. The Sandia's are fractured and faulted granites, capped with carbonates on the eastern slope, while the Magdelenas include substantial volcanic rocks. In both mountain blocks we have sampled springs and streams at various times of the year, while we've also sampled mine water inflow to a deep mine in the Magdelenas.

Figure 4 illustrates the conceptual models used in this study. The entire mountain block, on the order of 10 km in horizontal scale, and hundreds of meters to a couple of kms in vertical scale, constitutes the domain we are studying to understand the net processes leading to mountain front recharge. The hillslope scale, a few tens of meters both vertically and in horizontal extent, is used to understand how rainfall and snowmelt is partitioned into soil moisture and runoff, and how soil moisture is partitioned in the shallow subsurface into recharge into the mountain block, downslope interflow in the soil layer (and then to runoff), and ET. Our current modeling treats this partitioning crudely and focuses mostly on the partitioning between interflow and recharge.

Key but preliminary results for the first year of this project consider a range of space and time scales.

At the hillslope scale we have the following preliminary results:

• If the bedrock (a combination of fractures and matrix) is insufficiently conductive, there is effectively no recharge. All soil moisture recycles to the atmosphere via ET or moves downslope in the shallow soil cover. This would be more common in crystalline rocks.



• If the bedrock conductivity is high then there is no barrier to deep recharge into the mountain block, and the soil cover only serves to store water for this recharge and ET. There is little downslope interflow in the soil cover. This would be more common is volcanic rocks.



• Intermediate to these conditions the amount of recharge depends on conditions, with apparently neglected properties, like the topography at the bedrock surface, playing a larger role than expected. Where only a small percentage of soil moisture will recharge with a uniformly sloped bedrock surface, a majority may recharge on a rough slope, presumably because of increased storage.



• While we have seen evidence in these simulations for the effects of percent soil cover, vegetation, slope, aspect, and other features, our preliminary findings are incomplete. We propose to emphasize these additional features in future years.

At the mountain block scale we have the following preliminary findings concerning groundwater flow and mountain front recharge:

Geology is a dominant control on mountain block hydrology.

• Highly permeable carbonates and volcanics can carry virtually all the recharge nature can provide, and conduct it to discharge into adjacent valleys. Fractured and faulted crystalline rocks, on the other hand, cannot carry as much water.
• Bounding faults play a significant role in redistributing the water along the mountain front, due to the enhanced conductivity of damage zones located around the fault.
• Range bounding fault zones, due to reduced permeability, can reduce the capacity of a mountain block to deliver water to adjacent valley aquifers by redirecting water to the surface springs and streams.
• It is important to note that the geology that allows significant deep movement of groundwater is the same geology that maximizes recharge at the surface.

• Although cross-sections do not allow us to properly examine the effect of topography in mountain blocks, it is clear that topography drives the flow, from snowmelt in the mountain highs to discharge along the mountain front.
• Topographic variation within the mountain block leads to smaller-scale circulation systems and discharges to local streams and springs.

It is not surprising that geology and topography play the dominant roles once water percolates deep below the soil cover. They are recognized as important controls in almost every hydrogeological study. Nor is it surprising that the some rocks yield more water than others. This has long been observed in basic scale hydrogeologic studies around the southwest (e.g., the Alamosa Basin portion of the Rio Grande, with crystalline rocks in the mountains on the east and volcanics on the west). However, we hope to move this from subjective understanding to more quantitative understanding and prediction.

Plans

We intend to continue our preliminary mountain block scale research by conducting extensive 2D cross-sectional simulations of various mountain block architectures, then expanding that to 3D, taking into account complex terrain and surface drainage patterns, and finally synthesizing these simulations with spring and stream samples of isotopes and chemistry to build a more complete picture of processes at this scale.

• Cross-sectional simulations of vadose zone and groundwater hydrology across a range of geological mountain block architectures.
• Three dimensional simulations of vadose zone and groundwater hydrology, and related "base flow" for a variety of mountain terrains, geomorphologic drainage patterns and underlying geology, for mountain blocks typical of the basin and range (and rift) mountains of the Rio Grande and Colorado River basins.
• Synthesis of modeling studies with isotopic and chemical samples from springs and streams.