A long-term hydrological model for the northern Espanola Basin, New Mexico

A three dimensional groundwater model was prepared for the northern Española Basin along the Black Mesa Reach of the Rio Grande bounded by the USGS Embudo Gauging Station on the north and the confluence with Rio Chama on the south. The modeled region encompassed an area of approximately 208 square kilometers.;The main objective of the model was to predict the aquifer effects of irrigation and to explain the hydrologic system dynamics with consideration of the acequia irrigation practices in the area. Acequias are community irrigation systems some of which have existed since Spanish colonization. In northern New Mexico, the acequia water-use regime and attendant acequia-related cultural values are at particular risk due to increasing urbanization pressures and the potential impacts on actual water use, water quality, and riparian vegetation along the Rio Grande and irrigation ditches.;The characterization of shallow groundwater-acequia-river interactions was the main aspect of this research. The first stage of the research included conceptualization of the problem, data collection through literature review and field work, and preparation of input data for the model. The second stage included the development of a geological model and three classes of groundwater simulation models. The final stage of the research was to investigate the potential effects of potential climatic changes on the hydrologic system through seven different model scenarios.;The USGS modeling environment MODFLOW2000, a groundwater modeling code, was used for the hydrologic simulations. The aquifer system in the model is represented by an equally spaced grid (300 m × 300 m), 3 layers, 65 rows and 59 columns with a total depth of 100 meters at its deepest point. The steady-state and short-term transient models provided input for the subsequent transient runs. The main transient model simulated the groundwater conditions in the region from 1974 to 2007 with 408 monthly stress periods. The Streamflow Routing Package (SFR1) was implemented in the main transient model to simulate the groundwater-surface water interactions. An acequia (irrigation canal)-river network consisting of 361 reaches and 33 segments was constructed to represent the river flow and diversions from the Rio Grande to 10 major irrigation canals in the region.;The calibrated long-term transient model simulated the heads with an absolute residual mean of 0.51 m in the target wells. The computed flow at the very end cell of the model was compared to the flow records from the USGS San Juan Pueblo Gauging Station. The flow records from 1974 to 1986 matched well with the model simulated flow values. The difference between the recorded and simulated values ranged between +15% and -39% of the recorded flow at the San Juan Pueblo Gauging Station with an average between +8% and -12%.;The calibrated model then was compared to a transient model that only considered the river as a surface water feature and no diversions for irrigation. The model with no irrigation diversion yielded a higher river flow by 1.2%. Having no irrigation diversion from the river decreased the groundwater levels in the region by 17.7 cm on the average thereby decreasing phreatophytic ET. The canal-shallow groundwater interactions showed that, except for the small portion in the northern part of the area, all the canals recharged the shallow groundwater in the region.;To investigate the potential effects of climatic change on the modeled system, seven alternative model scenarios were created and tested. The original calibrated model was modified to include 648 stress periods (1974-2027) for future predictions. The new models were used to simulate the response of hydrologic system to the changes in areal recharge (precipitation) and evapotranspiration (ET).;The largest difference in flow was observed between Scenarios 1 (original calibrated model extended to 2027) and 7 (reduced ET and areal recharge, no-diversion). The comparison of predicted river flows showed that the Scenario 7 (reduced ET and areal recharge, no-diversion) conditions increased the river flow by 1.07% on the average compared to the Scenario 1. Overall, evapotranspiration and canal diversions seemed to have the strongest impact on the hydrologic budgets and river-aquifer-canal interactions based on the comparison between the model scenarios.