Spatiotemporal dynamics of grounwater in the Andean Highlands in northern Ecuador

Abstract

Climate change and some adaptive human actions will affect water resource access and availability for environmental requirements and human supply. Hydrological modelling combined with saturated zone routed techniques is an effective strategy for evaluating the renewable water resources under stationary conditions prior forecasting climate and human impacts under plausible scenarios. This methodology has proven to be suitable for sparse-data mountainous basins intended to supply densely populated urban areas. In this context, this Thesis used the SWAT model to evaluate the renewable groundwater resource (RGR) under a stationary condition as a step to assess the impact of climate change on this water component in the Pita River basin (PRB), which is a representative area of the páramo ecosystem in the Andean Highland in northern Ecuador proposed to meet a fraction of the future water demand in the Metropolitan District of Quito (MDQ). Tracer techniques and Darcian-based formulations were used to determine the groundwater dynamics by means of some essential variables that SWAT model does not provide.

A SWAT model was configured over the period 2006¿2015. A stationarity analysis prior implementing the numerical model was implemented to asseverate that the modelling period is suitable to generate stationary evaluations of the RGR. The SWAT evaluation provided average annual streamflow and aquifer recharge rates around 0.45- and 0.01-fold the average annual precipitation in the area, which is 1,004 mm. The conclusion was that aquifers in the PRB are virtually full, thus repealing most of the infiltration water amount. As in other pristine Andean Highland basins, in the PRB the natural hydrological regime determined by thick, porous andosols over virtually full moderate-permeability volcanic and volcano-sedimentary aquifers induces high streamflow and low aquifer recharge rates.

Five regional climate models (RCMs) based on two Representative Concentration Pathway (RCP) emission scenarios (4.5 and 8.5) were implemented to assess the climate change impact on the RGR for the mid- (2040¿2069) and long-term (2070¿2099) time horizons. All the RCMs indicated some less aquifer recharge in the mid-term. However, this pattern was softened, and even reversed, in several scenarios in the long term. Seasonal differences in streamflow and aquifer recharge in the baseline scenario were predicted to increase to + 23 %. Future hydrological regime could place the highly sensitive soil¿vegetation dynamics of the páramo ecosystem at risk of degradation, with negative consequences for stream water provision. Hence, the RGR fraction seems to be the safest option for water provision in the future, despite that much more knowledge about its dynamics is needed to asseverate that its quantity and quality properties will remain.

Chemical, stable isotope and radioisotope tracer techniques corroborated the existence of shallow and deep aquifers based on the different average groundwater age and sources of salinity associated to each one. The processes controlling the chemical and isotopic baseline of shallow groundwater contributed to baseflow were atmospheric bulk deposition and water-rock interaction. For the deep groundwater discharge, volcanic CO2 degassing and emanations of halides, lithium and sulphate adds to those above two processes. Based on measurable tritium contents, the contribution of old groundwater was interpreted much minor than modern groundwater. For old groundwater, the renewal time of several millennia agrees with large low-yielding aquifers and low aquifer recharge rates. The low recharge rates estimated with SWAT model corroborates a part of this assertion, the relative one to the assumed high thickness (even some thousand meters) and low specific yielding (lower than 0.01) of deep aquifers remains unknown until conclusive data are available.

Darcian-based formulations provided tentative evaluations of average groundwater residence (renewal) time and total groundwater discharge. Since aquifers are virtually full, groundwater storage is much higher than aquifer recharge, the groundwater residence (renewal) times are typically low (some to dozen years) for shallow groundwater and long to very long, even millennia for deep groundwater. Regarding the RGR, average annual discharge was around 340 mm, which was similar to the sum of baseflow and aquifer recharge calculated with SWAT model. The discharge fraction concerned to average annual subsurface groundwater transmitting through the aquifer to the basin outlet was around 0.28 mm, which was coincident with deep aquifer recharge calculated with SWAT model. Average aquifer thickness and saturated volume were tentatively estimated around 1,350 m and 0.64 km3, respectively. This volume does not amount to a significant reserve because the large part is associated to high-mineralized (low quality) old groundwater having a very low renewal.

This Thesis evidences how severe could impact an unplanned exploitation regime on the RGR and especially on the old groundwater reserve; the outlook would worsen if the unplanned exploitation was carried out without considering the additional impact of climate change. Future research will be addressed towards improving the current geological knowledge. This constraint and the low groundwater monitoring prevent implementing coupled surface water-groundwater numerical models for quantitative groundwater assessments. This handicap is similar to that reported in other adjacent basins of the Andean Highland and prevents a sustainable groundwater management.