MAPPING RECHARGE AND DISCHARGE ZONES BASED ON CORRELATIONS BETWEEN DYNAMIC PRESSURE INCREMENT AND STREAM ORDER ANALYSES

Defining vertical flow of groundwater is important for developing conceptual hydrologic/hydrogeologic models and water budgets. Defining vertical flow components can be difficult with sparse well data. Vertical flow varies in 3D space between local-to-regional flowpaths. A GIS-based analysis of dynamic pressure increments (Δp) and proximity of wells to streams of different Strahler order maximizes the use of sparse data to map upward- and downward-directed flow. The method was tested in part of the Pedernales River watershed in Central Texas to study recharge and discharge and movement of groundwater between the Trinity Hill Country aquifer in Cretaceous karstic limestone and the Llano Uplift aquifer in lower Paleozoic sandstone and karstic limestone. The Pedernales River is a Strahler 5^th-order stream. Its tributaries in the maturely dissected study area include 1^st to 4^th-order streams. A potentiometric surface map shows the Pedernales River is the regional base level; groundwater is recharged in upland areas and flows toward discharge areas in the floor of the river valley. There are also local components of groundwater flow to springs and seeps in its tributary streams. A dynamic pressure increment (Δp) as defined by Tóth measures departure from hydrostatic and indicates where groundwater flow is directed upward (+Δp; discharge) or downward (-Δp; recharge). The Δp value was calculated from measured water columns in 237 wells with records provided by local and state agencies and field measurements. Posting Δp on a GIS-based topographic map showed an obvious correlation—positive values in tributary valleys and negative values in upland settings. Recharge and discharge in local-to-regional scales of flow were assumed to depend on degree of topographic relief and, therefore, would be associated with Strahler stream order. GIS tools were used to determine distance of wells from channels of various stream order. Correlation of Δp and well distance is statistically significant for 3^rd - to 5^th-order streams. Recharge and discharge areas for individual streams were mapped as a function of distance from each stream and considered the linear regression’s prediction interval. A composite map of the recharge area was derived by superposing recharge areas of the various stream segments.