What data layers do I need for a GIS hydrogeology screening?

At minimum: a geological/lithology map, a DEM with slope and drainage derivatives, land cover, soil or surficial permeability, rainfall, and any well, spring, or borehole data. Lineaments from terrain or remote sensing and a water-table surface add fracture and gradient context.

How do I estimate aquifer potential or recharge in GIS?

Screening typically reclassifies each thematic layer into ranked influence classes, assigns weights reflecting hydrogeological importance, and combines them in a weighted raster overlay to produce a relative potential index. It is a relative screening tool, not a quantitative recharge measurement.

What is a weighted overlay and how is it used for groundwater?

A weighted overlay reclassifies several rasters to a common scale, multiplies each by an importance weight, and sums them. For groundwater it combines lithology, slope, drainage density, land cover, and rainfall into a single relative groundwater-potential index.

How do I screen for groundwater vulnerability with GIS?

Vulnerability screening uses indices such as DRASTIC, which combine depth to water, recharge, aquifer media, soil, topography, vadose-zone media, and conductivity as weighted rasters to map relative contamination susceptibility.

GIS for Hydrogeology Screening: A Practical Workflow

The short version

GIS hydrogeology screening is a reconnaissance method that combines geology, terrain, soils, land cover, rainfall, and well data into a single relative index — most often groundwater potential or aquifer vulnerability — built with a weighted raster overlay. It is a screening tool: it tells you where to look and where to drill exploratory boreholes, not how many litres per second a well will yield. The output is only as defensible as the input layers, their CRS alignment, and the transparency of the weights you choose.

The reason GIS is the right tool here is that groundwater occurrence is controlled by several spatial factors at once — what the rock and soil let through, how the land sheds or holds water, where fractures concentrate flow — and overlaying them is exactly what GIS does well. The discipline is in keeping every layer on the same grid and CRS, justifying the reclassification and weights, and being honest that the result is relative.

The factor layers and why each matters

A groundwater-potential screen typically combines these themes. Each is reclassified into ranked classes (high to low favourability) before overlay.

Lithology / geology. The dominant control. Unconsolidated sands and gravels and fractured/karstic rock favour storage and flow; massive crystalline rock and clay do not. Derive from a digital geological map; if scanned, georeference and digitize units rather than trusting colours.
Slope (from the DEM). Gentle slopes give water time to infiltrate; steep slopes promote runoff. Reclass low slope as high favourability. Compute on a projected DEM (gdaldem slope).
Drainage density. Higher drainage density often indicates lower permeability (more runoff), so it tends to correlate negatively with recharge potential. Derive a stream network from flow accumulation, then line density.
Lineament density. Fractures and faults are conduits in hard-rock aquifers. Map lineaments from hillshade and remote sensing, then compute density; high density raises potential in fractured terrain.
Land use / land cover. Vegetated and bare permeable surfaces allow recharge; impervious urban cover and open water change the picture. Use a classified land-cover raster (e.g. Copernicus land cover).
Soil / surficial permeability. Sandy, well-drained soils favour infiltration; clays impede it.
Rainfall. The recharge source. Interpolate gauge data or use a gridded climate product; more rainfall, higher potential, all else equal.

Optional but valuable: a water-table surface interpolated from well levels (gradient and depth to water), and spring/borehole locations for validation.

Get every layer onto the same grid

The most common failure in overlay screening is silent misalignment. Before any reclassification, bring all rasters to one CRS, one cell size, one extent, and one snap origin. Pick a projected metric CRS for the basin:

gdalwarp -t_srs EPSG:25831 -tr 30 30 -te <xmin ymin xmax ymax> -r bilinear lithology.tif lith_30m.tif
gdalwarp -t_srs EPSG:25831 -tr 30 30 -te <xmin ymin xmax ymax> -r bilinear slope.tif slope_30m.tif

Use bilinear/cubic for continuous data (slope, rainfall) and nearest neighbour for categorical data (lithology classes, land cover) so class codes are not blurred. Confirm with gdalinfo that all layers report identical size, extent, and pixel size; if they differ by even one row the map algebra will misregister.

Reclassify to a common scale

Convert each thematic layer to a common ranked scale (commonly 1–5, low to high favourability). Categorical layers map by lookup (e.g. alluvium = 5, sandstone = 4, shale = 2, granite = 1); continuous layers map by ranges (slope <3° = 5, 3–8° = 4, … >25° = 1). In QGIS use Reclassify by table or the raster calculator; in ArcGIS, Reclassify. Document every class boundary — these choices are where local hydrogeological judgement enters, and a reviewer must be able to see and challenge them.

Weight and combine

Assign each factor a weight reflecting its hydrogeological importance (lithology usually highest, rainfall and slope moderate, land cover lower). Weights should sum to 1 (or 100%). Combine in a weighted linear overlay:

GWP = 0.30*lith + 0.20*slope + 0.15*lineament + 0.15*drainage + 0.10*soil + 0.10*rainfall

In QGIS this is the Raster Calculator; in ArcGIS, the Weighted Overlay or Weighted Sum tool. Where you want a defensible, repeatable weighting rather than gut feel, the Analytic Hierarchy Process (AHP) is the standard method: pairwise-compare factors, derive weights from the comparison matrix, and report the consistency ratio (keep it below 0.1). Whatever method you use, publish the weights — an overlay with hidden weights is not auditable.

Vulnerability screening: DRASTIC

When the question is contamination susceptibility rather than yield, the established index is DRASTIC, which combines seven weighted layers: Depth to water, net Recharge, Aquifer media, Soil media, Topography (slope), Impact of the vadose zone, and hydraulic Conductivity. Each is rated and multiplied by a published weight; the sum is a relative vulnerability index. It is built in GIS exactly like the potential overlay above. Treat DRASTIC as a relative, regional screening index — it is widely used and equally widely cautioned to not be over-interpreted at parcel scale.

Worked example: a hard-rock basin

A 400 km² crystalline-basement basin needed siting for exploratory boreholes. We assembled lithology, a 30 m DEM (slope and drainage from it), lineaments mapped from multidirectional hillshade and Sentinel-2 imagery, land cover, soils, and gauge rainfall. Everything was warped to EPSG:25831 at 30 m on a common grid. Each layer was reclassified 1–5, weights set by AHP (lithology and lineament density highest, because in basement terrain storage is fracture-controlled), and combined in the raster calculator. The resulting potential map concentrated "high" zones along mapped fracture corridors on gentle slopes — and crucially, of the 22 existing productive boreholes in the basin, 17 fell in the high or very-high classes, giving the screen credibility before any new drilling.

Validate against reality

A screening map you cannot test is a guess with colours. Validate by:

Overlaying known wells/springs and checking that productive ones fall in high-potential classes (as above). Compute the proportion in each class.
Cross-checking depth-to-water from monitoring wells against the index where data exist.
Sensitivity testing — re-run with the weights perturbed; if the high zones move drastically, the result is weight-driven, not data-driven, and you must say so.
Inspecting basin edges and NoData boundaries where overlay artifacts collect.

Bathyl perspective

We deliver hydrogeology screens as transparent, reproducible overlays: every factor layer documented with its source, CRS, and reclassification table, every weight justified (typically by AHP) and published, and the output explicitly framed as relative potential validated against known wells. The value is not a coloured map — it is a defensible, auditable basis for deciding where to spend drilling budget.

GIS for Hydrogeology Screening