How Polluted Seepage Distribution Diagrams Actually Work in the Field

Most people think these diagrams are just pretty color-coded maps you slap into a report. They're not. A Polluted Seepage Distribution Diagram is a functional tool that tracks how contaminants move through subsurface layers, and getting it wrong means missing the actual plume while your client signs off on a remediation plan that won't hold up. I spent three years doing environmental site assessments before I stopped treating these as boilerplate deliverables and actually learned how to make them useful.

Polluted Seepage Distribution Diagram: What It Is and What It Isn't

A polluted seepage distribution diagram visually represents the spatial extent and concentration gradient of contaminant-laden water moving through soil and groundwater systems. It combines hydrogeological data with contaminant tracking to show where polluted water is going, how fast it's moving, and which pathways it's following. The typical inputs are groundwater monitoring well data, soil boring results, hydraulic conductivity values from slug tests or pump tests, and seasonal water table fluctuation records. You layer that against the subsurface stratigraphy to produce a contour map showing concentration vs. distance. Here's the part nobody tells you: the diagram is only as good as the interpolation method you choose, and most consultants default to kriging because it sounds impressive in a report. Kriging assumes spatial continuity that often doesn't exist in fractured bedrock or heterogeneous fill. I've seen two kriging-based diagrams drawn from the same dataset that looked completely different because the analysts used different variogram models. One showed a tight plume. The other showed it spreading three times further. Both were technically "correct" by their chosen parameters.

The Practical Workflow

Start with your raw data. That means every monitoring well reading, every lab result, every boring log. Don't cherry-pick. I once had a project where excluding three outlying well readings made the plume look contained and remediable for under fifty thousand dollars. Including them pushed the boundary past the property line and triggered a state-level investigation that cost eight times more. The client was upset I'd found those readings late, but I'd been asked to verify the preliminary assessment, not confirm it. From there, establish the potentiometric surface. Map the water table elevations across your monitoring network using contour interpolation. This defines the hydraulic gradient, which controls the direction of seepage. If you skip this step or use stale data from a previous season, your flow directions will be wrong, and your entire distribution diagram will point contaminants in the wrong direction. Next, overlay the contaminant concentration data. Draw concentration contours between your monitoring points. This is where the diagram earns its name — you're mapping where the polluted seepage is actually present, not where you suspect it might be. For the interpolation itself, I use inverse distance weighting (IDW) as my primary method with kriging only as a secondary check when the data density supports it. IDW is less fancy but more predictable. When you have sparse data points, fancy methods will fabricate structure that isn't there.

A Real Problem I Dealt With

On a former dry cleaner site with PCE contamination, the monitoring well network was installed on a rectangular grid pattern. The Polluted Seepage Distribution Diagram I produced showed the plume migrating steadily northeast. Everything looked clean and predictable. Then I pulled the regional hydrogeologic report for the area and noticed something the original assessment had missed. There was a buried paleochannel — an old river bed filled with coarse gravel — running diagonally through the site at roughly a forty-five-degree angle to the monitoring grid. The gravel had hydraulic conductivity two orders of magnitude higher than the surrounding clay loam. The plume wasn't migrating northeast uniformly. It was funnelling through that gravel channel at accelerated speed, and the monitoring wells on the grid were completely missing the leading edge because they were installed outside the channel's path. The workaround was straightforward but tedious. I requested additional test pits and piezometer installations along the suspected channel axis. Once I had data points actually intersecting the high-permeability zone, I redrew the diagram with a segmented interpolation approach — separate conductivity zones for the channel versus the matrix. The revised Polluted Seepage Distribution Diagram showed the plume had advanced twice as far as the original indicated, and the contaminant concentrations inside the channel were four times higher than the surrounding matrix wells suggested. That project saved the firm from looking incompetent during a regulatory review six months later. The state inspector independently identified the same paleochannel issue.

Counter-Intuitive Things Beginners Miss

Temporal variation matters more than spatial resolution. A diagram based on a single sampling event is almost always misleading. Seasonal rebound can shift the potentiometric surface by two to three meters, completely reversing apparent flow directions. I always recommend at least two sampling events separated by a significant hydrologic change — spring melt, heavy rain period, or irrigation shutdown. The difference between those two datasets is where the real answers live. Vertical heterogeneity gets ignored way too often. Most diagrams are presented as flat 2D maps because that's what fits in a report appendix. But seepage doesn't move in a plane. It moves through layered media with different permeabilities. A proper diagram should show at least three depth slices — shallow, intermediate, and deep monitoring intervals. contaminants frequently migrate preferentially through mid-depth permeable lenses while the shallow and deep zones remain relatively unaffected. Another thing: don't trust contour interpolation alone. Always cross-reference with the actual raw data points. I've seen diagrams where the 100 ppm contour looped around an area that had no monitoring wells above fifty ppm. The interpolation was creating a phantom high-concentration zone because of how it weighted distant outliers. The fix is simple — add a point-data overlay on top of your contours so readers can see where the actual measurements fall relative to the interpolated lines.

Software and Deliverables

ArcGIS and QGIS are the standard tools. Surfer is still used by some older hydrogeologists but it's clunky for anything beyond basic contouring. For the interpolation itself, I keep it simple. IDW in GIS with a power of two, bounded search radius set to roughly the average spacing between monitoring wells, and output resolution no finer than half that spacing. Finer resolution gives a false sense of precision. The final diagram should include: the contour lines with labeled values, the potentiometric surface contours, monitoring well locations with their measured concentrations, the regional geological framework, and a scale bar plus north arrow. Include a legend that distinguishes between measured data points and interpolated contours so anyone reading it understands what's actually known versus what was estimated. Save the project file alongside the published diagram. Regulatory reviewers increasingly ask for the underlying workspace to verify methodology. I've had two projects where I couldn't reproduce a diagram after six months because I'd saved only the exported PDF and not the GIS project with its symbology and interpolation parameters intact.

When These Diagrams Fail Completely

They don't work in karst terrain. Dissolved limestone creates conduit flow paths that are impossible to map with standard monitoring networks. The seepage moves through caves and solution channels at velocities that make any interpolation meaningless. If you're working in carbonate bedrock and someone hands you a kriged concentration map, flag it immediately. The data density required to capture conduit flow would need wells every ten to twenty meters, and even then you'd likely miss the active passages. In those cases, dye tracing is the only reliable method, and it doesn't produce a conventional distribution diagram — it produces flow path maps based on tracer recovery points. They also fail when the contaminant is non-aqueous phase liquid. DNAPLs like chlorinated solvents sink through the water table and pool on low-permeability layers. A standard seepage distribution diagram assumes dissolved-phase transport and will dramatically underestimate contamination below the water table. You need dedicated sub-slab and lower-aquifer sampling to catch this, and the resulting diagram looks nothing like the clean plume shapes you'd expect from textbook examples.

Quick Reference Checklist

Verify your monitoring well data covers at least two hydrologic seasons. Confirm the potentiometric surface is mapped from current data, not inherited from a previous study. Use interpolation parameters that match your actual data density — don't let software defaults create finer resolution than your wells justify. Overlay raw data points on your contours. Include depth-specific slices rather than a single flat map. Document every interpolation method and parameter used. Flag any terrain or geochemical conditions where the diagram methodology may be unreliable.