Case Study: Pesticide Degradation

This case study illustrates the simulation of groundwater contamination by pesticides and the use of coupling between concentrations, metabolites, and CSIA isotopic signatures to distinguish between active degradation, dilution, and adsorption. The practical question: why is the atrazine concentration dropping faster than expected decay suggests, and does this decrease reflect real degradation or a simple spreading of the plume into a larger volume?

Context

The setting is an alluvial aquifer in an agricultural zone. Atrazine, banned in France since 2003, is still regularly detected in several monitoring wells in the basin, often accompanied by its metabolites DEA and DIA. Atrazine concentrations are generally decreasing year after year, but at a rate that varies significantly from one piezometer to another. The key question is the nature of this decrease: is it true natural attenuation through microbial degradation, or simple transport and dilution shifting the problem elsewhere without actually resolving it?

Three piezometers in the network have been equipped for atrazine CSIA (δ¹³C and δ¹⁵N) for the past three years. These isotopic measurements are available and can be compared against the simulation results.

Scene Setup

Scene Element	Configuration
Horizontal Footprint	2000 × 800 m, alluvial agricultural zone
Depth	12 m, referenced to ground level
Calculation Grid	100 × 40 × 24 = 96,000 cells
Surface Layer (0 to 2 m)	Agricultural soil, organic matter 1.5 to 3%, pH 6.5 to 7.5
Alluvial Aquifer (2 to 10 m)	Gravelly alluvium, porosity 0.20, K = 5e-4 m/s
Substratum (10 to 12 m)	Marly clay, porosity 0.05, K = 5e-9 m/s
Sampling Points	12 piezometers, including 3 CSIA-equipped

Compounds and Parameters

The simulation tracks atrazine and its two primary metabolites, DEA and DIA. Three degradation pathways are activated in the engine, each producing a different dominant metabolite and each having its characteristic carbon epsilon.

Pathway	Dominant Metabolite	Epsilon δ¹³C (‰)	Favorable Conditions
Microbial Aerobic Dealkylation	DEA	-2 to -5	Dissolved O₂ > 1 mg/L, available organic matter
Surface Photodegradation	DEA and DIA	Near 0	Shallow zones, < 0.5 m below surface
Abiotic Hydrolysis	Hydroxyatrazine	Very low, < 1	Acidic or basic pH

The degradation rate *k* varies cell-by-cell based on interpolated geochemical conditions. In aerobic zones rich in organic matter, the microbial pathway dominates with a *k* around 0.003 /day. In deep anoxic zones, degradation is much slower with a *k* around 0.0001 /day. The engine simultaneously propagates concentration and isotopic signatures according to the Rayleigh equation.

Unlike a traditional simulation where a global *k* is imposed, here the effective *k* emerges from the coupling with local geochemical conditions. This means that if the user accurately describes the site's spatial geochemistry, the model naturally produces the degradation heterogeneity required to explain observed differences between piezometers.

Source

The source is modeled as an active diffuse surface source from 1970 to 2003, reflecting historical agricultural applications of atrazine. After 2003, direct input ceases, but slow desorption from the soil continues to feed the aquifer for several more years. The source intensity reflects typical regional agricultural use, calibrated to reproduce the order of magnitude of concentrations measured at the start of monitoring.

Results

The 55-year simulation produces a plume that is globally homogeneous at the site scale but exhibits marked local variations. Simulated atrazine concentrations at the 12 piezometers reproduce the observed decreases, including the discrepancies between wells: some show rapid decay while others are much slower. This differentiation is consistent with the local geochemical conditions interpolated on the grid.

The simulated atrazine/DEA ratio is particularly informative. In zones where aerobic degradation is active, the ratio drops below 1, with DEA becoming the majority compound. In low-degradation zones, the ratio remains near or above 1 even after several decades. This spatial differentiation is a direct footprint of degradation heterogeneity.

Piezometer	Geochemical Zone	Measured Atrazine (ng/L)	Measured DEA (ng/L)	Measured δ¹³C Atrazine (‰)	Simulated δ¹³C (‰)
PZ-03	Aerobic, High OM	80	220	-22.1	-22.5
PZ-07	Intermediate	180	150	-25.3	-25.0
PZ-11	Deep, Low Oxygen	320	40	-27.4	-27.2

Interpretation

The simultaneous reproduction of concentrations, parent/metabolite ratios, and isotopic signatures at the three CSIA-equipped piezometers validates the central hypothesis: the observed decrease of atrazine in the aquifer is not merely due to transport and dilution, but reflects microbial degradation that is actively occurring in certain areas of the site. The measured isotopic enrichment cannot be explained by simple mixing or preferential adsorption; it is the direct signature of kinetic fractionation associated with degradation.

This conclusion has practical implications: in the aerobic zones of the site, natural attenuation is effective and concentrations will continue to decrease. In less oxygenated zones (PZ-11 and its geochemical vicinity), atrazine will persist for several more decades. Areas requiring reinforced monitoring are thus clearly identifiable, while zones that are self-remediating can be left to their natural evolution.

Without CSIA, the simulation would likely have adjusted a uniform global *k* based on the average concentration drop. The result would have seemed correct on average but would have missed the spatial heterogeneity that is precisely the most useful information for site management. Integrating CSIA and simulation is far more than a check: it is an interpretive lever that guides operational decisions.

Extensions

"Persistent source" variant: simulate a scenario where soil desorption lasts longer than expected to evaluate how long the aquifer will remain contaminated even without new inputs.
Addition of simazine and terbuthylazine: extend the simulation to other triazines in the panel to see if they exhibit the same heterogeneities or if persistence varies by molecule.
Integration of δ¹⁵N on atrazine: add the nitrogen dimension to distinguish dealkylation (which primarily fractions C) from hydrolysis (which primarily fractions N), useful if pH conditions vary.
Point-source scenario: test the hypothesis that some high concentrations are due to more recent inputs rather than soil memory alone.

Learn More

Pesticides: Triazine family, metabolites, CSIA.
Degradation Pathways: Atrazine metabolite chains.
Speciation and Propagation: Spatially variable kinetics.
CSIA Isotopy: Principles of residual enrichment.