Martin Scheringer
RECETOX, Masaryk University, Czech Republic
ETH Zürich, Switzerland
Brno, September 19, 2025
Environmental Fate Modeling
of Chemicals, Lecture 1:
Motivation and Foundations
Introduction
Chemical Pollution as a Global Impact
https://www.unenvironment.org/explore-topics/chemicals-waste/what-we-do/policy-and-governance/global-chemicals-outlook
“Findings of the GCO-II indicate that the
sound management of chemicals
and waste will not be achieved by 2020.
Trends data suggest that the doubling
of the global chemicals market
between 2017 and 2030 will increase
global chemical releases, exposures,
concentrations and adverse
health and environmental impacts
unless the sound management of
chemicals and waste is achieved
worldwide.
Business as usual is therefore
not an option.”
From: key messages for policy makers
Chemical Pollution as a Global Impact
Global Environment Outlook 6
Modern society is living in the most
chemical-intensive era in human history,
the pace of production of new chemicals
largely surpasses the capacity to fully
assess their potential adverse impacts
on human health and ecosystems
(...) and now chemical pollution is
considered a global threat. (p. 76 and 88)
https://www.unep.org/resources/global-environment-outlook-6
1940s: Discovery of DDT
è First synthesis in 1874 by O. Zeidler
è Insecticidal effect: discovered in 1938
by P. Müller, Geigy (Basel)
è First production: 1943,
use in World War II against
typhus, malaria
è DDT is the „ideal“ insecticide,
1948 Nobel Prize for P. Müller
1950s: Widespread Use of DDT
Side Effects of DDT: Birds of Prey
èPopulation decline in, e.g., peregrine falcons
ÆEggshell thinning
ÆCaused, among other chemicals, by DDT and its transformation
product DDE
ÆBut: exact cause-effect
relationship complicated
https://daily.jstor.org/the-case-of-the-thinning-eggshells/
„Silent Spring“
èRachel Carson publishes Silent Spring in 1962
ÆPresents effects of pesticides on wildlife and on humans
ÆPoints out connection betweeen research into DDT and
chemical industry
ÆInitiates heated debate, has long-term impacts
Polychlorinated Biphenyls in Wildlife
Polychlorinated Biphenyls in Wildlife
http://response.restoration.noaa.gov/about/media/pcbs-why-are-
banned-chemicals-still-hurting-environment-today.html
The New York State’s 'eat none' advisory and the
restriction on taking fish for this section of the Upper
Hudson has been in place for 36 years.
The Global Presence of Chemicals
α-HCH
Iwata, H.et al., Environ. Sci. Technol. 27 (1993), 1080–1098
èalpha-HCH in the global oceans
The Global Presence of Chemicals
https://doi.org/10.1002/anie.199204873
“During the last 50 years
the prediction that many
manmade chemicals would
reach every corner of the
global environment
has become a reality.”
(p. 512)
The Global Presence of Chemicals
https://doi.org/10.1126/
science.7569923
The Global Presence of Chemicals
èResults from an Africa-wide network of passive
samplers
K. White et al. (2021)
Environ. Sci. Technol. 55,
9413–9424, https://doi.org/
10.1021/acs.est.0c03575
The Global Presence of Chemicals
èSampling locations
in the African
PAS network
and their
airflow patterns
K. White et al. (2021)
Environ. Sci. Technol. 55,
9413–9424, https://doi.org/
10.1021/acs.est.0c03575
F. Wania,
D. Mackay,
Environ. Sci.
Technol.
30 (1996),
390A–396A
The Global Presence of Chemicals
the solution to pollution is NOT dilution
Old Problems Come Back ...
è... or, rather, have been with us all the time:
Jepson and Law, Science 352 (2016)
Old Problems Come Back ...
èOrca whale with
950 mg/kg
PCBs in blubber
èAccumulated over
22 years
Source: The Guardian, May 2017
Environmental Fate Modeling
with Box Models:
Concepts and a First Application
Environmental Fate of Chemicals
èPhase partitioning:
Ædistribution between air, water, soil, sediment,
biota, ice/snow, ...
èTransformation and degradation:
Æformation of transformation products; finally: mineralization
èTransport:
Æmovement of chemical with air and water;
issue: long-range transport
(Stockholm Convention)
CO2, H2O, Cl–
, SO4
2–
air
water
Why Model the Environmental Fate?
èWe cannot measure everywhere:
Æinterpolation
Æextrapolation:
investigate problems before they become manifest
èProcess understanding
èUncertainty analysis
èInvestigation of scenarios: warmer climate, ...
èOverall: models help create the big picture!
What Is a Model?
è Selection of processes
Æ Considered relevant for the problem investigated:
here: phase partitioning, degradation, transport
Æ Quantitatively described within a consistent mathematical framework
here: mass balance equations for a chemical
è Solution of the model:
Æ Calculate masses of a chemical in all compartments of the model
as function of time
è Purposes:
Æ „realistic“ description of the environment: simulation models
Æ Sketch of the environment: evaluative models.
Æ Understanding of processes and their interplay;
Æ Compare environmental fate of different compounds,
Æ „screening“, „ranking“ of chemicals.
Mechanistic Models
èBased on “laws of nature” ...
Æconservation of mass
Ælaws of chemical thermodynamics
Ælaws of chemical kinetics, in particular: law of mass action
Ælaws of diffusion
è... and on empirical data: rain rate, wind speed,
chemical property data, etc.
Multimedia Box Models
è Model types: there are many specific models for chemicals in
Æ air
Æ groundwater
Æ soil
Æ ...
è These different models are difficult to integrate into an
overall picture of a chemical‘s fate.
è Multimedia box models:
Æ overall mass balance of chemical in a system of several linked
environmental media and geographical regions.
Æ Often less highly resolved than media-specific models,
but: overall picture, consistent level of complexity
Multimedia Box Models
èConvenient analytical framework for investigating
chemical fate in connected environmental media
Cover picture of Environ. Sci. Technol. 40:
issue on “Emerging Contaminants” (December 2006)
Multimedia Box Models
èConvenient analytical framework for investigating
chemical fate in connected environmental media
èConvenient analytical framework for investigating
chemical fate in connected environmental media
Multimedia Box Models
A First Example: CCl4, Global
èEmission rate: E (kg/yr)
èVolume of air: V (m3)
èLoss rate constant: k (1/yr)
èConcentration
in air: c (kg/m3)
èMass balance equation: E = c·V·k
Æestimate E, V, k; solve for c
Æestimate c, V, k; solve for E
Æestimate c, V, E; solve for k
A First Example: CCl4, Global
èWith actual data:
ÆE = 80 kt/yr = 5.19 ·108 mol/yr
ÆV = 5·1018 m3
Æk = 1/30 yr = 0.0333 yr–1
Æthis yields:
c = 3.1·10–9 mol/m3 = 70 ppt
Æmeasured concentration:
of CCl4 in the troposphere: c = 90 ppt
Mass Balance Equation for Two Boxes
èair, a, and water, w
èmass-balance equation
for one box: E = c·V·k
èfor two boxes:
E = ca·Va·ka + cw·Vw·kw
and
ca/cw = Kaw
Global Box Models
èGlobo-POP (Wania, Mackay 1995)
èCliMoChem (Scheringer et al. 2000)
èBETR-Global (MacLeod et al. 2005)
Globo-POP
CliMoChem
BETR-Global
General Approach: Mass Balances
è For all boxes, set up mass-balance equations
dm
dt
= Fin – Fout
Fin: emission sources and
inflow from other boxes
Fout: degradation and
outflow to other boxes
and out of the system
all fluxes: F = k·m
needed: rate constants, k
Parameter Requirements (I)
èModel parameters:
ÆChemical specific: partition coefficients, degradation rate
constants
ÆEnvironmental parameters: temperature, vegetation types,
rain rate, aerosol concentrations, etc.
èProblems:
ÆVariability and uncertainty
Parameter Requirements (I)
èModel parameters:
ÆChemical specific: partition coefficients, degradation rate
constants
ÆEnvironmental parameters: temperature, vegetation types,
rain rate, aerosol concentrations, etc.
èProblems:
ÆVariability and uncertainty
{ks
Parameter Requirements (I)
èModel parameters:
ÆChemical specific: partition coefficients, degradation rate
constants
ÆEnvironmental parameters: temperature, vegetation types,
rain rate, aerosol concentrations, etc.
èProblems:
ÆVariability and uncertainty
{ks, … ks }
T
OC
humidity
mechanisms?
??
distribution of values?
1 n
see: Fenner, K., Lanz, V., Scheringer, M., Borsuk, M.
Environ. Sci. Technol. 41 (2007), 2840–2846
Parameter Requirements (I)
èModel parameters:
ÆChemical specific: partition coefficients, degradation rate
constants
ÆEnvironmental parameters: temperature, vegetation types,
rain rate, aerosol concentrations, etc.
èProblems:
ÆVariability and uncertainty
{ks, … ks }
T
OC
humidity
mechanisms?
??
distribution of values?
1 n
Kow
Parameter Requirements (I)
èModel parameters:
ÆChemical specific: partition coefficients, degradation rate
constants
ÆEnvironmental parameters: temperature, vegetation types,
rain rate, aerosol concentrations, etc.
èProblems:
ÆVariability and uncertainty
{ks, … ks }
T
OC
humidity
mechanisms?
??
distribution of values?
1 n
accuracy?
Kow
*
* from: J. Pontolillo, R. Eganhouse, USGS Report 01-4201 (2001)
Parameter Requirements: Emissions (I)
èEmission data: spatially and temporally resolved list
of amounts of chemical released to the environment
èExample: PCBs
source:
Palina Tsytsik, Tamara Kukharchyk,
Belarus
K. Breivik et al., Sci Total Environ 290 (2002) 199–224
Parameter Requirements: Emissions (II)
è Emissions peak only after 2000 (30 years after PCBs)
è Decrease after 2015: too optimistic estimate
Z. Wang et al., Environment International 70 (2014), 62–75
Z. Wang et al., Environment International 69 (2014), 166–176
Environmental Fate Calculations
with Box Models
Regional Scale – Diuron in Queensland
Environmental Fate Models
as an “Integrative Platform”
èFour main components:
chemical properties:
partition coefficients,
degradation half-lives,
degradation schemes
emission inventory:
spatial and temporal
distribution
model results:
concentrations,
mass fluxes,
persistence, ...
field data:
measured
concentrations
and mass fluxes
?
model:
description of the
environment,
here: Tully catchm.
volume, V
loss rate constant, k
emission rate, E
concentration, c
measured
concentration, c
The Tully River Catchment, Queensland
A Less Persistent Chemical
in a Small Region
Australian Bureau of Meteorology
Very Wet and Highly Dynamic
Chemical and Model
diuron emissions
derived from
application to sugarcane:
1.3 kg a.i. per ha and year
degradation half-lives
and partition coefficients:
3 h (air)
42 d (water),
221 d (soil)
log Kow = 2.7
log Kaw = –7
Diuron Concentration in Seawater
Modeled and measured monthly-averaged diuron concentrations
model results
model uncertainty
L. Camenzuli, M. Scheringer et al. (2012)
Sci. Total Environ. 440 (2012), 178–185
Environmental Fate Calculations
with Box Models
Global Scale – Endosulfan in the Arctic
èEndosulfan: insecticide, since 1950s, global pollutant
èAssessed under Stockholm Convention in 2009
Endosulfan: Two Isomers, One TP*
α- and β-endosulfan,
released at ratio of 7:3,
10% to air, 90% to soil
endosulfan
sulfate
minerali-
zation
L. Becker, M. Scheringer, U. Schenker, K. Hungerbühler,
Environ. Pollut. 159, 1737–1743 (2011)
* TP: transformation
product
Global Model “CliMoChem”
yellow: bare soil orange: grass land
green: deciduous forests brown: coniferous forests
soil cover information
from remote sensing data
Scheringer, M., Wegmann, F., Fenner, K., Hungerbühler, K.
Environ. Sci. Technol. 34 (2000), 1842–1850
èDescribes South-North
transport of chemicals
èWell-mixed latitudinal
zones
èEmpirical information:
Ætemperature
ÆOH radicals
Æprecipitation
Æsoil composition
Ævegetation
Æand more ...
Data for CliMoChem: Ice and Snow
è Snow depth from NASA EOS data
è Challenges:
Æ Changing snow density and water content
Æ Dynamics of snow melt
Æ Air-snow phase partitioning
J. Stocker, M. Scheringer, F. Wegmann, K. Hungerbühler
Environ. Sci. Technol. 41 (2007), 6192–6198
January June
Model Inputs: Chemical Property Data
è Partition coefficients and degradation half-lives (days)
log Kow
log Kaw
t1/2, air
t1/2, water
t1/2, soil
a-endosulfan b-endosulfan
endosulfan
sulfate
4.93 4.78 3.71
–3.56 –4.75 –4.78
6; 10; 18 4.4; 8; 13 4; 7; 12
12.6; 22; 38 17.4; 30; 52 58; 100; 173
24; 42; 73 90; 156; 270 180; 312; 540
air: values based on measurements and AOPWIN
water: experimental data for hydrolysis in seawater
soil: selection based on assessment by US EPA R.E.D. (2002)
Model Inputs: Emission Data (I)
èOverall historical emissions (global):1
èTotal historical usage (here assumed to be
released): about 300 000 t
1: Li and Macdonald, Sci. Total Environ. 342 (2005), 87–106
Model Inputs: Emission Data (II)
èContributions of latitudinal zones:1
Æareas of crops under endosulfan treatment:
cocoa, coffee, cotton, fruits, soy, tea, vegetables
èModel input:
1: data from FAO; Bayer Crop Sciences
Model Inputs: Emission Data (II)
èContributions of latitudinal zones:1
Æareas of crops under endosulfan treatment:
cocoa, coffee, cotton, fruits, soy, tea, vegetables
èModel input:
1: data from FAO; Bayer Crop Sciences
zone number
Results: Concentrations in Air
è Good agreement
for all three substances:
Æ latitudinal trend: ok
Æ α > β > sulfate: ok
latitude
model field data below detection limit
N S
L. Becker, M. Scheringer, U. Schenker, K. Hungerbühler, Environ. Pollut. 159, 1737–1743 (2011)
Results: Time Trends in the Arctic
è concentrations of α-endosulfan
in Arctic air
Æ maximum in spring and fall
Æ why?
field data model
L. Becker, M. Scheringer, U. Schenker, K. Hungerbühler,
Environ. Pollut. 159, 1737–1743 (2011)
field data for Alert (82°N) from
H. Hung, Environment Canada
Results: Concentrations in Ocean Water
è field data only north
of 65° N
è sulfate > α ≈ β: ok
è all three substances:
model by factor 10 too low
latitude
model field data
N S
L. Becker, M. Scheringer, U. Schenker, K. Hungerbühler, Environ. Pollut. 159, 1737–1743 (2011)
Results: Concentrations in Ocean Water
è field data only north
of 65° N
è sulfate > α ≈ β: ok
è all three substances:
model by factor 10 too low
è most plausible explanation:
activation energy
of hydrolysis too low
latitude
model field data
N S
L. Becker, M. Scheringer, U. Schenker, K. Hungerbühler, Environ. Pollut. 159, 1737–1743 (2011)
Results: Transfer to the Arctic
èWhat fraction of endosulfan in the Arctic stems
from emissions in different latitudinal zones?
share
of emissions in 2000
share of contribution
to amount in Arctic
seawater in 2000:
N temperate
(40–70 °N)
N subtropic
(20–40 °N)
N tropic
(0–20 °N)
a-endosulfan
b-endosulfan
endosulfan
sulfate
16% 33% 12%
57% 40% 2.5%
66% 33% 1%
65% 34% 1%
transport distance
to the Arctic (km)
2000 5000 8000
L. Becker, M. Scheringer, U. Schenker, K. Hungerbühler, Environ. Pollut. 159, 1737–1743 (2011)
Other Applications ...
è Environmental fate of nanoparticles
è Chemicals in organisms: pharmacokinetic modeling
60
Modeling Engineered Nanoparticles
in the Environment
Environ. Sci. Technol. 46 (2012), 6705–6713, dx.doi.org/10.1021/es204530n
Conclusions
èMulti-compartment models are highly flexible:
èsize and type of environmental system covered
ètemporal and spatial resolution
ètypes of chemicals and processes covered
èMulti-compartment models ake it possible to:
Æcompare different processes
Æanalyze importance of processes
Æintegrate field data, chemical property data, and emission data
Almost 40 Years of Contaminant Fate
Models: from 1979 ...
... to 2010
M. MacLeod et al.,
Environ. Sci. Technol. 44,
2010, 8360–8364
... and 2018
A. Di Guardo et al., Environ. Sci.: Processes Impacts 20, 2018, 58–71
Environmental Fate Modeling
as a Technique
A. Buser et al.,
Integrated
Environmental
Assessment &
Management
8, 2012,
703–708