Report 56 — North American Acid Deposition: Control and Pressure Reduction Preceded Uneven Monitored Surface-Water Chemistry Recovery

v1.8 status note: This is a bounded retrospective comparison. Its adequacy gates were specified after the historical outcomes were known, not preregistered in advance. The case can refine or challenge the theory, but it does not by itself establish a general Repair-Lag mechanism.

v1.8.1 endpoint correction: Headlines now identify the observed recovery measure as monitored surface-water chemistry. Direct biological recovery remains separately undefined; REV012 records the correction.

Case question

When sulfur and nitrogen emissions and deposition fall but monitored surface-water chemistry recovery remains uneven, is the remaining slowness institutional repair lag or biogeochemical recovery time?

Short answer

The United States–Canada acid-deposition policy package achieved substantial control and pressure reductions. The 2024 joint assessment reports that both countries had met their Acid Rain Annex commitments since 2007. Between 1990 and 2020, it reports sulfur-dioxide emissions down 78% in Canada and 93% in the United States, and United States nitrogen-oxide emissions down 70%. Power-sector indicators, wet-sulfate deposition, and critical-load exceedance also fell sharply; 2022 Acid Rain Program facilities held enough allowances for full programme compliance.

Monitored surface-water chemistry recovery was real but uneven. EPA reports improvement in acid-neutralizing capacity at 90% of monitored Adirondack sites, 77% in New England, 67% in the Catskills and Northern Appalachian Plateau, and 19% in the Central Appalachians over the reported period. These chemistry indicators do not establish biological-community recovery, whose canonical threshold remains undefined. Soils, catchments, geology, remaining pollutant loads, and biogeochemical time can govern recovery after the pressure has fallen.

This is a boundary case because it challenges a simple equation: lag exists, therefore institutional repair failed.

Bounded repair episode

The unit is the layered policy, technology, fuel, market, monitoring, and ecological-response package. It is not the United States Acid Rain Program treated as an isolated cause.

Four-endpoint assessment

The numerical gates below are retrospective analytic operationalizations. They are useful for making the case auditable, but they are not a preregistered causal test.

Empirical anchors by evidence type

Legal and institutional record

• United States Title IV set statutory sulfur-dioxide and nitrogen-oxide control objectives and monitoring requirements. Source: S145.
• The 1991 Canada–United States Air Quality Agreement established binational acid-rain commitments and review architecture. Source: S146.
• In 2022, Acid Rain Program facilities held sufficient allowances for reported programme compliance. Compliance is an administrative endpoint, not an ecological endpoint. Source: S152.

Pressure reduction

• The 2024 joint assessment reports commitments met since 2007. From 1990 to 2020, it reports sulfur-dioxide emissions down 78% in Canada and 93% in the United States, and United States nitrogen-oxide emissions down 70%. Source: S148.
• EPA power-sector reporting shows sulfur-dioxide emissions down 95% and nitrogen-oxide emissions down 89% from 1995 to 2023. This is a combined policy, technology, fuel, and market outcome and is not attributed to the Acid Rain Program alone. Source: S149.
• EPA reports a 73% decline in eastern-United States wet sulfate deposition between the 2000–2002 and 2020–2022 comparison periods. Total deposition also uses modeled dry-deposition estimates. Source: S150.

Recovery indicators

• The joint assessment reports calculated critical-load exceedance in the studied United States acid-sensitive lakes and streams declining from 38% in 2000–2002 to 5.8% in 2019–2021. The remaining 5.8% is not zero, and the endpoint is model-based. Source: S148.
• EPA reports regionally uneven improvement in acid-neutralizing capacity: 90% of monitored Adirondack sites, 77% in New England, 67% in the Catskills and Northern Appalachian Plateau, and 19% in the Central Appalachians over 1990–2022. Source: S151.
• The long-term surface-water chemistry network documents the monitored domain and its selection boundary. It does not represent every North American water body. Source: S153.

What the case changes conceptually

Repair-Lag analysis must distinguish at least three temporal statements:

1. institutions were late or inadequate in reducing the pressure;
2. institutions reduced the pressure, but the affected substrate recovers slowly; and
3. institutions reduced part of the pressure, while residual scope or coordination problems remain.

Only the first is straightforward institutional repair failure. The second can be successful control with delayed recovery. The third requires a bounded diagnosis of what remains unaddressed.

For CS005, the primary remaining stage is replenishment. The leading explanation is biogeochemical response time, recorded as not_applicable in the institutional lag-type field and explained separately. A secondary remaining scope or coordination problem stays not_yet_defined unless evidence isolates it. Strategic lag is not established.

Power and distribution

United States and Canadian legislatures, regulators, binational bodies, utilities, large emitters, allowance holders, fuel suppliers, technology providers, and monitoring networks held different forms of authority and information. Source and receptor regions were often geographically separated. Firms chose among controls, fuel switching, allowance transactions, operational changes, and closure, while public institutions set targets, monitoring rules, and adequacy definitions.

The current official record does not adequately establish who ultimately paid through electricity prices, taxes, shareholder losses, public subsidy, employment change, or regional economic adjustment. It also does not establish how benefits and remaining harms were distributed among downwind communities or how Indigenous participation, ecological dependence, and rights were affected. Those are explicit gaps, not grounds for inference.

Serious rivals

• Binding regulation and monitorability: enforceable caps, measurement, penalties, and compliance flexibility may explain control success without a general Repair-Lag mechanism.
• Technology, fuel, and market change: scrubbers, low-sulfur fuel, natural gas, efficiency, plant closures, and energy-market change contributed to the trend.
• Policy-package confounding: the Acid Rain Program, later federal and state rules, Canadian measures, technical change, and market conditions overlap.
• Biogeochemical memory: depleted base cations, stored sulfur and nitrogen, soil and catchment processes, geology, hydrology, and remaining loads predict uneven recovery after deposition falls.
• Endpoint definition: legal compliance, precursor reduction, deposition, chemical recovery, and biological recovery are different outcomes.
• Monitoring selection: long-term networks emphasize acid-sensitive regions and continuously sampled sites; their results cannot be generalized to all waters.

The boundary interpretation is strongest if post-control recovery variation tracks catchment, soil, geological, hydrological, and residual-load conditions more closely than continuing institutional non-repair.

Retrospective decision rules

• Institutional/control gate: declared commitments are materially met, regulated facilities are monitored, and annual compliance is demonstrated.
• Pressure gate: precursor emissions fall by more than 80% from the declared baseline and wet sulfate deposition falls by more than 70% over the declared comparison.
• Chemical-recovery gate: site-appropriate acid-neutralizing-capacity and critical-load criteria are assessed by region rather than collapsed into one score.
• Biological-recovery gate: not_yet_defined until direct biological indicators and thresholds are registered.

These rules were formed after reading official summaries. They must be locked before lower-level reanalysis and cannot be described as prior confirmation.

What would lower confidence

Lower confidence in the boundary classification if:

• persistent non-recovery is explained chiefly by continuing, remediable implementation, scope, financing, or coordination failures;
• emissions or deposition trends reverse materially;
• reported chemical improvements do not survive broader monitoring or alternative endpoints;
• biological recovery remains absent in ways not explained by soil, catchment, geology, hydrology, or remaining loads;
• independent attribution shows that the registered control package contributed little to the pressure decline; or
• retrospective endpoint choices are shown to select a favorable story.

Current theory disposition

C1 and C7 remain developing.

Next test

Test whether regional post-control recovery variation is better predicted by catchment chemistry, geology, historical depletion, and remaining loads than by continuing institutional implementation differences. Add biological outcomes, independent causal attribution, distribution and legitimacy evidence, and prospectively specified thresholds before strengthening the verdict.

Misuse warning

Do not treat the strong control record as complete ecological repair, harmless delay, proof that allowance markets alone produced the result, or evidence that one regulatory architecture works everywhere. Do not erase communities, Indigenous rights, local burdens, remaining exceedances, non-power sources, biological recovery, or distributional questions. Slow recovery does not automatically prove policy failure, but formal compliance and falling emissions do not prove complete ecological repair.

Canonical sources used

• S145 — Title IV: Acid Deposition Control
• S146 — Canada–United States Air Quality Agreement
• S147 — 2011 NAPAP integrated assessment
• S148 — 2024 Canada–United States joint review
• S149 — EPA emissions reductions
• S150 — EPA atmospheric deposition
• S151 — EPA ecosystem response
• S152 — EPA programme compliance and market activity
• S153 — EPA surface-water chemistry monitoring

Official legal, administrative, and monitoring sources are strong records for their stated domains. Several are produced by institutions that designed or administer the programmes. They do not substitute for independent causal, ecological, distributional, or legitimacy review.