Watershed and Lake Restoration: Diagnose, Prioritize, Act, and Verify

Build a jurisdiction-specific escalation card for emergency services, environmental incidents, HAB or public-health concerns, fish mortality, drinking-water receptors, and after-hours handoff; use a tabletop scenario without entering water or collecting a sample.

Would the team protect people and animals, preserve observations from a safe location, notify the right owner, and avoid cleanup, sampling, source attribution, treatment, or public-status decisions unless directed and authorized?

Create a responsibility table with primary and backup owner, jurisdiction, current procedure, evidence needed, permitted action, prohibited inference, handoff trigger, and dated source for each decision.

Are observation, technical recommendation, landowner consent, permit, authorization, implementation, verification, and public claim separate attributable states?

List candidate regional, historical, lake-type, within-lake, project, and regulatory comparisons; record authority, date, applicability, uncertainty, and why each can or cannot support the intended decision.

Did you avoid turning a trophic index, concentration, model output, grant target, or national recommendation into a universal safety grade, legal requirement, or guaranteed restoration endpoint?

Draw a source-pathway-lake-response model and give every link an evidence class, confidence, time scale, responsible reviewer, competing explanation, and next discriminating test.

Does the model preserve observations, associations, model estimates, source attribution, management eligibility, and causal conclusions as different claims?

Create a term ledger with sign convention, unit, datum, time coverage, method, evidence state, owner, uncertainty, and missing or residual terms; never assign a residual automatically to groundwater or a named source.

Are control volume, analysis period, units, storage change, missing terms, and the limits of V/Q explicit before any nutrient budget is interpreted?

Specify stations, discharge methods, event triggers, analytes and fractions, sampling and integration intervals, missing-data rules, QA/QC, and uncertainty for one tributary or outfall; state which source questions the design cannot answer.

Are concentration, delivered load, source attribution, whole-lake budget, and required reduction still distinct, and is sampling-method detail left in the load-monitoring route rather than reinvented here?

Build a practice release containing accepted, estimated, qualified, rejected, missing, and corrected records; retain the raw evidence and document which budget or source claims each record may and may not support.

Are raw, screened, validated, modeled, corrected, released, superseded, and withdrawn states visible before a value enters the nutrient ledger?

Create a versioned ledger for tributaries, direct drainage, stormwater, septic and sewer hypotheses, atmospheric input, groundwater context, waterfowl or other candidate sources, outflow, retention or export, sediment-water exchange, storage change, and missing terms; preserve low, central, and high bounds with provenance.

Do all terms share a defined period and unit, reconcile within stated uncertainty, retain missing and residual terms, and avoid assigning a legal reduction target or named-source responsibility without the controlling evidence and authority?

Audit one candidate hotspot: document objective, data vintage, model domain, cutoff rationale, sensitivity, source-mobilization-delivery evidence, landowner or Tribal controls, and what would verify or overturn the screen without naming a responsible party.

Is the map still a planning hypothesis rather than parcel-scale observation, culpability, compliance, source proof, or automatic project priority?

Build an alternatives table covering no action, protect intact areas, roofs and driveways, erosion and vegetation, septic records, agricultural and stormwater systems, operational changes, and any in-lake concept; record why each advances, pauses, or is rejected.

Does the program avoid remote source or septic diagnosis, universal BMP efficiencies and buffer widths, one-size designs, treatment selection from concentration alone, and the collapse of eligibility, design, permit, authorization, implementation, maintenance, and effectiveness?

Enter a traceable baseline and one scenario using explicit low, central, and high load estimates where available; mark unchanged sources as assumptions and missing sources unquantified, then export the complete provenance ledger.

Can another reviewer reproduce every subtotal and arithmetic planning envelope and distinguish them from TMDL, design, permit, compliance, verified reduction, causation, and lake response?

Run a return demonstration with one unresolved access condition, one method deviation, one protected-resource stop, and one task outside the approved role; separately create a gate register with controlling source, evidence reference, owner, status, expiration, scope, and stop or postpone consequence.

Are observed proficiency, assignment, supervision, permission, qualified design, permit, project authorization, and implementation represented as different decisions?

Complete a tabletop project with one rejected alternative, one unresolved gate, one authorized scope, one as-built deviation, an O&M owner and trigger, a missed-maintenance response, and a generalized or no-public-release status; leave unknowns visible.

Does the record prevent planning, recommendation, grant award, permit, contractor mobilization, installation, operational acceptance, maintenance completion, and effectiveness from collapsing into a single ‘done’ state?

Build a verification table for as-built conformance, maintenance, source change, delivered load, timing and lag, lake response, protected uses, adverse effects, null or inconclusive outcomes, corrective action, next review, and the owner of each conclusion.

Can the design distinguish installed, functioning, maintained, expected reduction, observed change, technically verified effect, no effect, adverse effect, inconclusive result, and authority-approved public claim?