Skip to content
Lake and pond expertiseSoftware, monitoring, and consulting
Learning CenterField methods chapterFind Critical Nutrient Source Areas Without Turning a Map Into a Verdict
Field method · Source-area verification

Find Critical Nutrient Source Areas Without Turning a Map Into a Verdict

Screen where nutrient sources, mobilization, and connected transport pathways may coincide, then verify the evidence, uncertainty, access, privacy, and treatability before prioritizing action.

For
Watershed coordinators, Tribes, conservation districts, municipalities, lake associations, agricultural and shoreland programs, consultants, GIS analysts, monitoring teams, and community partners
Reading time
25 minutes
Reviewed
Next review
Direct answer

What to do first

A critical source area is a planning hypothesis about the overlap of a nutrient source, a mechanism that mobilizes it, and a connected pathway that can deliver it to the receiving water during a relevant period. Begin with a documented decision and a versioned conceptual model, then preserve the origin, date, scale, resolution, processing, uncertainty, and permitted use of every layer. Keep source potential, transport potential, delivery evidence, observed water quality, modeled estimates, records, and field observations separate. PLET is a planning-level estimator and SPARROW is regional; neither is a property-level measurement or verdict. A high score, land use, proximity, modeled contribution, septic record, or monitoring result does not establish culpability, a violation, or a universal treatment priority. Do not invent a universal CSA cutoff: test rankings and thresholds against sensitivity, data gaps, season, model assumptions, and the decision consequence. Verify provisional areas through lawful, consent-based field observations, records review, hydrologic evidence, and representative monitoring. Protect household, farm, business, Tribal, archaeological, and infrastructure information; use the minimum geography and access needed. Publish a bounded confidence state and a revision trigger, then evaluate feasibility, permissions, equity, maintenance, and expected pathway interruption separately from source ranking.

Use this guide to
  • Define a decision-specific CSA hypothesis without assigning blame or regulatory status
  • Preserve the lineage, scale, uncertainty, and permitted use of every screening input
  • Separate source potential, mobilization, transport, delivery, observations, and model estimates
  • Test provisional rankings and thresholds instead of presenting a universal cutoff
  • Verify priority areas through consent-based field evidence and responsible records review
  • Protect sensitive information and publish confidence, limitations, and revision triggers

1. Define the decision and the CSA hypothesis

The screen must serve a bounded planning decision, not an open-ended search for a responsible property.

Record the receiving-water concern, relevant nutrient form, season or events, decision owner, geographic extent, authority, and consequence of a false positive or false negative. State whether the output will prioritize monitoring, outreach, records review, feasibility study, or voluntary conservation planning.

Define a provisional CSA as the intersection of source potential, mobilization, and connected delivery potential. It is not proof of contribution, responsibility, noncompliance, or treatment need.

Sources: [2], [1]

2. Keep source, mobilization, transport, and delivery distinct

A source cannot affect the lake unless relevant mass is mobilized and reaches a connected pathway.

Evidence domains that must not be collapsed
DomainExample evidenceLimit
SourceNutrient use, storage, generation, or inventoryDoes not prove release
MobilizationRunoff, erosion, overflow, leaching, or disturbance mechanismDoes not prove lake delivery
TransportFlow path, ditch, drain, groundwater, or conveyanceConnection may vary by event
DeliveryObserved or supported mass reaching the receiving systemMust match period and scale

Sources: [2], [1]

3. Build a versioned data and assumption register

A polished map is not reproducible unless every layer and transformation has lineage.

  • Owner, origin, collection date, update date, and license or permitted use
  • Spatial scale, resolution, positional accuracy, missingness, and censoring
  • Processing, reclassification, weighting, joins, and coordinate system
  • Model version, calibration domain, base year, scenario, and applicability
  • Privacy class, retention rule, access owner, and public-display geography

Sources: [1], [6]

4. Inventory plausible source categories without parcel inference

Use multiple evidence streams for wastewater, agriculture, urban runoff, shoreland, and other hypotheses.

Record point discharges, onsite wastewater hypotheses, livestock and cropland systems, fertilizer and soil-loss hypotheses, urban and road runoff, construction, shoreland disturbance, atmospheric deposition, wildlife, and legacy stores where applicable. Absence from a readily available dataset is not evidence of zero contribution.

A pumped septic-system record documents a service event; it does not by itself demonstrate system performance or exclude wastewater transport. Likewise, land cover or nutrient application does not equal delivered load.

Sources: [7], [8], [2]

5. Match tools and models to their supported scale

Screening estimates are useful when their scope is explicit and harmful when treated as observations.

EPA's PLET can support planning-level estimates using watershed characteristics and management scenarios. USGS SPARROW relates monitoring and landscape information at regional scales. Preserve model identity, version, inputs, calibration domain, base year, scenario, and output units.

Do not downscale or re-label a model output as a measured parcel contribution. Compare modeled and observed evidence only after reconciling spatial boundary, period, nutrient form, and uncertainty.

Sources: [3], [6]

6. Make uncertainty and sensitivity visible

The rank order may depend more on assumptions and missing data than on actual differences among areas.

  1. Vary assumptions

    Test reasonable weights, thresholds, connectivity rules, seasons, source factors, and missing-data treatments.

  2. Compare ranks

    Flag areas whose position changes materially and identify the assumptions driving that change.

  3. Map confidence

    Display evidence sufficiency and uncertainty separately from priority score.

Sources: [2], [6]

7. Treat ranking and cutoffs as provisional decisions

No national score or percentile is a universal boundary between critical and noncritical land.

Document why a ranking method fits the decision and how many areas can realistically be verified. If a working cutoff is needed, label it as local and provisional, evaluate alternatives, and show inclusions and exclusions under sensitivity cases.

Do not imply that a low rank proves no delivery or that a high rank proves harm. Use tiers to schedule additional evidence, not to manufacture certainty.

Sources: [2], [1]

9. Evaluate treatability after screening

Priority evidence and a practical, lawful intervention are different questions.

For each verified or supported area, identify the pathway to interrupt, plausible system of practices, expected mechanism, scale, access, owner interest, constraints, permissions, cost, maintenance, co-benefits, displacement risks, and evidence needed for design. Retain no-action, monitoring, outreach, repair, prevention, and policy alternatives as appropriate.

Do not assume a BMP label carries a universal efficiency or that a national conservation standard is a site design. Current state and local criteria, field-office guidance, qualified design, and applicable authority control.

Sources: [2], [1]

10. Protect people, publish limitations, and schedule revision

Useful prioritization can be shared without converting sensitive evidence into a public accusation map.

Apply data minimization, role-based access, aggregation, retention limits, consent, and review appropriate to household wastewater, farm operations, businesses, Tribal resources, archaeological sites, and infrastructure. Respect Tribal jurisdiction and data governance; do not presume a county or state workflow controls Tribal lands.

Publish methods, source dates, uncertainty, confidence state, exclusions, verification status, appropriate uses, prohibited uses, feedback route, owner, and revision triggers. Correct a map when new records, field evidence, land management, drainage, or model versions materially change the interpretation.

Sources: [1], [9]

Evidence base

Sources and review notes

Educational watershed nutrient-planning guidance only. This material is not a source-attribution, liability, compliance, impairment, permit, engineering-design, septic-system, agricultural nutrient-management, pesticide-use, public-health, or treatment decision. It does not establish a universal phosphorus target, critical-source cutoff, buffer width, setback, pumping interval, fertilizer rate, storm design, BMP efficiency, load reduction, or expected lake response. Current state, territorial, Tribal, federal, watershed, and local authorities; adopted standards and TMDLs; property and access rights; current NRCS state Field Office Technical Guide materials; product labels; qualified professionals; and site-specific evidence control. LakeTech does not identify responsible parties or authorize work.

  1. Handbook for Developing Watershed Plans to Restore and Protect Our WatersU.S. Environmental Protection Agency · agency guidance
  2. Critical Source Area Identification and BMP Selection: Supplement to Watershed Planning HandbookU.S. Environmental Protection Agency · agency guidance
  3. Pollutant Load Estimation Tool (PLET)U.S. Environmental Protection Agency · reference
  4. Guidance: Monitoring and Evaluating Nonpoint Source Watershed ProjectsU.S. Environmental Protection Agency · agency guidance
  5. Nonpoint Source Monitoring: TechNOTESU.S. Environmental Protection Agency · reference
  6. Everything You Need to Know about SPARROWU.S. Geological Survey · reference
  7. Septic System Impacts on Water SourcesU.S. Environmental Protection Agency · agency guidance
  8. Sources and Solutions: StormwaterU.S. Environmental Protection Agency · agency guidance
  9. Tribal Nonpoint Source ProgramU.S. Environmental Protection Agency · agency guidance