climate-risk-adaptation-method

# Climate Risk Adaptation Method ## Freeze the basis 1. Record system/asset/population boundary, geography, present-day baseline, hazard sources, scenarios/horizons, models/ensembles/downscaling, units, decision question, equity/community limits and qualified owners. 2. Assign stable IDs to hazards, sources, scenarios, horizons, exposed objects/groups, vulnerabilities, dependencies, consequences, options, pathways, triggers and decisions. 3. Run hazard/scenario, exposure/vulnerability/consequence, and option/pathway work independently. Join only complete source-addressable results. ## Apply the method - Keep physical and transition risk, and acute and chronic hazards, distinct. - Link every risk through explicit hazard, exposure and vulnerability evidence. Preserve baseline, model, ensemble, scenario, horizon, spatial support, units and source date. - Report ranges, member counts, confidence basis, model/data/structural uncertainty and gaps. Do not average incompatible scenarios or treat model spread as total uncertainty. - Estimate physical or financial consequence only with a supplied validated vulnerability/damage function and compatible units; otherwise keep consequence qualitative and evidence-linked. - Trace direct, indirect and cascading dependencies without double counting. Record differential exposure, adaptive capacity, privacy, distributional effects and community evidence. - Structure options by prerequisites, lead time, life, reversibility, performance evidence, triggers, co-benefits, maladaptation and residual risk. Build conditional pathways; do not invent scores, probabilities, costs, rankings or default scenarios. ## Produce evidence Use all five assets: hazard-source register, exposure/vulnerability/dependency ledger, scenario matrix, option/pathway/trigger register, and decision pack. Every material row carries ID, unit, source/version/date, owner/reviewer, applicability, uncertainty, assumptions and decision status. Stop on incompatible baselines, missing scenario/model metadata, unjustified downscaling, unvalidated consequence functions, sensitive locations, missing community authority or professional-design questions. ## Review discipline - Preserve observed climate, model output, scenario assumption, exposure fact, vulnerability function, attributed stakeholder statement and analyst inference as different evidence kinds. - A scenario is not a forecast probability. Never combine scenario outputs into an expected value unless an authorized decision method provides probabilities and supports that use. - Match hazard support to exposure geography and decision scale. Record coarse-resolution, bias-adjustment, model dependence and local-data limitations beside every derived result. - Separate direct, indirect and cascading consequences; name dependency direction and apply a declared double-counting rule. - Evaluate each adaptation option across scenarios for prerequisites, lead time, useful life, reversibility, trigger observability, residual risk, maladaptation and distributional effect. - Record who bears costs, receives benefits and controls the decision. Missing affected-community evidence remains a gap, not assumed consent. - At the join, show robust enabling actions and scenario-contingent pathways without converting evidence into a professional recommendation or capital ranking. ## Authority boundary Do not issue engineering, actuarial, insurance, legal, disclosure, emergency, land-use, relocation, procurement, investment or capital decisions. Require qualified climate, engineering, actuarial, emergency, regulatory, finance, public/community and asset-owner review. Read [upstream provenance](references/upstream.md). Apply only its bounded concepts; the local Skill is the runtime source of truth.