+14.2% max 1-day rainfall
Full-ensemble projection with 32 of 34 CMIP6 models passing skill screening and 78% trusted-model agreement on increase.
Climate-risk evidence for pilots
Know which climate models to trust before using them for 2050 risk decisions.
Arasense helps cities, infrastructure owners, and climate-risk consultants produce defensible evidence for flood-driving rainfall, heat, and drought. For each location, the platform shows which models earn trust, which models are rejected, and what the trusted models imply for mid-century risk.
Trust
Local model scoring before projection, using the Aras Diagram.
2050
Mid-century hazard reports with uncertainty and model agreement shown.
Pilot
Start with one geography, one decision question, and one evidence pack.
Concrete outputs beat broad claims
Arasense is strongest when a visitor can see real evidence quickly. The public case-study page packages the current proof around Bologna rainfall, the Italian flood-exposure portfolio, and a Puglia-ready pilot path.
Rome ranks highest
Portfolio view ranks cities by worsening flood-driving rainfall signal, with Rome at +19% and Venice at +3% in the sample.
Bari as the next local proof
A focused Bari/Puglia pilot can turn the same workflow into local evidence for adaptation, infrastructure, or advisory conversations.
Research-grounded methods, framed for institutional use
Arasense is built around interpretable model evaluation and transparent evidence generation. The platform connects the Aras Diagram framework with operational geospatial data sources and validation-stage screening workflows.
Aras Diagram
Structured decomposition of model behavior across bias, variability, and phase alignment, supporting clearer judgement than a single aggregate score.
ERA5-Land and CMIP6
Climate diagnostics and projections are framed around established reanalysis and global climate model datasets, with model skill made explicit.
Sentinel-1 evidence
Flood-screening pilots can be compared against satellite-derived evidence windows to document where the workflow performs and where it remains uncertain.
A private analytical console for climate-risk evidence
Arasense combines climate model evaluation, bias-aware interpretation, projection reporting, portfolio screening, and validation-stage flood analytics into one technical stack. The objective is not another generic dashboard. The objective is a more defensible basis for planning, screening, and risk communication.
Climate Intelligence
Compare climate model behavior through Aras Diagram signal decomposition, model-trust tiers, and transparent ranking logic.
Projection Reports
Estimate mid-century changes for rainfall, heat, drought, and scenario differences using trust-weighted models.
Portfolio Screening
Rank locations by worsening hazard signal so teams can see which exposures deserve attention first.
The enhanced console is built around spatial decisions
The current Arasense console is available through private demos, pilot studies, and selected collaborations. Users select climate points and flood-screening regions directly on a map, then generate structured outputs for model trust, projections, portfolio ranking, and validation review.
Map-first workflow
Click, draw, analyze
Set a climate region of interest with one map click, draw flood-screening boxes, and keep the spatial inputs synchronized across forms and reports.
Model trust
Aras Diagram plus trust engine
Score models by bias, variability, and alignment, then surface which models are credible enough to drive projections at the selected location.
Hazard reporting
2050 projections and multi-hazard profiles
Generate mid-century reports for rainfall extremes, heavy-rain frequency, heat, and drought, including SSP2-4.5 vs SSP5-8.5 comparisons.
Flood pilot
Terrain, climate, and Sentinel evidence
Combine hydrological graph structure, climate-driven precipitation features, GNN screening probabilities, and Sentinel-1 validation windows for regional pilots.
Start with teams that must explain climate-risk decisions
The first market should stay focused: public authorities, infrastructure owners, and climate/adaptation consultants that need transparent climate evidence for planning, prioritization, advisory work, or investment-facing communication.
Public and infrastructure
Authorities, utilities, and asset owners
Support resilience planning, regional screening, asset prioritization, and early-stage adaptation decisions with more structured evidence.
Advisory and finance
Consultancies, analysts, and risk teams
Strengthen client deliverables, due diligence workflows, and comparative climate-risk assessments with clearer interpretability.
Why the approach is credible
Scientific basis
- Aras Diagram framework for structured climate-model evaluation.
- Model Trust Engine for location-specific skill tiers and projection weights.
- Regional and multi-model assessment logic grounded in research practice.
- Focus on bias, variability, and alignment rather than presentation-only summary scores.
Technical basis
- FastAPI backend for deployable diagnostics, reports, and API workflows.
- Map-first console with climate diagnostics, projection comparison, portfolio ranking, and raw JSON validation.
- Graph neural network flood pipeline currently scoped to validation-stage regional pilots with Sentinel-1 comparison.
What clients should expect
Faster first-pass insight
Arasense is aimed at shortening the path from complex data to a first defensible interpretation, especially where teams need to compare scenarios, rank locations, or communicate uncertainty clearly.
How work can start
Pilot-led engagement
Early collaborations can begin with a geography, asset class, or decision question, then expand into repeatable workflows as the value is demonstrated.
Validation note
Flood outputs are validation-stage screening evidence. They are not a replacement for hydraulic modelling, field validation, local calibration, or engineering-grade flood forecasting.
The clean first sale is a focused pilot
The first sale should not be a large platform promise. It should be one geography, one decision question, and one defensible evidence pack that proves the method matters.
Near-term applications
One geography, one question
Examples: Bari coastal rainfall, Bologna flood-driving rainfall, heat-risk screening for infrastructure, or a small portfolio ranking for an advisory client.
Delivery model
Custom pilot, then platform access
Start with focused pilots or decision-support engagements, then expand into live product access and private API workflows as needs mature.
Contact
Bring one location or portfolio question.
Arasense is currently best shared through guided demos, pilot studies, and selected institutional collaborations.