DEPLOYED TOOL
CoSMoS: Complete Stochastic Modelling System
CoSMoS is an award-winning open-source R package that generates realistic synthetic time series and spatial fields of any hydroclimatic variable (precipitation, streamflow, wind speed, humidity) in seconds. You define the statistical properties you need — CoSMoS produces the ensemble.
- Univariate, multivariate, and random field simulation of any hydroclimatic variable
- Preserves marginal distributions, correlation structure, and intermittency across scales
- Space-time storm simulation including advection from local rainfall to cyclone-scale fields
Maintained by TycheLab — Papalexiou, Serinaldi, Shook
35,000+ downloads on CRAN since 2019
INDUSTRY CONNECTION
Members of the group have sustained long-term engagement with the global insurance and reinsurance sector. For over a decade, a core group researcher held a position funded through the Willis Research Network — a programme connecting academic researchers directly to catastrophe risk modelling practice. This engagement has shaped the group's understanding of how probabilistic hydroclimatic analysis is applied in real-world risk assessment and pricing.
Reinsurance and Catastrophe Modelling
Catastrophe models depend on simulated hydroclimatic fields — wind, precipitation, flood — generated at scale, preserving spatial dependence and tail behaviour. CoSMoS was designed for exactly this class of problem. The reinsurance sector applies wind field simulation, copula- based dependence models, and stochastic precipitation ensembles in portfolio risk and pricing: these are the methods our group has developed and continues to advance.
Infrastructure Design and Flood Standards
Designing bridges, drainage systems, and flood defences requires robust estimates of extreme precipitation and streamflow — return levels, IDF curves, uncertainty bounds — across present and future climate conditions. Our methods address the full chain from observational record analysis through nonstationarity and uncertainty quantification to design-event specification under climate change.
Urban Flood Risk and Climate Adaptation
Cities investing in flood adaptation need to know which measures hold up across a range of plausible future conditions, not just under one design storm. Our probabilistic scenario generation and uncertainty quantification methods provide the framework for ensemble-based performance assessment — moving beyond static design events.
Operational Forecasting and Water Utilities
Water utilities and forecasting agencies operating at large scales need tools that are computationally efficient, physically grounded, and calibrated to local conditions. Our stochastic simulation and machine-learning surrogate work addresses ensemble-scale forecasting without the overhead of full physics-based models.
Climate Adaptation Policy and Investment
National and regional adaptation programmes need to translate probabilistic climate projections into actionable investment decisions. Our decision-support frameworks — built on robust decision-making and scenario discovery methods — connect quantified uncertainty to the governance and planning processes where it must become a decision.
Water Governance and International Policy
Where water crosses borders, risk quantification must meet governance reality. Our research connects probabilistic hydrology to water conflict analysis, SDG 6 implementation, and the institutional frameworks that determine how risk information becomes policy.
