Climate change is often described through global averages: global temperature increases, mean sea-level rise, or worldwide emissions pathways. Yet communities rarely experience climate change at the global scale. People experience it locally: through flooded neighborhoods, prolonged heat waves, changing rainfall patterns, coastal erosion, and disruptions to ecosystems that support livelihoods. One of the most important challenges facing the climate science community today is therefore translating large-scale Earth system processes into meaningful local decisions.
During the CMIP 2026 session on Advancing Climate Projections and Applications: From Global Processes and Impacts to Responsible Local Decisions, including with Marine Ecosystems (POS3.1), discussions centered on this critical challenge: moving from global climate projections to actionable information for regional planning, adaptation, and ecosystem management.
From global climate models to local realities
Climate projections generated through frameworks such as the Coupled Model Intercomparison Project (CMIP) have become fundamental tools for understanding future climate risks. These Global Climate Models (GCMs) simulate interactions across the atmosphere, oceans, land, and cryosphere to project future conditions under different socioeconomic and emissions pathways.
However, there is an inherent challenge. Most GCMs operate at relatively coarse spatial resolutions, often representing the Earth in grid cells that span tens to hundreds of kilometers. Such scales are highly valuable for understanding planetary processes but are often insufficient for local decision-making.
For instance, a single model grid cell may encompass urban areas, forests, agricultural land, rivers, and coastal ecosystems simultaneously. Critical local phenomena—including urban heat islands, localized flooding, coastal circulation patterns, or ecosystem-specific climate responses—can therefore remain unresolved.
Bridging this gap requires moving beyond simply producing global projections toward creating climate information that communities and policymakers can directly apply.
Downscaling: translating global signals into actionable knowledge
A key topic of discussion during POS3.1 involved the role of downscaling approaches in transforming global climate outputs into higher-resolution local projections.
Bias Correction and Spatial Downscaling (BCSD) methods are increasingly used to refine coarse-resolution model outputs into locally relevant climate information. These methods help correct systematic model biases while increasing spatial detail, allowing projections to be represented at scales more suitable for decision-making.
Through such approaches, researchers can derive localized climate indicators including:
- Temperature anomalies relative to historical baselines
- Changes in seasonal and annual precipitation totals
- Frequency and intensity of extreme weather events
- Heat stress indicators
- Flood and drought risk metrics
Equally important is the use of multi-model ensemble synthesis. Rather than relying on a single model, combining outputs across multiple CMIP models helps address uncertainties arising from differences in model structures and assumptions. Ensemble approaches create more robust projections and improve confidence when informing adaptation decisions.
The challenge is therefore not only producing more detailed climate projections but ensuring these projections remain scientifically credible and decision relevant.
Responsible local decisions and policy applications
High-resolution climate projections increasingly influence policy and development planning.
Urban planners are using localized projections to assess how changing rainfall patterns may affect infrastructure and flood risk. Cities can evaluate how events historically considered “1-in-100-year” floods may become more frequent under future climate conditions. Similarly, projections of extreme heat can guide urban design, energy planning, and public health preparedness.
In Southeast Asia, these applications have particular relevance. Rapid urbanization, dense coastal populations, and exposure to climate hazards make localized climate information essential. In Malaysia, climate projections increasingly contribute to adaptation planning and support national efforts toward risk-informed socioeconomic development.
Importantly, climate data accessibility is also improving. Platforms that standardize and pre-process projection datasets are reducing barriers for policymakers and practitioners who may lack specialized climate modeling expertise. This democratization of climate information helps bridge the longstanding divide between climate science production and practical implementation.
Marine ecosystems: local impacts within global systems
An important dimension of POS3.1 was the inclusion of marine ecosystems within discussions of local climate applications.
Marine environments illustrate particularly well the tension between global processes and local impacts. Rising ocean temperatures, changing circulation patterns, marine heatwaves, and ocean acidification occur within a globally interconnected system, yet their consequences are experienced locally by coastal communities and ecosystems.
Coral reefs, fisheries, mangroves, and coastal biodiversity can exhibit highly localized responses to changing environmental conditions. Small-scale ocean processes—including currents, eddies, and coastal circulation—may significantly influence ecological outcomes but remain difficult to capture in coarse-resolution climate models.
For many regions, especially across Southeast Asia and small island states, improving climate projections for marine systems is not simply a scientific question but also a socioeconomic one.
Fisheries, food security, livelihoods, and coastal resilience depend on understanding how global climate signals manifest in local marine environments.
Looking ahead: climate science for decision-making
One of the strongest messages emerging from the session was that climate science increasingly operates at the interface between research and action.
Advances in model resolution, downscaling techniques, and Earth system understanding are creating opportunities to provide more relevant information for society. Yet producing increasingly sophisticated datasets alone is not sufficient.
Responsible local decision-making requires continued collaboration between climate scientists, policymakers, ecosystem experts, planners, and communities themselves. Scientific credibility must be paired with accessibility, transparency, and practical usability.
As climate impacts intensify, the question is no longer only what changes are occurring globally? Increasingly, the question becomes: what do these changes mean for specific places, communities, and ecosystems—and how do we act on that knowledge responsibly?
The discussions in POS3.1 highlighted that bridging global processes and local decisions is not merely a technical exercise. It represents a fundamental shift in how climate science serves society
