Climate Models
Global and Regional Climate Models
Global Climate Models (GCMs) simulate climate across the entire planet. They work by dividing Earth's surface into a three-dimensional grid, where each cell represents a specific location. Cells are typically 100–200 km wide horizontally and about 1 km deep vertically. Within each cell, the model calculates atmospheric and oceanic processes, land surface interactions, and sea ice dynamics, then passes that information to neighboring cells at each time step.
The tradeoff with GCMs is resolution. A 100 km grid cell can capture large-scale circulation patterns, but it can't resolve a thunderstorm or a mountain valley. That's where Regional Climate Models (RCMs) come in. RCMs are nested within GCMs: they take the GCM's output as boundary conditions and then run a higher-resolution simulation (typically 10–50 km) over a specific area. This makes them useful for studying climate change impacts in places like California or the Mediterranean, where local topography and coastlines matter a lot.
- Well-known GCMs include the Community Earth System Model (CESM) and the Hadley Centre Global Environment Model (HadGEM)
- RCMs sacrifice global coverage for finer spatial detail, which is critical for regional planning and impact studies
Energy Balance Models
Energy balance models are among the simplest climate models. They're built on one core principle: incoming solar radiation must balance outgoing thermal radiation for Earth's temperature to remain stable. If something tips that balance, the planet warms or cools.
These models account for factors like albedo (how reflective Earth's surface is), greenhouse gas concentrations, and heat transport between latitudes. Because they simplify the climate system so heavily, they're not used for detailed forecasting. Instead, they're valuable for isolating how a single variable, such as a change in atmospheric or a shift in solar output, affects global temperature.
- The Budyko-Sellers model is a classic example that explores how ice-albedo feedback can lead to dramatically different climate states
- The Daisyworld model demonstrates how biological feedback (in this case, hypothetical black and white daisies altering surface albedo) can regulate planetary temperature
Earth System Models
Coupled Models
Earth System Models (ESMs) go beyond climate models by simulating interactions between the atmosphere, ocean, land surface, cryosphere, and biosphere all at once. They track the exchanges of energy, water, carbon, and other substances between these components, which means they can capture feedbacks that a standalone atmospheric or ocean model would miss entirely.
The term "coupled" refers to how these models are built. Separate models of individual components (say, an atmospheric model and an ocean model) are linked so they exchange information and fluxes at each time step. For example, the ocean model sends sea surface temperatures to the atmospheric model, and the atmospheric model sends wind stress and heat fluxes back to the ocean model.
- This coupling is what allows ESMs to study critical feedbacks like ocean-atmosphere interactions (e.g., El Niño) or land-atmosphere interactions (e.g., how deforestation changes regional rainfall)
- CESM and HadGEM are both ESMs, meaning they function as GCMs and include coupled biogeochemical and ecological components
Biogeochemical Models
Biogeochemical models focus specifically on how elements and compounds cycle through the Earth system. They track the movement and transformation of substances like carbon, nitrogen, and phosphorus through living organisms, soils, the atmosphere, and the ocean.
Key processes these models represent include photosynthesis, respiration, decomposition, and nutrient uptake. By simulating these cycles, researchers can study questions like: how much carbon will terrestrial ecosystems absorb as levels rise? Or how will warming soils release stored nitrogen?
- The CASA model (Carnegie-Ames-Stanford Approach) estimates global carbon fluxes from vegetation and soils
- The LPJ-DGVM (Lund-Potsdam-Jena Dynamic Global Vegetation Model) simulates how vegetation types shift in response to changing climate, which in turn alters carbon and water cycling
Simplified Models
Box Models
Box models strip the Earth system down to a set of interconnected reservoirs, or "boxes." Each box represents a major component (the atmosphere, the surface ocean, the deep ocean, terrestrial biomass, etc.), and arrows between boxes represent fluxes of energy, water, or chemical species.
What you gain is computational efficiency and conceptual clarity. Because box models don't resolve spatial detail within each reservoir, they run quickly and make it easier to see how the overall system responds to a perturbation. For instance, a global carbon cycle box model can show how a pulse of emitted into the atmosphere gradually redistributes among the ocean, land, and atmosphere over centuries.
- Box models are especially useful for studying long-term, system-wide behavior rather than regional detail
- An ocean box model might represent the surface and deep ocean as two boxes connected by thermohaline circulation fluxes
Integrated Assessment Models
Integrated Assessment Models (IAMs) are unique because they bridge the natural and social sciences. They combine models of the economy, energy systems, land use, and the physical climate system into a single framework.
The purpose is to analyze how human decisions and Earth system responses interact over time. For example, an IAM can estimate the economic costs of a carbon tax, project how that tax would reduce emissions, and then feed those reduced emissions into a climate module to see the resulting temperature change. This makes IAMs a primary tool for policymakers evaluating options like carbon pricing, renewable energy subsidies, or land-use regulations.
- The DICE model (Dynamic Integrated Climate-Economy) is a relatively compact IAM that links a simple climate model to an economic growth model
- The GCAM (Global Change Assessment Model) is more detailed, with explicit representations of energy technologies, agriculture, and land use across world regions
- Because IAMs integrate so many disciplines, their results depend heavily on assumptions about economic growth, technology development, and discount rates, so it's important to interpret their outputs as scenarios rather than precise predictions