The Simulator

About the Simulator

Overview

The Simulator is an interactive tool for exploring stratospheric aerosol injection (SAI) and its potential climate impacts. It is meant to help researchers, policymakers, and the broader public see what SAI could mean for our planet’s climate decade by decade.

Even under optimistic scenarios where emissions fall rapidly and carbon removal scales up, we are likely to face serious near-term climate impacts. The Simulator lets users explore what adding SAI to that mix might do — not as a prediction, but as a way to understand trade-offs.

It is part of a broader effort to democratize access to climate tools. Running a full climate model can take weeks of supercomputer time and deep expertise, and the outputs and analyses are hidden behind paywalls, written in scientific jargon, and often explore the effects of only one SAI deployment scenario (among millions of potential scenarios). With the Simulator, you can explore a broad range of comparable results in seconds, built from thousands of pre-run simulations.

The hope is that by making these dynamics visible and explorable, we can deepen conversation around climate intervention.

Like all modeling, the Simulator has limitations. Its results depend on the underlying simulations used to build it, which do not capture all feedbacks, variability, or sources of uncertainty in the climate system. These results are intended to help users understand trends and trade-offs rather than to predict precise future climate conditions.

Methodology

Under the hood, the Simulator is built from a simple idea: you don’t need to rerun a full Earth System Model every time you want to explore a new scenario. It uses a technique called “pattern scaling” to estimate regional climate responses (like temperature and rainfall) based on global mean temperature.

This approach is validated against full climate model runs (see Visioni et al., 2023), and described in detail in Farley et al., 2025 and the v1.0 Emulator Explainer (PDF).

The Simulator combines:

• A simple emulator for global mean temperature (similar to FaIR),
• Regional scaling patterns from full climate models, and
• A set of algorithms that translate user choices — like start date or target temperature — into required injection amounts, costs, and logistics.

That means you can see, instantly, how a later start date or a tighter cooling target would change outcomes, both globally and locally.

Publications and Presentations

The simulator has been cited in publications, reports, and external write-ups, including:

• Simulating the Impacts of SAI Deployment — blog overview of the methodology and first release
• v1.0 Emulator Explainer (PDF) — technical overview
• The Stratospheric Aerosol Injection (SAI) Simulator: An Open-Source Web Tool for Exploring Climate and SAI Deployment Scenarios — submitted for publication

Explore

You can try the Simulator directly at simulator.reflective.org. Start with the default scenario, or follow the walkthrough in the User Guide below to build your own.

The full Simulator codebase is publicly available on GitHub, including instructions for downloading the open access training data.

User Guide

Introduction

This guide provides a walkthrough of the Simulator’s core functions and helps you begin exploring scenarios related to Stratospheric Aerosol Injection (SAI).

You will start by learning how to build a scenario, then move on to visualizations and exploring regional effects. Finally, we will look at the advanced options for custom scenarios.

Start Up

When you open simulator.reflective.org, you will see the welcome screen. If a message appears saying “desktop required”, adjust your browser window to a wider layout.

From here, you can either jump straight into exploration or navigate through the introductory walkthrough to learn more about SAI. If you are new to the topic, we recommend spending a few minutes on the walkthrough before diving in.

Once the pop-up closes, you will arrive at the main Simulator interface, which has three key components:

1. Input Panel – where you define your scenario and configure how results are visualized.
2. Output Charts – a map and a plot that show the simulated impacts of your chosen scenario.
3. Metrics Box – a summary of deployment requirements and approximate costs.

Quick Tour

• In the input panel, you will find two main menus:
  • Build Your Scenario – define the SAI deployment and climate assumptions.
  • Configure Visualizations – choose what climate impacts and decades to display.
• The map and plot visualize your scenario’s outcomes for a world with and without SAI.
  • Use the slides on the map to toggle between these views in the map.
  • On the righthand plot, you can switch between a plot that shows how a given impact changes with time and a distribution.
• The metrics box (bottom of the screen) shows what a deployment would involve:
  • annual SO₂ requirements
  • their equivalence to volcanic eruptions
  • estimated aircraft fleet size
  • deployment costs

(Note: these metrics are rough estimates. There has been remarkably little engineering work done to date, so we must make significant assumptions about the payload of potential planes and flight costs.)

At the top of the page, you will also find menu and sharing options—these are covered later in the Advanced Features section. For additional help, click the blue question mark icon in the lower-right corner to open the Help Center and FAQs.

Simulation Inputs: Building a Scenario

Let’s begin by selecting the Simulator inputs. Although the Simulator opens with a default configuration and it can be tempting to start exploring the maps, it is important to understand the assumptions underlying the visualizations.

Selecting a baseline warming scenario:

You start by selecting a baseline warming scenario. This input captures fossil fuel emissions and decarbonization for this simulated world in which SAI is carried out. These scenarios are standard amongst climate scientists (you can learn more here).

When selecting a scenario, consider: will the world make considerable progress towards decarbonization? Will fossil fuel consumption increase? As a default, we have chosen SSP2-4.5, the “middle of the road” scenario now seen as most likely.

Next, you choose the parameters of a deployment: your SAI cooling target and when you would like the deployment to start.