Training climate thinkers for the policy fight

A new Harvard-MIT pilot asks undergraduates to turn climate evidence into action.

This spring, the Salata Institute-based Global Climate Policy Project brought together a dozen students, many from technical fields, and asked them to examine a climate problem as policymakers would.

Policymakers need clear arguments that can survive institutions, budgets, politics, and power. So each student was tasked with writing a policy memo – what should be done, who should do it, and why – that tackles an urgent issue. Topics included making methane data compatible across platforms, and how climate models often overlook small island states.

Not all the students’ ideas are ready for implementation, but that is part of the point. The memos show students testing ambitious proposals against real constraints and learning how to present them in forward-looking policy vernacular. Here, we highlight four of them.

Martina Jaramillo, a first-year at Harvard studying economics and government, examined why disaster aid in the Sahel can still fail even when quality forecast data and well-funded aid programs are already in place. Her project narrowed in on Niger, where the problem is not simply knowing that droughts or floods are coming. It is getting help to the right households in remote, insecure, or low-connectivity areas.

Jaramillo unpacked anticipatory action strategies. That is the use of early warning signs, such as rainfall deficits, drought indices, food-insecurity projections, river levels, and flood forecasts, to send aid before a crisis does its worst damage. The goal is to help families buy food, protect assets, and avoid desperate choices such as selling livestock or taking on debt.

In her policy memo, Jaramillo recommended a hybrid cash-transfer system that uses mobile payments where possible and local agents or payment points where digital systems do not reach. Aid often arrives after families have already absorbed damage. Early aid can prevent those losses.

“Once you give people these tools, then they’re better equipped for the next climate disaster, which are ever increasing in this region,” she said.

Analy Vega, a first-year at Harvard studying biology and environmental science, examined why Arctic methane monitoring remains as much a policy problem as a technical one. Her project focuses on methane released by melting permafrost in the northern Bering, Chukchi, and Beaufort Seas, where warming is accelerating thaw, remote-monitoring satellites struggle in Arctic conditions, and permafrost methane is still largely missing from national greenhouse gas inventories.

Vega argues that the central failure is governance.

“No technically adequate, institutionally resilient, or Indigenous-data-sovereign framework currently exists” for monitoring these emissions, she writes. That gap matters because better methane data affects climate targets, funding decisions, and the credibility of international climate pledges. It also affects Indigenous communities, including Inuit, Iñupiaq, Chukchi, and Yupik, whose knowledge of Arctic change has often been excluded from formal monitoring systems.

Sensors alone will not solve the problem, she cautions. A credible monitoring system needs financing, scientific standards, political realism, and authority for the communities that have been observing Arctic change the longest to have a say in what is collected, how it is stored, and who uses it.

Her memo weighs three options: expanding funding for Arctic methane monitoring, creating an Arctic Council protocol that pairs satellite data with Indigenous data sovereignty, and building a Canada-led agreement to fund shared sensors, coordinate data infrastructure, and produce standardized methane estimates.

Luna Yin, a Harvard sophomore studying environmental science and engineering, examined how the world is collecting more methane data than ever, but much of it cannot easily be compared in a way that will help reduce leaks of this powerful greenhouse gas. Her interest grew out of research she did on MethaneSAT in the lab of Steve Wofsy, which inspired her to look more closely at global methane emissions data.

Methane matters because it warms the atmosphere far faster than carbon dioxide in the short term. Satellites, aircraft, and ground sensors can all detect methane, but each sees a different part of the problem and produces data in different formats. “The current methane data that we have is really fragmented across different sources,” Yin says.

Her recommendation is for an international body such as the UN Environmental Program to establish a global methane data standard so researchers and governments can compare measurements across instruments, projects, and countries. The proposal does not require building a new monitoring system from scratch. It asks policymakers to make the systems already being built speak the same language. Once that standard is in place, UNEP could more easily strengthen its International Methane Emissions Observatory and bring in more aircraft and ground-sensor data.

Mitigating methane depends on knowing where it is being released. Without cleaner, more compatible data, governments cannot see the full picture, researchers lose time untangling incompatible files, and oil and gas companies face less scrutiny over leaks.

Sophia Zhang, a Harvard sophomore studying statistics and comparative literature, examined how small island states are forced to make climate decisions with data that often lacks the resolution to see them clearly. As her memo puts it, “Small island developing states are making existential adaptation decisions based on climate data that was not built for them.”

It is a structural problem built into climate models developed by wealthy nations. Global projections often operate at scales too large to capture islands such as Barbados or Fiji’s 332 scattered islands. That matters because governments use this data to decide where to build infrastructure, how to protect freshwater sources, and how to price climate risk.

Zhang recommends that the World Meteorological Organization fund regional climate projection hubs, starting with institutions in the Caribbean and the Pacific. Small island states would not build global modeling systems from scratch, but gain authority over downscaling, validating, and interpreting the climate projections that shape their futures. Better climate data, in this case, is also about power: who gets to define the risks, and who has to live with the consequences.