2 days Ago
In one of his first acts in his second term as president of the United States, Donald Trump fired several hundred employees of the US National Oceanic and Atmospheric Administration (NOAA). The move triggered a lot of discussion in the climate community and beyond about how it will hurt critical weather and climate services. The climate community has been working hard to help people adapt to the effects of climate change and build resilience against consequences in the future, so it doesn’t help that the NOAA has now been downsized.
While this is unfortunate, this isn’t entirely surprising considering the NOAA was also under considerable pressure during Trump’s first term as US president. Predictions and projections The weather is local while climate is global, but a good weather forecast still requires global patterns to be captured and accounted for. Climate predictions on the other hand focus on meteorological changes that occur more slowly across multiple seasons.
Climate projections, on the other hand, offer various possible scenarios for multiple decades to come. These projection efforts are coordinated closely by the UN Intergovernmental Panel on Climate Change (IPCC). All the research centres involved in preparing these projections need to follow particular protocols as well as focus on certain previously agreed upon future scenarios.
Once every few years, the IPCC prepares a grand synthesis of all simulations from tens of models to produce an assessment report. The latest such report was issued in 2021-2022. Climate predictions on the other hand are national efforts with some coordination under the UN World Meteorological Organisation (WMO), especially for global observational systems.
Climate predictions need the models to be prepared by ‘initialising’ them before each forecast begins. Data from all the relevant sources — including weather-monitoring stations and satellites — are fed into the models responsible for simulating ocean, atmosphere, and land systems. Different prediction centers follow different methodologies during this data assimilation step.
Since no single country can cover the globe with its observation systems, global coordination in this enterprise is inevitable. Climate predictions also tend to be internal efforts. The participating countries under the WMO also merge multiple such predictions to produce a so-called multi-model ensemble.
But as Trump’s decision to take a sledgehammer to the NOAA indicates, we may need to make climate predictions the same way we prepare the well-coordinated climate projections, with redundancies in the global to regional predictions, in order to protect the overall prediction enterprise from political vagaries and other debilitating perturbations. On the plus side, a globally coordinated climate predictions system will also bring about higher-resolution models and more accurate predictions for all countries. They could also help governments respond better to the rapidly emerging suite of extreme events.
Many such events assailed the earth during the record-setting global warming of 2023 and 2024, and this is likely to continue in 2025. Coordinated predictions will also benefit from regular global stocktakes that record the numbers of events that were correctly predicted and how many provided meaningful inputs to governments to prepare for, mitigate, manage, and recover from climate-related disasters. Towards K-scale modelling The other critical question is whether the predictions that are available have the requisite spatial resolution required for governments to respond to location-specific disasters.
The answer is a clear ‘no’. Even climate projections don’t offer information at scales required for regional and local adaptation and resilience-building. There have been repeated calls now to move beyond the current suite of coarse resolution models used for predictions and projections, to move towards the use of 1-km scale, or K-scale, models.
Such modelling will require considerable computing resources that no one country can afford — yet it also offers a valuable opportunity to incorporate climate predictions into international climate action. As other experts have also suggested, a global effort can make this happen, with each region and country receiving more accurate and more location-specific early warnings and seasonal outlooks. Such coordinated K-scale modelling for climate predictions and projections should be a high priority.
Need for cost-benefit analysis This author has already suggested that modellers focus less on projecting the climate until the year 2100 and more on that at the more socially relevant timescales of multiple years to a decade or two. Modellers currently understand that uncertainties in projections for the first couple of decades are dominated by the natural variability of the climate system plus limitations in the models themselves. The ‘IPCC-class’ models thus consider innovation in energy and transportation, population growth, carbon capture, and the effects of various climate policies to understand the possible levels of warming by 2100.
These projections are envisioned to capture all eventualities, including Russia’s invasion of Ukraine, the West Asia conflict, and so on — yet it misses the downsizing of the NOAA and the exit of a highly industrialised country from international climate talks. A crucial requirement to build resilience in any sector that depends on government funding or market forces is a cost-benefit analysis that justifies its existence. Obviously, it doesn’t suffice to claim that the value of a service is self-evident.
If one prediction center is doing better than another, questions can easily be raised about the size of their workforces and their operational efficiency. If a smaller workforce is consistently able to make better predictions, we need to understand how and replicate it. This goes beyond Trump, who has no regard for the effects of the US’s continuing industrialisation on the world’s climate.
It is instead about a world in which modelling centres and efforts still matter, where public funding is limited even when it isn’t a zero-sum game, to the extent that the centres need to rationalise their contributions rather than sidestepping audits in the name of a climate “emergency”. Each center has to be prepared to defend itself with convincing answers to the difficult questions. They may not like the questions but they can still be fair in the larger economic picture.
In the same vein, a cost-benefit analysis of the IPCC’s projections is also needed to make the climate centres more resilient. Overall, it is crucial to justify the need for continued long-term projections. All these factors underscore the need for resilient climate prediction efforts as a global enterprise.
Any negligence or delay on this front will only leave the centres vulnerable to being pared back themselves. Any system is only as strong as the weakest link. Raghu Murtugudde is retired professor, IIT Bombay, and emeritus professor, University of Maryland.
Published - March 07, 2025 05:30 am IST Copy link Email Facebook Twitter Telegram LinkedIn WhatsApp Reddit weather science / climate change.
