Philippines WWC APAC Desk

AI Weather Forecasting in the Philippines: Can Better Data Help Protect Farms, Fisheries and Businesses?

AI Weather Forecasting in the Philippines: Can Better Data Help Protect Farms, Fisheries and Businesses?

The Philippines is expanding the use of artificial intelligence in weather forecasting, with the Department of Science and Technology (DOST) and the Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA) working with technology partners to produce more detailed and frequently updated forecasts. The bigger challenge now is turning those forecasts into decisions that help farmers, fisherfolk, local governments and businesses prepare for changing weather conditions.

The country already has an AI-powered forecasting project under development with US-based Atmo Inc. Its potential value goes beyond providing another weather map: more detailed forecasts could help users assess conditions at a local level and plan activities that are sensitive to rainfall, temperature, wind and other weather variables.

But there is an important distinction between developing forecasting technology and deploying it across industries. Official records show that the government is evaluating AI forecasts and strengthening climate-data sharing, particularly for agriculture and fisheries. They do not yet establish that airports, energy companies or private farms have signed separate agreements to use the system.

How the Philippine AI weather project works

The DOST Advanced Science and Technology Institute (DOST-ASTI) identifies the initiative as the AI-Powered Weather Forecasting for a Resilient Philippines, or AI-4RP.

The project is a collaboration between DOST-ASTI and PAGASA, with Atmo Inc. providing the artificial intelligence and deep-learning forecasting system through an annual software subscription agreement.

DOST-ASTI’s 2024 annual report says the publicly accessible Atmo model can produce forecasts extending up to two weeks, with weather updates every 15 minutes. The model was initially deployed for short- and medium-range forecasts before its second version was expanded to cover the entire Philippines. asti.dost.gov.ph

The project’s technical design also aims for forecasts at a resolution of 2 kilometres by 2 kilometres. In practical terms, this level of detail is intended to help users examine weather conditions over smaller geographic areas rather than relying only on broad regional forecasts. DOST-PCIEERD

That does not mean every forecast will be correct at every location. Weather prediction remains uncertain, particularly when thunderstorms, rapidly changing wind conditions and local terrain influence what happens over a short period.

Why AI forecasts still need PAGASA evaluation

One of the most important details in the government’s project record is that AI forecasts are not simply being accepted without testing.

DOST-ASTI reported that PAGASA evaluates the model through specialised verification procedures, quarterly performance reviews and comparisons with numerical weather prediction models. Operational forecasters are also surveyed to assess the model’s accuracy and consistency.

The AI forecasts have been incorporated into daily discussions for Impact-Based Forecasting in Metro Cebu and Manila. This approach considers not only the expected weather but also the possible consequences for people, infrastructure and economic activity. asti.dost.gov.ph

This evaluation process matters because a forecast that identifies rain over a wide area may not provide enough information for a specific decision.

A local government deciding whether to prepare evacuation centres, for example, needs information that can be interpreted alongside flood exposure, river conditions and the location of vulnerable communities. A farm operator needs to consider crop stage, soil moisture and water availability as well as rainfall forecasts.

AI can help process weather information, but local expertise remains essential when translating predictions into action.

Agriculture and fisheries already have a data-sharing framework

Agriculture is among the clearest areas where the government has formalised the use of weather and climate information.

On Sept. 3, 2026, the Department of Agriculture and PAGASA signed a Memorandum of Agreement on weather and climate data-sharing under the Adapting Philippine Agriculture to Climate Change Project. The seven-year programme involves the two agencies and the Food and Agriculture Organization of the United Nations, with support from the Green Climate Fund.

The agreement provides for regular sharing and joint use of weather and climate data. It also formalises the joint preparation and dissemination of agrometeorological advisories through the Department of Agriculture’s National Climate Information Service Centre, regional and provincial centres, and PAGASA. USDA Portal

For farmers, better access to timely information could help with planting schedules, irrigation planning and decisions about when to carry out field activities. Fisherfolk could also benefit from weather information that supports decisions about sea conditions and operational safety.

The need is particularly relevant as PAGASA’s Nov. 4 climate advisory confirmed a very strong El Niño in the tropical Pacific, expected to persist until the first half of 2027 and potentially peak by the end of 2026. The agency warned of increased rainfall-deficit risks and possible dry spells and drought in several parts of the country. PAGASA

WWC has previously examined the wider consequences of this climate outlook in its report on the possible economic impact of El Niño on Philippine agriculture.

A second AI partnership is being developed for climate resilience

The Atmo project is not the only artificial intelligence initiative involving PAGASA.

On Oct. 28, 2026, PAGASA announced an agreement involving South Korea’s DI Lab Inc. and Skymap Global Philippines for Project CLAIRE, short for Climate AI for Resilience and Early Warning.

The project is supported by the Korea International Cooperation Agency under its Creative Technology Solution programme. Its stated purpose is to localise and evaluate AI-powered weather monitoring and forecasting technologies that could complement existing forecasting capabilities.

The agreement brings together PAGASA’s meteorological expertise, DI Lab’s AI technology and Skymap Global Philippines’ local support role. The official announcement says further implementation details and expected outputs will be guided by the agreement and subsequent arrangements among the partners. PAGASA Public Files

Project CLAIRE and AI-4RP should not be treated as the same programme. They involve different technology partners and separate arrangements, although both seek to strengthen the country’s use of advanced forecasting tools.

The additional partnership illustrates the government’s broader interest in testing technologies that could improve weather monitoring and early warning systems.

What about airports and energy companies?

More detailed weather information could also be useful to aviation and energy operations.

Airports and airlines need weather information for flight planning, visibility assessments, wind conditions and decisions involving potentially disruptive weather. Energy providers monitor rainfall, temperature and weather conditions that can affect electricity demand, renewable-energy output and infrastructure operations.

However, the existence of an AI forecasting platform does not establish that these industries are already using it commercially.

The public records reviewed for this report confirm the national forecasting project, PAGASA’s evaluation activities and the formal data-sharing agreement for agriculture and fisheries. They do not confirm specific contracts, subscriptions or operational deployments with individual airports, power companies or private agricultural businesses.

That distinction is important. Industry interest is not the same as an implemented partnership, and a government research project is not automatically a commercial service available to every organisation.

For context, WWC has also reported on the growing agricultural risks associated with dry conditions in Davao Occidental, illustrating why more localised weather information could matter to communities with different rainfall conditions.

Why access and reliability will determine the project’s value

A forecasting system becomes useful when people can understand its information and act on it.

For farmers and fisherfolk, that means advisories must be accessible, timely and relevant to local conditions. For local governments, forecasts need to connect with flood maps, evacuation plans and disaster-response procedures. For businesses, the information must be reliable enough to support operational planning.

The government will also need to explain how the AI forecasts compare with established models, how often their performance is reviewed and how users should interpret uncertainty.

PAGASA’s continued role is especially important. AI forecasts can provide additional information, but official warnings and public safety guidance must remain grounded in the agency’s meteorological assessment and established procedures.

The country’s investment in AI weather forecasting could eventually benefit several sectors. Its success, however, will depend less on the technology’s novelty than on forecast quality, transparent evaluation and whether the information reaches the people who need it.

What happens next

The next stage for AI-4RP involves continued model evaluation, technical development and decisions on the project’s future operating arrangements. The Department of Agriculture and PAGASA will also need to carry forward their data-sharing framework so that weather information becomes more useful to farming and fishing communities.

Project CLAIRE, meanwhile, is expected to proceed under its own implementation arrangements with the Korean and Philippine partners.

For now, the public record shows that the Philippines is developing and testing AI-supported weather services, while strengthening the systems that connect climate information with agriculture and disaster preparedness.

The decisive question is whether these projects can turn more detailed forecasts into better decisions on the ground.

Frequently asked questions

What is AI-4RP in the Philippines?

AI-4RP stands for AI-Powered Weather Forecasting for a Resilient Philippines. It is a DOST-ASTI and PAGASA project using Atmo Inc.’s artificial intelligence and deep-learning forecasting technology.

How frequently does the Atmo AI weather model update forecasts?

DOST-ASTI’s 2024 annual report says the publicly accessible model provides weather updates every 15 minutes and forecasts extending up to two weeks.

Can AI weather forecasts help Filipino farmers?

They could support farm planning by providing more detailed weather information. The Department of Agriculture and PAGASA have also formalised data-sharing and the joint dissemination of agricultural weather advisories.

Does AI replace PAGASA forecasters?

No. The government’s project records describe AI forecasts as an additional tool that PAGASA evaluates against established numerical weather prediction models. Forecasters remain important in interpreting information and issuing official guidance.

Are Philippine airports and power companies already using the AI system?

The primary records reviewed confirm the national project and its agricultural data-sharing applications, but do not establish specific operational contracts with individual airports or energy companies.

Get our stories first on Google

More in Philippines

See all in Philippines