Cross-sector partnerships in MHEWS
MHEWS cannot be designed and executed by disaster management and meteorological agencies alone. They often require partnerships with telecommunications regulators and operators, research institutions, humanitarian organizations, media, health, education and other sectoral ministries, and community structures, all working towards a common purpose. EWCM illustrates the range of stakeholders involved. It was developed by ITU with Microsoft's AI for Good Lab and the satellite company Planet Labs, and draws population data from the Institute for Health Metrics and Evaluation at the University of Washington. As in Liberia, the products are then validated with national telecommunications regulators and mobile operators, who can use them for CB. The example case studies below further demonstrate how these partnerships work and point to lessons and best practices.
Give a single authority the mandate to convene.
A single authority with the mandate to convene ensures consistent, trusted, and timely coordination and communication of early warnings. By enabling all stakeholders to act on a common message, this approach reduces confusion, delays and the likelihood conflicting information. This is applicable both in general, but also in specific contexts of conflict and fragility. For instance, in early 2026, Kuwait had an urgent need for a public warning system amid active regional conflict in the Middle East. It had no operational mechanism to send geo-targeted emergency alerts to mobile users across its three operators. In the face of this challenge, the Communications and Information Technology Regulatory Authority worked with the private alerting vendor Intersec to bring a nationwide CB system into operation in under 10 days, whereas the industry norm is several months. This timeline was made possible by the regulator’s authority to mandate participation and partnership with a private vendor. Rather than waiting for local infrastructure, Intersec deployed the CB system on its secure cloud in France while the system was set up locally. When the local infrastructure was ready, the system was migrated to sovereign infrastructure.
Involve the private sector as co-designer rather than supplier.
Along the Hong Kong–Singapore flight corridor, the Hong Kong Observatory and the Meteorological Service Singapore developed convective weather products that were tested by pilots from Cathay Pacific and Singapore Airlines and by air traffic controllers, whose feedback on usability, workflow integration and barriers to uptake drove redesign. In Viet Nam, Google and the private forecasting company Weathernews Inc. worked with the national hydrometeorological administration to integrate AI-derived flood and cyclone forecasts into an operational dashboard before the 2025 typhoon season. In Belize, a specialist from the radar manufacturer Leonardo Germany GmbH [BV1] delivered hands-on training that allowed national technicians to repair and recalibrate the national weather radar within days of its failure mid-hurricane season.
Use research institutions to supply capability that national services cannot sustain alone.
In Madagascar, the Bureau National de Gestion des Risques et des Catastrophes (BNGRC, the national disaster management office) partners with the Pacific Disaster Center (an applied research centre managed by the University of Hawaiʻi), to use its multi-hazard DisasterAWARE Pro platform. The platform complements the nation’s warning and alerting system, equipping the BNGRC with hazard monitoring, alert dissemination and geospatial mapping capabilities. The platform tracks up to 30 hazards and provides continuous remote analyst support. This arrangement enabled rapid monitoring and support when the country confirmed its first cases of mpox in December 2025. Because of the partnership, BNGRC did not need to build capabilities from scratch or wait for external technical support. DisasterAWARE Pro was operational, operators were trained and the platform was actively used throughout the response for hazard monitoring, geospatial tracking of outbreak progression and alert dissemination. Other examples of such partnerships include the Malawian meteorological service piloting AI-based forecasting with the Norwegian Meteorological Institute and the European Centre for Medium-Range Weather Forecasts; Algeria working with the Center for Development of Advanced Technologies, the Ministry of Higher Education and Scientific Research, and the Algerian Space Agency on a wildfire detection system; and the African Disaster Mitigation Research Center leading a post-event assessment and seismic modelling following the Hayli Gubbi eruption.
Partnerships with civil society can help to reach populations state agencies cannot. In Nigeria, Plan International, International Medical Corps and the Dutch Relief Alliance used community networks, ward leaders and camp management structures to disseminate information and target aid in Bakassi and Dalori camps for internally displaced persons. In the Brazilian Amazon, the alert system in Indigenous territories originated in a request from Indigenous leaders and is run by the Hutukara Yanomami Association with the Instituto Socioambiental and UNICEF. Other examples include IOM working with communities and emergency operations centres to reach geographically isolated populations in Fiji; and Save the Children establishing school-based clubs to reach children and their families in Bangladesh.