Page 11 - Disaster Management: The Standards Perspective

Disaster Management: The Standards Perspective

Background

More than 125 million people were affected by disasters each year globally between
2015-2023. The rising frequency of disasters over the past five decades has significantly
1
heightened concerns about potential further loss of life and property. The adoption of Artificial
Intelligence (AI) in managing these natural hazards has seen considerable growth over the
years, propelled by enhanced computing capabilities, data access, and refined algorithms.
Initial efforts concentrated on crafting basic models for predicting meteorological conditions
and natural events such as earthquakes and floods. The introduction of big data, machine
learning, and deep learning technologies has broadened the scope of application, allowing
for more holistic approaches to disaster management in line with the targets of the Sendai
Framework for Disaster Risk Reduction.
AI is applied across a spectrum of activities, ranging from forecasting disasters and evaluating
risks to conducting real-time surveillance, and aiding in recovery after a disaster. AI-based
modelling techniques can also be employed to process extensive datasets on weather and
geological conditions to forecast potential natural hazards like floods, earthquakes tsunamis
and landslides, thereby facilitating timely evacuations. Additionally, AI-enhanced imagery
and video analysis from satellites and drones help in the immediate monitoring of disaster-
stricken zones, which aids in swift damage evaluation as well as in resource prioritization and
distribution. Incorporating AI into disaster management comes with its own set of challenges.
Concerns about data privacy, biases in algorithms, and the necessity for comprehensive, high-
quality data are prominent. Furthermore, the effective utilization of AI solutions demands
specialized knowledge, which may pose barriers.

International standards can facilitate the smooth integration of various AI technologies,
improving the efficiency of disaster response and mitigation activities. They help ensure data
privacy and safeguard AI systems against cyber threats, preserving their dependability in
emergency situations and transparency in decision-making. By setting best practices, global
standards direct the creation and application of AI tools, making sure they are strong, equitable,
and effective in handling natural hazards. Standards can also support the advancement in AI
technologies, while serving as a link between technological innovation and humanitarian efforts,
presenting new opportunities to lessen the harsh effects of natural hazards and disasters.

The cooperative efforts of the World Meteorological Organization (WMO), the International
Telecommunication Union (ITU), and the United Nations Environment Programme (UNEP)
can play a key role in formulating standards for managing natural hazards, thereby boosting
worldwide resilience and response abilities. Each of these United Nations agencies contributes
critical knowledge and expertise—WMO with its meteorological insights, climatological and
hydrological, UNEP with its understanding of environmental risks and ITU serves as both a
standards developing organization (SDO) and the UN agency for digital technologies—all
of which are vital for the development of a comprehensive and interoperable frameworks,
guidelines and best practices for disaster management.

In line with this vision, the ITU/WMO/UNEP Focus Group on AI for Natural Disaster Management
(FG-AI4NDM) was established in December 2020.

1 https:// www .undrr .org/ implementing -sendai -framework/ monitoring -sendai -framework

1