绘制地球观测的未来：气候智能的技术创新

Charting the Futureof Earth Observation: W H I T EP A P E R Images:Getty Images Contents Foreword Executive summary Introduction 1Bytes to insights 1.1Enhanced resolution and diversification in sensing capabilities1.2Reduced EO data processing time to enable near-real-time 2New era in climate forecasting for adaptation and resilience 3Democratizing climate insights ConclusionContributorsEndnotes Disclaimer This document is published by theWorld Economic Forum as a contributionto a project, insight area or interaction.The findings, interpretations andconclusions expressed herein are a resultof a collaborative process facilitated and © 2024 World Economic Forum. All rightsreserved. No part of this publication maybe reproduced or transmitted in any formor by any means, including photocopying Foreword Sebastian BuckupHead, Network andPartnerships; Member, Dava NewmanDirector, MIT Media Lab;Apollo Program Professorof Astronautics, MIT With escalating temperatures, increasingly severeweather events and unprecedented levels ofgreenhouse gas emissions, the world stands at acrossroads. Scientific consensus underscores thatimmediate measures are essential to mitigate themost catastrophic impacts of climate change. In thisnew paradigm, Earth observation (EO) technologyand innovation are championing a new era forclimate intelligence, offering unprecedented insightsand solutions to address these urgent challenges.Recent EO innovations, combined with the growth observable measurements from satellites withthe highest revisit times (satellites that visit certaingeographic regions more frequently). In turn, this This white paper, written in collaboration with theMassachusetts Institute of Technology (MIT) MediaLab, highlights the transformative potential of EO forclimate intelligence and forecasting. By combiningthe research capabilities of the MIT Media Lab withthe global platform of the World Economic Forum,the paper identifies technology pipelines acceleratingthe processing and analysis of satellite EO data toprovide unparalleled insights into climate change. Executive summary Earth observation technologies and advanceddata processing are revolutionizing climate Systemic challenges have historically preventedEO data from being fully integrated into climatesolutions, primarily due to its large volume andcomplexity. Rapid technological advancements insatellite and sensor technologies are addressingthese systematic issues alongside new and openartificial intelligence (AI) algorithms, machine learning nations and small- and medium-sized enterprises(SMEs) to launch their own satellites, increasing thevolume of publicly available EO data. At the sametime, there is an increase in the development oflarger, more sophisticated satellite platforms. These Higher resolution climate forecasting: ClimateML-based models and foundation models areincreasing the resolution of climate and weatherforecast models twelvefold. These enhanced Recent advancements in satellite EO sensors driveimproved global coverage, resolution, accuracyand a wider array of observable measurements.This enables monitoring of larger swaths of land andmore frequent revisit rates. These advancements Contextual data for end-user needs: Dataimmersion through augmented reality (AR) andvirtual reality (VR) are transforming complex EOdatasets into interactive models that help usersunderstand the data and intuitive visual insightsthat improve decision-making. Digital twins useadvanced analytics, ML and AI to analyse data Unprecedented data processing speeds:from multiple sources and simulate complex “what Sophisticated AI and ML algorithms enable moredetailed climate impact assessments (such asthose used in post-disaster management) in hoursor minutes. This task can take weeks when usingtraditional models or on-site inspections. ClimateML-based models trained on existing data can Key next steps include: –Expanding EO data access for climate-vulnerable communities–Investing in technology pipelines to drive furtherinnovation in EO-derived climate insights Evolution of large and small EO satellites: EOsatellite systems have advanced on two oppositefronts. The rise of small satellites and miniaturization Introduction Integrating complementary technologieswith satellite EO converts complex data into By 2032, satellite EO is expected to generate up to2 exabytes (2 billion gigabytes) of data cumulatively,accounting for approximately 86% of the total dataproduced by the space application segment forthe forecast period.3However, the full potentialof satellite EO data in managing climate impactsremains underutilized. This is partially due to theinherent complexity of large satellite EO datasets Since 1980, the US has experienced 391 weatherdisasters causing damages of over $2.755 trillion,1including severe storms, hurricanes, floods andwildfires. The World Meteorological Organization By 2032, satelliteEO is expected togenerate over 2 Earth obser