使用geeLite R包从Google Earth引擎构建和管理本地数据库

11115 Building and Managing Local Databasesfrom Google Earth Engine withthe geeLite R PackagePublic Disclosure Authorized Marcell T. KurbuczBo Pieter Johannes Andrée Development EconomicsDevelopment Research GroupMay 2025 Policy Research Working Paper11115 Abstract Google Earth Engine has transformed geospatial analysisby providing access to petabytes of satellite imagery andgeospatial data, coupled with the substantial computationalpower required for in-depth analysis. This accessibilityempowers scientists, researchers, and non-experts aliketo address critical global challenges on an unprecedentedscale. In recent years, numerous R packages have emergedto leverage Google Earth Engine’s functionalities. How-ever, constructing and managing complex spatio-temporaldatabases for monitoring changes in remotely sensed dataremains a challenging task that often necessitates advancedcoding skills. To bridge this gap, geeLite, a novel R package, isintroduced to facilitate the construction,manage-ment, and updating of local databases for Google EarthEngine-computed geospatial features, which enables usersto monitor their evolution over time. By storing geospatialfeatures in SQLite format—a serverless and self-containeddatabase solution requiring no additional setup or admin-istration—geeLite simplifies the data collection process.Furthermore, it streamlines the conversion of stored datainto native R formats and provides functions for aggregat-ing and processing created databases to meet specific userneeds. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about developmentissues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry thenames of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely thoseof the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank andits affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Graphical Abstract Building and Managing Local Databases from Google Earth Enginewith thegeeLiteR Package Marcell T. Kurbucz, Bo Pieter Johannes Andrée Highlights Building and Managing Local Databases from Google Earth Enginewith thegeeLiteR Package Marcell T. Kurbucz, Bo Pieter Johannes Andrée •Google Earth Engine offers vast satellite imagery and computationalpower.•geeLiteis an R package designed to leverage the power of GoogleEarth Engine.•It enables creating, updating, and managing custom databases for real-time tracking.•It stores data in SQLite format, enhancing both accessibility and porta-bility.•It streamlines the reading, aggregation, and processing of created databases. Building and Managing Local Databases from GoogleEarth Engine with thegeeLiteR Package Marcell T. Kurbucza,b,1,∗, Bo Pieter Johannes Andréeb,1 aInstitute for Global Prosperity, The Bartlett, University College London, 149 TottenhamCourt Road, London, W1T 7NF, United KingdombDevelopment Economics Data Group, World Bank, 1818 H Street NW, Washington,D.C., 20433, USA Keywords:GoogleEarthEngine,GeographicInformationSystem,RemoteSensing,RasterData,Spatio-TemporalData,Software JELCodes:C21,C63,C81,C88,L17 1. Introduction The ever-growing volume of Earth observation data presents both oppor-tunities and challenges for scientific inquiry.While vast datasets hold thepotential to revolutionize our understanding of Earth systems, traditionaldesktop-based analysis methods often struggle with the computational bur-den associated with such data (Amani et al., 2020).Google Earth Engine(GEE) has emerged as a powerful solution, offering a cloud-based platformfor efficient management, analysis, and visualization of geospatial big data(Gorelick et al., 2017).Its core strength lies in its ability to overcome thelimitations of traditional approaches by providing (Tamiminia et al., 2020): •Petabytes of public geospatial data:A comprehensive data cata-log readily accessible through a web interface, including historical andcurrent satellite imagery, environmental variables, and other geospatialinformation.•High-performance parallel processing:Leveraging Google’s cloudinfrastructure for large-scale analysis, enabling researchers to tacklecomplex problems that would be infeasible on personal computers.•Accessible development environment:A web-based interface andapplication programming interfaces (APIs) in JavaScript and Python,supporting widely-used programming languages. This combination of features allows scientists to investigate new scientificquestions in a wide range of subjects, particularly those requiring large-scalespatial and temporal analysis.2Examples include studies on climate change(Banerjee et al., 2024; Kazemi Garajeh et al., 2024), environmental degrada-t