Design and Evaluation of a Google Earth Engine-Based Training Framework for Water Resource Analysis in Educational Settings

Orazbek Zhanatayev, Shakhnazar-Sultan Manbay, Baglan Tolebay, Azamat Serek

2026

Closed access Metadata (CSL-JSON)
DOI 10.1109/icecco67619.2026.11488826
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This study will outline the establishment and evaluation of a training module based on Google Earth Engine (GEE) as a simple learning data analytics tool. The suggested framework will be based on a three-part pedagogical model that involves (i) conceptual backgrounds of remote sensing and hydrological processes, (ii) analytical experiences through Google Earth Engine capabilities of image collections, reducers, time-series analysis, spectral indices (NDWI, MNDWI), and (iii) application-based projects where the learners will be expected to study realworld water systems independently. A pilot group comprising 22 geography, computer science and environmental science students were provided with a threesession program, which included the basics of remote sensing, GEE based water resource analysis and the utility of the Normalized Difference Water Index (NDWI) to detect surface water. All the course participants completed the course and carried out their own research projects with openaccess datasets in GEE. This indicated the ease and flexibility of the framework. The test result increased significantly,$58.6 \%(S D=12.4)$to$86.3 \%(S D=8.9)(p<0.01)$. The accuracy of water masks constructed by participants, when compared to the reference JRC Global Surface Water dataset, was 91.7 %, with a Kappa coefficient of 0.88. The time to complete a task reduced by 27 percent and the usefulness, clarity, and overall experience average ratings stood at 4.7, 4.6 and 4.8 respectively. While most participants improved their coding and data processing skills, 14 % initially encountered difficulties with JavaScript syntax. This highlights the need for introductory programming support in future iterations of the framework. The framework is effective in enhancing the analytical skills and understanding hydrological systems. It provides a resource-saving model of teaching environmental data that is easy to expand especially in schools which do not possess many computers. Programming assistance should be provided to beginners that is specific in future versions.