Course Introduction
Development programmes increasingly operate across large geographic areas where conventional field monitoring alone may be insufficient to establish the extent, distribution, timing, and persistence of observed changes. Satellite imagery and geospatial analysis can help teams examine agricultural conditions, forest cover, water availability, settlement expansion, ecosystem restoration, infrastructure development, and disaster exposure. This course equips participants to use Google Earth Engine to transform extensive geospatial datasets into structured, repeatable evidence that supports programme design, implementation monitoring, evaluation, adaptive management, and accountability.
The course develops advanced practical capabilities in Earth Engine's JavaScript Code Editor and, where appropriate, its Python interface. Participants learn to work with raster imagery, image collections, vector features, geographic boundaries, temporal filters, reducers, and statistical summaries. They practise constructing analytical workflows that combine multiple data sources and produce interpretable indicators for development programmes. Emphasis is placed on understanding how data structures, spatial resolution, acquisition dates, missing observations, and processing choices influence the reliability and interpretation of results.
A central component of the programme is the design of satellite-derived indicators that correspond to explicit programme theories, monitoring questions, and evaluation criteria. Participants explore vegetation indices, land-cover classification, surface water indicators, burn and disturbance measures, built-up area proxies, and other relevant measures. They learn to distinguish direct observations from proxies, select appropriate reference periods, account for seasonal patterns, and document indicator definitions. These skills help ensure that geospatial results are not merely technically sophisticated but are also relevant to programme objectives, beneficiary needs, and management decisions.
Advanced time-series analysis and change detection are addressed through practical exercises involving image compositing, cloud and shadow masking, trend summaries, before-and-after comparisons, anomaly detection, and spatially explicit change assessment. Participants examine how changes in image availability, sensor characteristics, observation frequency, and environmental variability can affect comparisons over time. They learn to test assumptions, assess uncertainty, compare alternative analytical choices, and validate outputs using reference observations or independent datasets where available. The course emphasises that observed spatial change does not automatically demonstrate programme attribution; credible causal conclusions require suitable evaluation designs and corroborating evidence.
The training also addresses the operational requirements of producing reliable, reusable analytical outputs. Participants develop modular scripts, apply efficient filtering and reduction techniques, manage exports and assets, create charts and map layers, and establish quality assurance checks. They explore approaches for documenting datasets, assumptions, parameters, validation results, limitations, and version changes so that colleagues can reproduce and review the analysis. Attention is given to computation efficiency, task management, project permissions, data governance, platform access arrangements, and the distinction between research prototypes and continuously operated production systems.
Throughout the ten-day programme, participants work through applied development-assessment scenarios and build a practical geospatial assessment product from research question to decision-ready output. Activities can be tailored to agriculture, natural resource management, climate adaptation, disaster risk reduction, infrastructure, public health, or urban development. By the end of the course, participants should be better prepared to integrate Earth Engine outputs with programme monitoring systems, field verification, administrative records, survey evidence, and evaluation frameworks. The intended outcome is a transparent and technically defensible workflow that improves the spatial reach, consistency, timeliness, and usefulness of development programme assessment.
Duration
10 days
Who Should Attend
-
Monitoring, Evaluation, Accountability and Learning (MEAL) specialists seeking to integrate satellite-derived evidence into programme monitoring and evaluation frameworks.
-
Development programme managers who need spatially explicit evidence to guide implementation decisions, resource allocation, and adaptive management.
-
GIS analysts and remote-sensing specialists responsible for producing geographic indicators and assessing environmental or socioeconomic change.
-
Monitoring and evaluation officers working on agriculture, food security, rural livelihoods, natural resource management, and climate resilience programmes.
-
Research officers and impact evaluation specialists interested in combining Earth observation with quantitative and mixed-methods evaluation designs.
-
Environmental officers assessing forest conservation, biodiversity protection, watershed management, ecosystem restoration, and land degradation.
-
Climate change adaptation and disaster risk reduction practitioners measuring exposure, environmental hazards, recovery, and resilience-related outcomes.
-
Infrastructure and urban development specialists examining settlement growth, transport corridors, construction footprints, and land-use transformation.
-
Data scientists and statisticians seeking to extend their analytical workflows with cloud-based geospatial processing and spatial-temporal data.
-
Programme data managers responsible for integrating geospatial indicators into institutional information systems and results dashboards.
-
NGO, government, multilateral agency, and donor personnel commissioning or reviewing geospatial evidence for programme performance reporting.
-
University lecturers, postgraduate researchers, and technical consultants undertaking applied Earth observation research or development evaluation.
-
Field monitoring coordinators who need to design targeted verification visits using spatial patterns, anomaly maps, and change-detection outputs.
-
Digital transformation and information management teams supporting reproducible analytical workflows and geospatial evidence services.
Course Objectives
-
Explain the architecture of Google Earth Engine and select appropriate imagery, vector datasets, processing methods, and analytical structures for specific development programme assessment questions.
-
Formulate measurable geospatial indicators from programme theories of change, results frameworks, evaluation questions, and clearly defined spatial and temporal units of analysis.
-
Locate, assess, and combine relevant satellite image collections and supporting datasets while documenting coverage, resolution, acquisition dates, provenance, and limitations.
-
Build efficient image-processing workflows using spatial and temporal filtering, cloud and shadow masking, band selection, compositing, and reusable functions.
-
Calculate and interpret vegetation, water, burn, built-up area, and other relevant spectral indicators while recognising the limits of each indicator as a proxy for programme outcomes.
-
Apply time-series analysis and change-detection techniques to assess spatial patterns, temporal trajectories, anomalies, and differences between selected reporting periods.
-
Develop land-cover classification workflows, select suitable training and validation samples, and evaluate classification performance using appropriate accuracy measures.
-
Use zonal statistics and spatial aggregation to summarise indicators by programme site, administrative unit, ecological zone, infrastructure corridor, or beneficiary catchment.
-
Assess seasonal variability, data gaps, cloud contamination, spatial resolution, measurement uncertainty, and other threats to the validity of geospatial comparisons.
-
Integrate Earth Engine outputs with field observations, administrative records, survey datasets, and evaluation designs without overstating causal attribution from spatial association alone.
-
Improve script reproducibility, computational efficiency, asset organisation, export management, validation documentation, and quality assurance for repeatable institutional workflows.
-
Produce clear maps, charts, indicator tables, and technical narratives that communicate geospatial findings, uncertainty, limitations, and actionable implications to programme decision-makers.
Comprehensive Course Outline
Module 1: Advanced Google Earth Engine Architecture and Development Assessment Applications
-
Examine Earth Engine's cloud-based processing architecture, supported geospatial data structures, analytical capabilities, and implications for large-scale development programme assessment.
-
Distinguish images, image collections, geometries, features, feature collections, reducers, and server-side objects through applied examples relevant to monitoring and evaluation.
-
Configure the Code Editor, project settings, code repositories or version-control arrangements, map visualisation, and debugging practices for a structured analytical environment.
-
Translate a development programme question into a spatially explicit assessment plan specifying indicators, geography, reporting periods, expected outputs, and validation requirements.
Module 2: Geospatial Data Discovery, Selection and Provenance
-
Navigate the Earth Engine Data Catalog to identify suitable optical, radar, elevation, climate, land-cover, surface-water, and other supporting datasets.
-
Evaluate spatial resolution, temporal frequency, sensor characteristics, processing level, acquisition dates, geographic coverage, and known limitations before selecting data.
-
Import and inspect programme boundaries, administrative units, monitoring sites, infrastructure alignments, and other vector assets needed for spatial analysis.
-
Create a data inventory documenting source identifiers, licensing or access conditions, collection versions, processing assumptions, and intended programme assessment uses.
Module 3: Advanced JavaScript and Python Workflow Development
-
Write reusable functions, loops, conditional logic, dictionaries, and modular processing routines that improve the clarity and maintainability of Earth Engine scripts.
-
Apply server-side programming concepts, deferred computation, mapped functions, and client-server distinctions to prevent common performance and execution errors.
-
Explore appropriate Python API workflows for integration with notebooks, statistical packages, and wider data science processes where the project environment supports them.
-
Establish naming conventions, comments, parameter configuration, code review procedures, and version tracking for collaborative geospatial assessment projects.
Module 4: Image Collection Processing and Quality Control
-
Filter image collections by date, geographic extent, sensor metadata, cloud conditions, and other properties relevant to the assessment question.
-
Apply cloud and shadow masking, band harmonisation where appropriate, scaling, quality flags, and observation screening before deriving indicators.
-
Generate median, percentile, seasonal, and other composite images while evaluating how compositing decisions affect spatial coverage and measured change.
-
Diagnose missing observations, edge effects, anomalous pixels, and inconsistent data availability using visual inspection, summary statistics, and quality-control layers.
Module 5: Spectral Indices and Programme-Relevant Indicator Engineering
-
Calculate and interpret vegetation indicators such as NDVI and EVI for carefully specified agricultural, restoration, and vegetation-condition assessment questions.
-
Derive surface-water and moisture-related indicators using suitable bands and datasets, documenting limitations caused by turbidity, terrain, mixed pixels, or seasonal conditions.
-
Explore burn, disturbance, built-up area, and land-surface indicators appropriate to environmental management, disaster recovery, and infrastructure monitoring.
-
Develop indicator specifications defining formulas, spatial units, temporal windows, aggregation rules, interpretation thresholds, and cautions against unsupported conclusions.
Module 6: Time-Series Analysis and Temporal Change Detection
-
Construct annual, seasonal, monthly, or event-based image series using consistent geographic boundaries, observation criteria, and processing decisions.
-
Use temporal reducers, charting functions, and summary statistics to examine trends, variability, anomalies, and differences between reporting periods.
-
Implement before-and-after comparisons and pixel-level difference methods while investigating seasonal effects, sensor inconsistencies, and changes in observation quality.
-
Assess breakpoints, persistent changes, and unusual trajectories using appropriate statistical methods, clearly distinguishing exploratory signals from validated findings.
Module 7: Land-Cover Classification and Accuracy Assessment
-
Design supervised classification workflows using appropriate spectral features, temporal composites, terrain variables, and other defensible predictor datasets.
-
Develop and inspect training samples that represent relevant land-cover classes, geographic variability, and the conditions encountered across the study area.
-
Evaluate classification outputs using independent or appropriately separated validation samples, confusion matrices, class-specific accuracy, and overall accuracy measures.
-
Quantify area estimates and classification uncertainty while explaining class imbalance, sampling limitations, and potential errors in land-cover transition analysis.
Module 8: Spatial Aggregation, Zonal Statistics and Geographic Comparisons
-
Calculate area-weighted or otherwise appropriate zonal summaries for programme sites, administrative areas, watersheds, ecological units, and beneficiary catchments.
-
Compare indicator distributions across implementation zones, intervention intensity groups, geographic strata, and relevant contextual categories.
-
Design aggregation methods that account for differences in polygon size, raster resolution, missing observations, and unequal data availability.
-
Produce consistent spatial summary tables and geographic comparison layers suitable for integration with results frameworks and programme performance reports.
Module 9: Agriculture, Food Security and Rural Livelihoods Assessment
-
Examine vegetation dynamics and seasonal crop-condition proxies to support agricultural monitoring while accounting for crop calendars, rainfall variation, and mixed land uses.
-
Assess spatial patterns of agricultural expansion, land-cover transitions, irrigation presence, and possible environmental stress using suitable datasets and validation procedures.
-
Integrate Earth Engine indicators with crop surveys, field observations, production statistics, and food security information to strengthen interpretation of programme results.
-
Develop maps and summary indicators for agricultural intervention targeting, landscape-level monitoring, and follow-up field verification without treating vegetation indices as direct yield measurements.
Module 10: Climate Resilience, Ecosystems and Natural Resource Management
-
Assess changes in vegetation condition, forest cover, surface water, land degradation proxies, and restoration areas using consistent spatial and temporal methods.
-
Examine flood, drought, wildfire, and other environmental exposure patterns using suitable satellite products and clearly defined event periods.
-
Monitor potential climate adaptation and ecosystem management outcomes by comparing observed environmental patterns with programme records and contextual climate information.
-
Document the limits of remote sensing for measuring resilience, biodiversity, ecosystem services, and household vulnerability, identifying where field or specialist evidence is necessary.
Module 11: Infrastructure, Urban Development and Service Coverage
-
Map built-up area patterns, settlement expansion, transport corridors, and surrounding land-cover changes to inform spatial monitoring of infrastructure interventions.
-
Examine geographic proximity and potential service catchments using vector data, buffers, and spatial summaries, while distinguishing proximity from actual service access or utilisation.
-
Assess construction footprints and environmental change over time using suitable imagery, reference boundaries, and independent verification where possible.
-
Combine geospatial outputs with project implementation schedules, asset inventories, service statistics, and site inspection records to assess progress and identify follow-up needs.
Module 12: Programme Evaluation Design, Attribution and Evidence Triangulation
-
Align spatial indicators with a programme's theory of change, evaluation questions, implementation geography, exposure definitions, and intended outcome pathways.
-
Design geographic comparison strategies and examine baseline and follow-up conditions while considering confounding factors, spillovers, selection effects, and temporal trends.
-
Explore how Earth Engine-derived variables can contribute to quasi-experimental designs, panel analysis, matching, or difference-in-differences when their assumptions are justified.
-
Triangulate satellite evidence with field verification, surveys, administrative data, qualitative findings, and implementation records before making claims about programme contribution or impact.
Module 13: Statistical Validation, Uncertainty and Sensitivity Analysis
-
Identify major sources of uncertainty arising from sensor characteristics, classification errors, cloud contamination, spatial resolution, sample design, and indicator construction.
-
Conduct sensitivity checks using alternative compositing methods, temporal windows, thresholds, spatial units, and inclusion criteria to assess result stability.
-
Evaluate reference data quality, sampling independence, validation representativeness, and the implications of measurement error for programme-level conclusions.
-
Present uncertainty ranges, confidence limitations, validation findings, and evidence gaps in a transparent manner that supports proportionate management decisions.
Module 14: Computational Efficiency, Exports and Reproducibility
-
Optimise spatial and temporal filters, mapped operations, reducers, scale parameters, geometry complexity, and intermediate computations for more efficient processing.
-
Configure image, table, and other supported exports with appropriate spatial resolution, projection, region, file format, naming conventions, and task monitoring.
-
Organise reusable scripts, assets, parameters, metadata, and processing logs so analytical outputs can be reviewed, refreshed, and reproduced by authorised colleagues.
-
Plan workflows in accordance with current platform quotas, applicable access terms, project permissions, and the operational or non-commercial status of the intended use.
Module 15: Geospatial Visualisation, Reporting and Decision Support
-
Create meaningful map layers, legends, colour palettes, charts, and summary tables that communicate patterns without exaggerating differences or hiding uncertainty.
-
Design reporting outputs that connect geographic evidence to programme indicators, evaluation questions, implementation milestones, and management actions.
-
Develop interactive map-based exploration or Earth Engine applications where appropriate, with clear controls, documented indicators, and audience-focused interpretation.
-
Prepare concise technical narratives that explain the data sources, analytical steps, findings, limitations, validation status, and implications for programme decisions.
Module 16: Integrated Development Programme Assessment Capstone
-
Define an applied assessment question and develop a complete workflow covering data discovery, indicator specification, image processing, analysis, and quality assurance.
-
Produce spatial and temporal outputs for a selected programme theme, incorporating appropriate comparison areas, reference observations, and validation checks.
-
Prepare a reproducible analytical package containing documented code, data provenance, indicator definitions, maps, tables, uncertainty notes, and key methodological decisions.
-
Present a decision-oriented assessment report and defend the interpretation of results, including limitations, follow-up verification needs, and recommendations for programme management.
Training Approach
This course will be delivered by our skilled trainers who have vast knowledge and experience as expert professionals in the fields. The course is taught in English and through a mix of theory, practical activities, group discussion and case studies. Course manuals and additional training materials will be provided to the participants upon completion of the training.
Tailor-Made Course
This course can also be tailor-made to meet organization requirement. For further inquiries, please contact us on: Email: training@upskilldevelopment.com Tel: +254 721 331 808
Training Venue
The training will be held at our Upskill Training Centre. We also offer training for a group (at a discount of 10% to 50%) at requested location all over the world. The Onsite course fee covers the course tuition, training materials, two break refreshments, buffet lunch, airport transfers, Upskill gift package, and guided tour.
Visa application, travel expenses, dinners, accommodation, insurance, and other personal expenses are catered by the participant
Certification
Participants will be issued with Upskill certificate upon completion of this course.
Airport Pickup and Accommodation
Airport pickup and accommodation is arranged upon request. For booking contact our Training Coordinator through Email: training@upskilldevelopment.com, +254 721 331 808
Terms of Payment:
Unless otherwise agreed between the two parties’ payment of the course fee should be done 3 working days before commencement of the training so as to enable us to prepare better.








