+254 721 331 808    training@upskilldevelopment.com

Home / Monitoring and Evaluation

Foundation Models for Earth Observation and Environmental Evaluation Training Course

Build practical expertise in Earth observation foundation models for environmental evaluation. Learn model selection, geospatial data preparation, fine-tuning, environmental applications, validation, uncertainty analysis, and GIS integration to produce more credible, reproducible, and decision-ready environmental evidence.
Duration: 10 Days Classroom & Online In-House Available Certificate on Completion

Training Registration & Schedule

Compare training options, fees and available dates, then register for your preferred session.

Online Training Registration

Training ModePlatformDurationFee per ParticipantRegistration
LIVE ONLINEInstructor-led training Zoom / Google MeetAttend from anywhere 10 days USD 1,740 Register Online →

Classroom / On-site Training Schedule

15 upcoming sessions

Swipe horizontally to see all columns →

Training DatesLocationTraining ModeFee per ParticipantRegistration
07 Dec 2026 – 18 Dec 202610 training days Nairobi CLASSROOM USD 2,900 Register Now →
07 Dec 2026 – 18 Dec 202610 training days Mombasa CLASSROOM USD 3,400 Register Now →
04 Jan 2027 – 15 Jan 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
04 Jan 2027 – 15 Jan 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
01 Feb 2027 – 12 Feb 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
01 Feb 2027 – 12 Feb 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
01 Mar 2027 – 12 Mar 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
01 Mar 2027 – 12 Mar 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
05 Apr 2027 – 16 Apr 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
05 Apr 2027 – 16 Apr 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
03 May 2027 – 14 May 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
03 May 2027 – 14 May 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
07 Jun 2027 – 11 Jun 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
07 Jun 2027 – 18 Jun 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
05 Jul 2027 – 16 Jul 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
Need registration assistance? Contact training@upskilldevelopment.com for training dates, invoices and invitation letters.

Nairobi, Kenya Classroom/On-site Training Schedule

Course Date Location Fee Enroll
07/12/2026 to 18/12/2026 Nairobi 2,900 USD Register
04/01/2027 to 15/01/2027 Nairobi 2,900 USD Register
01/02/2027 to 12/02/2027 Nairobi 2,900 USD Register
01/03/2027 to 12/03/2027 Nairobi 2,900 USD Register
05/04/2027 to 16/04/2027 Nairobi 2,900 USD Register
03/05/2027 to 14/05/2027 Nairobi 2,900 USD Register
07/06/2027 to 11/06/2027 Nairobi 2,900 USD Register
05/07/2027 to 16/07/2027 Nairobi 2,900 USD Register
02/08/2027 to 13/08/2027 Nairobi 2,900 USD Register
06/09/2027 to 17/09/2027 Nairobi 2,900 USD Register
04/10/2027 to 15/10/2027 Nairobi 2,900 USD Register
01/11/2027 to 12/11/2027 Nairobi 2,900 USD Register
06/12/2027 to 17/12/2027 Nairobi 2,900 USD Register

Mombasa, Kenya Classroom/On-site Training Schedule

Course Date Location Fee Enroll
07/12/2026 to 18/12/2026 Mombasa 3,400 USD Register
04/01/2027 to 15/01/2027 Mombasa 3,400 USD Register
01/02/2027 to 12/02/2027 Mombasa 3,400 USD Register
01/03/2027 to 12/03/2027 Mombasa 3,400 USD Register
05/04/2027 to 16/04/2027 Mombasa 3,400 USD Register
03/05/2027 to 14/05/2027 Mombasa 3,400 USD Register
07/06/2027 to 18/06/2027 Mombasa 3,400 USD Register
05/07/2027 to 16/07/2027 Mombasa 3,400 USD Register
02/08/2027 to 13/08/2027 Mombasa 3,400 USD Register
06/09/2027 to 17/09/2027 Mombasa 3,400 USD Register
04/10/2027 to 15/10/2027 Mombasa 3,400 USD Register
01/11/2027 to 12/11/2027 Mombasa 3,400 USD Register
06/12/2027 to 17/12/2027 Mombasa 3,400 USD Register

Kigali, Rwanda Classroom/On-site Training Schedule

Course Date Location Fee Enroll

Dubai, UAE Classroom/On-site Training Schedule

Course Date Location Fee Enroll

Dar es Salaam, Tanzania Classroom/On-site Training Schedule

Course Date Location Fee Enroll

Kampala, Uganda Classroom/On-site Training Schedule

Course Date Location Fee Enroll

Course Introduction

Foundation models for Earth observation are changing how environmental researchers and development institutions extract information from satellite imagery, radar observations, land-cover products, and other geospatial datasets. Pre-trained on extensive and diverse data, these models can learn reusable representations of Earth's surface and support downstream tasks with less task-specific labelled data than some conventional approaches. This course introduces participants to the concepts, architectures, applications, and evaluation methods required to apply geospatial foundation models to environmental monitoring, programme evaluation, climate resilience, natural resource management, and evidence-based development planning.

The programme examines the Earth observation data ecosystem and the technical foundations of geospatial artificial intelligence. Participants explore multispectral satellite imagery, synthetic aperture radar, time-series observations, digital elevation models, vegetation indices, land-cover classifications, and complementary environmental measurements. They learn how foundation models differ from conventional supervised machine learning, how self-supervised and multimodal pre-training create transferable representations, and why sensor characteristics, spatial resolution, temporal coverage, and geographic context influence model suitability. These foundations support informed selection of models for real-world environmental questions.

A major focus is the practical application of foundation models to environmental evaluation. Participants investigate how model-derived representations can support land-cover mapping, forest disturbance detection, agricultural monitoring, wetland assessment, urban expansion analysis, flood-impact assessment, soil-related applications, and ecosystem change monitoring. The course examines how to adapt pre-trained models through feature extraction, lightweight classification, fine-tuning, parameter-efficient adaptation, and suitable downstream learning methods. Participants also learn to connect geospatial predictions to evaluation questions, indicators, programme theories, and decision requirements instead of treating technical model performance as a substitute for meaningful environmental evidence.

The course emphasises rigorous validation and scientific credibility. Participants learn to design spatially and temporally appropriate training, validation, and test datasets; prevent geographic data leakage; compare foundation models against established remote sensing methods; and evaluate performance using metrics suited to classification, segmentation, regression, and change detection. They examine how results can deteriorate when models are transferred between ecological zones, seasons, satellite sensors, or observation conditions. Practical exercises address class imbalance, reference-data uncertainty, spatial autocorrelation, cloud contamination, model calibration, interpretability, and uncertainty reporting, helping participants determine when a model's outputs are reliable enough to support environmental decisions.

Participants also explore operational considerations that influence the value of geospatial foundation models in development institutions. These include access to imagery, computational resources, cloud processing, data storage, model licensing, reproducible workflows, technical skills, and integration with geographic information systems. The programme addresses responsible AI use, protection of sensitive geospatial information, representation of vulnerable communities and ecosystems, model transparency, and the energy and infrastructure costs associated with training and inference. Participants learn to assess whether a foundation model offers practical benefits over simpler alternatives and to document the conditions under which findings can reasonably be generalised.

By the end of the ten-day course, participants will be prepared to design an applied Earth observation workflow that links a clearly defined environmental evaluation question to suitable data, a selected foundation model, downstream analysis, independent validation, and decision-oriented reporting. They will develop practical outputs such as a model selection matrix, a geospatial data preparation protocol, an evaluation design, a validation report, and an implementation roadmap. The course supports environmental scientists, geospatial analysts, monitoring and evaluation professionals, and development practitioners who need to turn rapidly evolving geospatial AI capabilities into credible, reproducible, and useful environmental evidence.

Duration

10 days

Who Should Attend

  • Environmental scientists and researchers applying Earth observation data to ecosystem assessment, environmental change analysis, and sustainability research.

  • GIS specialists and remote sensing analysts seeking to use pre-trained geospatial models for more efficient spatial analysis.

  • Monitoring, evaluation, accountability, and learning professionals assessing environmental programmes, climate initiatives, and natural resource management projects.

  • Climate change adaptation and mitigation specialists monitoring land conditions, climate-related hazards, and ecosystem responses.

  • Forestry and biodiversity professionals analysing forest cover, vegetation disturbance, habitat conditions, and landscape change.

  • Agricultural development specialists evaluating crop conditions, land productivity, irrigation systems, and agricultural resilience.

  • Water resources and wetland specialists assessing water extent, catchment conditions, aquatic ecosystems, and changes in hydrological patterns.

  • Data scientists and machine learning engineers developing downstream applications from Earth observation foundation models.

  • Development programme managers responsible for environmental safeguards, climate resilience, land management, and geographically distributed projects.

  • Research evaluators designing environmental indicators, baselines, comparison strategies, and evidence synthesis workflows.

  • Government environmental officers and geospatial planning professionals supporting land administration, environmental monitoring, and policy implementation.

  • Conservation organisations and development partners seeking to strengthen spatial evidence for habitat protection and ecosystem restoration.

  • Remote sensing consultants and technical advisers supporting environmental assessments, geospatial analytics, and programme reporting.

  • Data engineers and geospatial technology professionals responsible for imagery pipelines, cloud processing, model deployment, and spatial data infrastructure.

  • Policy analysts and sustainability specialists translating geospatial evidence into environmental planning, investment decisions, and performance reporting.

Course Objectives

  • Explain the principles, architectures, pre-training strategies, and transferable representations of Earth observation foundation models and distinguish them from conventional supervised remote sensing approaches.

  • Identify suitable foundation models and data sources for specific environmental evaluation questions by assessing modality, spatial resolution, temporal coverage, geographic transferability, licensing, and computational requirements.

  • Prepare consistent satellite imagery and ancillary geospatial datasets through appropriate quality control, coordinate handling, temporal alignment, normalisation, and metadata documentation.

  • Apply feature extraction, downstream classification, segmentation, regression, and parameter-efficient fine-tuning techniques to environmental monitoring and evaluation problems.

  • Design environmental evaluation workflows that connect model outputs with programme objectives, environmental indicators, baselines, expected outcomes, and clearly defined decision requirements.

  • Develop representative training and validation strategies that account for spatial autocorrelation, geographic separation, temporal dependence, class imbalance, and uncertainty in reference observations.

  • Compare foundation-model performance against conventional machine learning, established remote sensing indices, and simple baseline methods to determine whether additional complexity produces meaningful benefits.

  • Evaluate classification, segmentation, regression, and change-detection outputs using appropriate performance metrics, uncertainty analysis, error diagnostics, and independent reference evidence.

  • Assess the geographic, seasonal, sensor-related, and ecological limitations of model transfer and establish procedures for local adaptation, recalibration, and responsible generalisation.

  • Integrate model-derived information into GIS environments, environmental indicator systems, spatial databases, maps, dashboards, and evaluation reports while preserving traceability and reproducibility.

  • Examine ethical, operational, computational, and environmental sustainability considerations associated with model selection, data access, processing infrastructure, deployment, and continued maintenance.

  • Produce an actionable implementation plan that specifies an environmental use case, data requirements, model strategy, validation framework, staffing, resources, performance indicators, and continuous improvement arrangements.

Comprehensive Course Outline

Module 1: Foundations of Earth Observation Foundation Models

  • Define foundation models and explain how large-scale pre-training produces reusable representations for multiple Earth observation tasks and environmental applications.

  • Compare self-supervised learning, supervised learning, transfer learning, representation learning, and conventional remote sensing classification approaches.

  • Examine how pre-trained geospatial representations can reduce dependence on task-specific labels while still requiring appropriate local adaptation and validation.

  • Identify environmental evaluation use cases where foundation models may provide practical advantages over conventional analytical methods and simpler baselines.

Module 2: Earth Observation Data Ecosystems and Sensor Characteristics

  • Examine multispectral satellite imagery, synthetic aperture radar, hyperspectral data, elevation products, and temporal observations used in environmental analysis.

  • Compare spatial resolution, spectral characteristics, revisit intervals, radiometric properties, coverage, and data availability across relevant satellite missions.

  • Assess the role of vegetation indices, land-cover maps, meteorological observations, soil information, and other ancillary datasets in environmental evaluation.

  • Develop a data selection matrix that links environmental questions to suitable sensors, temporal coverage, spatial scale, quality requirements, and access arrangements.

Module 3: Foundation Model Architectures and Pre-Training

  • Explain transformer-based architectures, convolutional encoders, image patch tokenisation, embeddings, and other representation-learning concepts used in geospatial models.

  • Examine masked image modelling, contrastive learning, temporal pre-training, self-supervised objectives, and multimodal learning for Earth observation.

  • Distinguish general-purpose geospatial models from models designed for specific sensors, spatial resolutions, environmental tasks, or observation modalities.

  • Evaluate how pre-training data composition, geographic representation, temporal coverage, and model architecture influence downstream suitability and transferability.

Module 4: Major Geospatial Foundation Models and Model Selection

  • Review representative models and initiatives, including NASA and IBM's Prithvi, ESA-supported TerraMind, and other relevant geospatial foundation model approaches.

  • Compare model input requirements, supported modalities, available weights, licences, computational demands, documentation, and downstream task compatibility.

  • Examine model cards, technical documentation, benchmark results, pre-training descriptions, and known limitations when evaluating candidate models.

  • Build a model selection framework that considers scientific suitability, local data availability, operational costs, transparency, reproducibility, and institutional capacity.

Module 5: Geospatial Data Preparation and Quality Assurance

  • Apply imagery loading, band selection, reprojection, resampling, tiling, normalisation, masking, and temporal alignment procedures appropriate to the selected model.

  • Identify data quality problems associated with cloud cover, shadows, sensor differences, missing observations, geolocation errors, and inconsistent acquisition conditions.

  • Integrate raster imagery with vector boundaries, field observations, environmental measurements, and reference datasets using consistent spatial identifiers.

  • Maintain metadata and processing records covering source datasets, coordinate systems, preprocessing parameters, quality checks, and data lineage.

Module 6: Embeddings, Feature Extraction and Downstream Learning

  • Generate and interpret geospatial embeddings that represent spatial patterns, land characteristics, environmental conditions, and relationships learned during pre-training.

  • Apply frozen feature extraction with conventional classifiers or regressors to build downstream environmental applications with limited additional training.

  • Compare feature-based approaches with end-to-end fine-tuning, lightweight prediction heads, and parameter-efficient adaptation methods.

  • Select downstream learning strategies according to the number of labelled samples, computing capacity, target task, expected accuracy, and deployment constraints.

Module 7: Fine-Tuning and Domain Adaptation

  • Prepare labelled datasets for adapting pre-trained models to local land-cover classes, vegetation conditions, agricultural systems, or environmental monitoring requirements.

  • Apply appropriate fine-tuning strategies, learning-rate controls, regularisation, checkpoint selection, and early stopping to manage overfitting.

  • Examine domain shift between regions, seasons, sensors, resolutions, and ecological settings, and identify when local calibration or additional labels are necessary.

  • Compare the cost, performance, and reproducibility of local fine-tuning against feature extraction and other lower-complexity adaptation strategies.

Module 8: Environmental Applications in Land Cover and Ecosystems

  • Use foundation-model representations to support land-cover classification, habitat mapping, wetland identification, and monitoring of changing landscape conditions.

  • Examine forest cover change, vegetation disturbance, ecosystem restoration, fragmentation, and other indicators relevant to conservation and environmental programmes.

  • Integrate satellite-derived classifications with ecological field observations and reference inventories to assess environmental change and programme performance.

  • Document class definitions, ecological context, spatial uncertainty, and validation limitations when translating model outputs into environmental indicators.

Module 9: Agricultural, Water and Climate Applications

  • Investigate foundation-model applications for crop type mapping, agricultural condition assessment, irrigation monitoring, and spatial analysis of farming systems.

  • Explore water extent mapping, flood-related change, catchment monitoring, wetland dynamics, and environmental indicators derived from satellite observations.

  • Examine potential applications in climate adaptation, wildfire monitoring, drought assessment, soil-related analysis, and other environmental risk contexts.

  • Distinguish directly observed surface characteristics from inferred environmental processes, ensuring that model outputs are not interpreted as definitive causal or biophysical measurements without supporting evidence.

Module 10: Temporal Analysis and Environmental Change Detection

  • Design temporal analysis workflows that compare environmental conditions across multiple observation periods while controlling for seasonal and sensor-related differences.

  • Apply suitable embedding comparisons, temporal classification, image differencing, and change-detection approaches to identify potential environmental transitions.

  • Investigate abrupt events and gradual trends involving vegetation loss, land conversion, water changes, ecosystem restoration, and urban expansion.

  • Separate genuine environmental change from observation artefacts, model instability, acquisition differences, and missing data through independent checks and sensitivity analysis.

Module 11: Multimodal Foundation Models and Data Fusion

  • Examine methods for combining optical imagery, radar observations, elevation data, vegetation indices, and other geospatial modalities within environmental workflows.

  • Assess the benefits and limitations of multimodal representations when one sensor is unavailable, observations are incomplete, or environmental conditions vary.

  • Develop data fusion strategies that preserve spatial and temporal consistency and account for differing measurement characteristics across input sources.

  • Validate whether multimodal inputs improve downstream environmental assessment compared with single-modality approaches under comparable testing conditions.

Module 12: Model Validation, Benchmarking and Scientific Reproducibility

  • Design spatially independent and temporally appropriate training, validation, and test splits that reduce leakage and overly optimistic performance estimates.

  • Select classification metrics, intersection over union, regression error measures, calibration diagnostics, and change-detection indicators according to the target task.

  • Compare foundation models with conventional remote sensing indices, random forest classifiers, established deep learning models, and other relevant baseline methods.

  • Record model versions, datasets, preprocessing decisions, random seeds, evaluation settings, software dependencies, and validation results to support independent reproduction.

Module 13: Uncertainty, Bias and Geographic Transferability

  • Identify geographic and ecological biases arising from uneven pre-training data coverage, unrepresentative labels, sensor differences, and underrepresented land-cover classes.

  • Evaluate uncertainty associated with ambiguous classifications, rare environmental conditions, incomplete reference data, and transfer to unfamiliar regions.

  • Conduct subgroup and regional performance analysis to identify locations, seasons, ecosystems, or environmental categories where model reliability is reduced.

  • Develop uncertainty communication and escalation procedures that prevent unvalidated predictions from being presented as confirmed environmental facts.

Module 14: GIS Integration and Environmental Evaluation Frameworks

  • Integrate model outputs with GIS layers, administrative boundaries, protected areas, project footprints, environmental baselines, and monitoring databases.

  • Convert model predictions into clearly defined environmental indicators with documented spatial units, temporal periods, aggregation rules, and reference conditions.

  • Link spatial results to evaluation questions, intervention logic, comparison groups, expected outcomes, and relevant contextual evidence.

  • Design maps, dashboards, and spatial summaries that communicate environmental trends, performance differences, uncertainty, and limitations to decision-makers.

Module 15: Deployment, Governance and Sustainable Operations

  • Assess cloud platforms, local computing environments, storage requirements, inference pipelines, model serving options, and operational resource constraints.

  • Examine model licensing, data permissions, privacy, sensitive-location disclosure, cybersecurity, environmental data stewardship, and responsible use requirements.

  • Estimate the computational and energy implications of model selection, inference volume, fine-tuning, and repeated environmental monitoring operations.

  • Establish governance arrangements for model approval, version control, performance monitoring, documentation, maintenance, staff capability, and retirement of unsuitable models.

Module 16: Applied Environmental Evaluation Capstone

  • Define a practical environmental evaluation problem and establish the target indicators, geographic extent, observation period, reference conditions, and intended decisions.

  • Prepare a representative geospatial dataset and select a foundation-model approach that fits the data, task, technical capacity, and validation requirements.

  • Conduct or interpret a downstream analysis, compare results against suitable baselines, and document performance, uncertainty, transferability, and operational limitations.

  • Present an implementation plan and technical report covering the workflow, reproducibility requirements, resource needs, environmental interpretation, and recommendations for responsible use.

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.

Choose Your Training Format

Attend a scheduled session or request a customized programme for your organization.

Classroom Training

Attend instructor-led training at one of our scheduled training locations.

Live Online Training

Join interactive instructor-led sessions remotely from your office or home.

In-House Training

Customize the course content, dates and delivery for your team or organization.

Certification

Participants who successfully complete the programme receive a Certificate of Completion from Upskill Development Institute.

Request Course Information

Need a Tailor-Made Programme?

We can customize the course content, duration, schedule and delivery format to match your organization's objectives.

Dates & Fees Enquire Now

Organizations we've served

Some of Our Recent Clients

We support professional development and institutional capacity building for organizations across a wide range of sectors. Explore a selection of organizations featured in our client portfolio.

SNV logoSNV
National Communications Authority logoNational Communications Authority
Masinde Muliro University of Science and Technology logoMasinde Muliro University of Science and Technology
Kenya Medical Research Institute logoKenya Medical Research Institute
Save Lamu logoSave Lamu
UNICEF logoUNICEF
Fairtrade Africa logoFairtrade Africa
World Agroforestry (ICRAF) logoWorld Agroforestry (ICRAF)
CPI logoCPI
Food and Agriculture Organization (FAO) logoFood and Agriculture Organization (FAO)
University of Southampton logoUniversity of Southampton
Ngali Holdings logoNgali Holdings
STOPAIDS logoSTOPAIDS
International Fertilizer Development Center (IFDC) logoInternational Fertilizer Development Center (IFDC)
Wide Communications logoWide Communications

Learn. Apply. Grow.

Build skills that make a difference at work

Upskill Development Institute offers practical professional training, customized group programmes and flexible learning options to help individuals and organizations strengthen performance.

Chat with us on WhatsApp