Across the Gulf, Earth Observation (EO) has moved from a niche capability to a national priority. Long seen as a tool for scientists and space agencies, EO has now become a backbone for climate monitoring, infrastructure planning, and economic decision-making. Two recent developments illustrate how quickly the region’s ambitions have expanded.
In Saudi Arabia, Neo Space Group (NSG), the Kingdom’s national space company and a Public Investment Fund subsidiary, launched the country’s first dedicated EO data marketplace last month. Powered by UP42, a Berlin-based EO platform recently acquired by NSG from Airbus, the service centralises access to high-resolution satellite imagery, geospatial data products, and advanced analytics, creating an integrated hub for users across sectors.
In parallel, the UAE’s Technology Innovation Institute (TII), the applied research arm of the Advanced Technology Research Council in Abu Dhabi, has been pushing the technical boundaries of EO through its AI-powered coastal and marine monitoring programme. By combining multispectral, hyperspectral, and Synthetic Aperture Radar (SAR) data with advanced machine learning, the institute has developed models capable of detecting oil spills, tracking coral reef health, and monitoring erosion in near real time.
Together, these initiatives cover both ends of the EO value chain. Saudi Arabia has focused on creating a marketplace for data access, while the UAE has advanced the science and AI that transforms that data into actionable insight.
Saudi Arabia opens the market
The Saudi EO Marketplace has been designed as a one-stop shop for public sector agencies, Saudi companies, and international users seeking geospatial data and analytical tools. Meeting regulatory requirements and hosted on secure national infrastructure, it caters to applications spanning environment, infrastructure, energy, real estate, mining, transportation, logistics, and agriculture.
“The launch of the EO Marketplace reflects the Kingdom’s remarkable trajectory of growth and the increasing demand for EO data,” says Martijn Blanken, CEO of Neo Space Group. “Given the Kingdom’s vast geography spanning over 2.15m square kilometres, nearly the size of Western Europe, this platform will serve as both a valuable marketplace and decision-making tool across a wide range of sectors.”
Frank Salzgeber, Acting Deputy Governor for the Space Sector at the Communications, Space, and Technology Commission, has emphasised the timing. “Value creation is shifting downstream in the EO supply chain where AI technologies intersect with space data. The market is now at a perfect inflection point.”
The acquisition of UP42, with its cloud-native platform for EO data access and processing, has provided NSG with an immediate technological foundation. Rebranded for the Saudi market and accessible via sa.up42.com, the marketplace allows seamless discovery, acquisition, and analysis of imagery from multiple satellite providers through automated workflows.
UAE advances the science
If NSG’s marketplace has focused on access, TII’s work has concentrated on capability. The institute’s Directed Energy Research Center (DERC) and AI and Digital Science Research Center (AIDRC) have collaborated on EO models that integrate physics-based algorithms with data-driven AI.
“The EO programme sits at the intersection of these domains, leveraging satellite imagery, remote sensing and AI-driven analytics to monitor and understand dynamic environmental processes, particularly in coastal and marine ecosystems critical to the UAE,” says Dr. Felix Vega, Acting Chief Researcher – Directed Energy Research Center, a part of TII.
The EO programme has deliberately been multidisciplinary. “We have a core team of about a dozen researchers, supported by collaborators across TII’s AI and hardware divisions. Expertise spans satellite remote sensing, photonics, generative AI and machine learning, geospatial analytics, and marine science. This mix allows us to integrate physics-based models with datadriven AI techniques, which has been essential for addressing the UAE’s complex challenges, from water quality monitoring to ship detection and bathymetry,” he adds.
Coastal monitoring and national context
The UAE’s coastline is central to its identity and economy. It supports critical shipping lanes, fisheries, mangroves, coral reefs, and rapidly growing coastal cities, Dr. Vega explains. “At the same time, it has faced unique pressures from rising sea temperatures, industrial activity, and increased maritime traffic. Accurate, timely monitoring has enabled early detection of harmful algal blooms, oil spills, and erosion, which can have significant ecological and economic impacts. Moreover, as the UAE has committed to ambitious sustainability goals, including its Net Zero 2050 strategy, EO has provided the data backbone for evidence-based policymaking and coastal resilience planning.”
TII has used multiple datasets for its observations, combining multispectral, hyperspectral, and SAR imagery. “Multispectral data, such as from Sentinel-2 or WorldView, is ideal for mapping vegetation like mangroves or detecting sediment plumes. Hyperspectral imagery adds finer spectral resolution, allowing us to distinguish subtle differences in water quality indicators such as chlorophyll-a or dissolved organic matter. SAR, on the other hand, is indispensable for oil spill detection, ship tracking, and monitoring in cloudy or nighttime conditions. We often fuse these datasets combining, for instance, optical water quality indicators with SAR-based vessel detection for a comprehensive situational picture,” says Vega.
To harmonise these inputs, TII has integrated data from MAXAR, Planet Labs, UP42, and the UAE Space Agency, despite differences in resolution, revisit rate, spectral bands, and licensing. “Harmonising these datasets has required careful radiometric and geometric correction, modular data pipelines, and cloud based co-located storage and processing,” Dr Vega notes.
AI and modelling innovations
The team has developed two types of AI models: classical machine learning models, including convolutional neural networks for image segmentation and physics-informed machine learning for parameter retrieval, as well as generative AI-based models.
“In classical models, training begins with curated datasets that combine satellite imagery with in-situ measurements from buoys and field campaigns. To handle the high variability of marine environments, changing turbidity, seasonal algae blooms, or sunglint effects, we employ data augmentation and synthetic data generation to expose the models to a wide range of conditions. Continuous retraining with new data ensures that models remain robust as environments evolve,” Dr. Vega explains.
Dr. Hacid adds: “For generative AI models, TII has followed an unsupervised learning approach by training an autoencoder to reconstruct the input imagery. Our design preserves spatial and spectral information effectively and can be adapted for multiple downstream tasks such as classification, segmentation, or change detection. The model uses both established benchmark datasets and a custom dataset built at TII, covering herbaceous wetlands, mangroves, moss and lichen, and ocean environments. This enhances robustness to variability in water turbidity, vegetation density, and seasonal shifts.”
SAR, InSAR and foundation models
SAR has become a cornerstone of monitoring. “Its cloud-penetrating, day-or-night capability has been indispensable,” Dr. Vega says. SAR detects surface roughness changes for oil spills, vessel wakes, or floods, while advanced InSAR tools have mapped millimeter-scale ground deformation, aiding earthquake assessment and infrastructure monitoring.
Hacid notes, “As it is usually difficult to align all the different satellite images in the time domain, the generative AI initiative is working on a foundation model performing intelligent Band-to-Band translation.”
Lessons from field projects
Projects in the Gulf of Oman and coral reef mapping have highlighted the value of sensor fusion and taught the team that no single dataset provides the full picture. “Combining multispectral, SAR, and bathymetric LiDAR has yielded far more actionable insights than any single source,” Dr. Vega says. Strategically, early engagement with policymakers and communities has ensured outputs address real decision-making needs.
“We’ve also seen the value of building scalable pipelines that can pivot to new regions or crises, such as quickly adapting our coral reef tools to mangrove health assessments,” he adds. High-resolution datasets have posed storage and compute challenges. “We have adopted cloud-native architectures with object-based storage and distributed computing frameworks, using tiered storage strategies for cost-effectiveness,” Dr. Vega notes. Continuity has been ensured via cross-calibration and inter-sensor validation, allowing long-term trend analysis even as satellites evolve.
Alignment with national goals and the UAE advantage
The EO programme supports multiple national priorities, from the UAE National Space Strategy 2030 to the Net Zero 2050 initiative. “Coastal and marine monitoring feeds directly into climate adaptation strategies, biodiversity protection, and disaster preparedness. By providing near real-time insights into oil spills, algal blooms, coral bleaching, and mangrove degradation, we empower regulators and industry to respond faster, minimising environmental and economic impact,” says Dr. Vega.
The UAE combines strategic geography, modern infrastructure, and a forward-looking vision. Its location at the nexus of major shipping routes offers a natural testbed for maritime monitoring, while its diverse coastal and desert ecosystems provide opportunities to develop and validate EO techniques for a wide range of environments. World-class data centers and cloud infrastructure support large-scale geospatial analytics.
Dr. Hacid notes that two current technological trends could serve as an advantage for the UAE to become one of the leading players on the field. He sees the emergence of more cost effective and lighter satellites with higher re-visit time and higher resolution rapidly changing the game for space-based EO empowering it with new capabilities but also new problems like synchronisation of data from multiple data sources for real time. Secondly, he sees LLMs and Generative AI transforming EO and offering “a whole new set of capabilities for better analysis, understanding, decision making and automation of downstream tasks.”
“A fully functional space-based Earth observation ecosystem must rely on a coordinated chain of components that work seamlessly from data acquisition to realworld applications. This includes the upstream, midstream, and downstream levels, the policy makers, as well as entrepreneurs and investors. TII plays a role with most of the components of this ecosystem starting with the upstream level with innovations in sensor technologies for example, the midstream level with innovations in communications and large data management, and at the downstream level with concrete applications for specific domains, e.g., costal monitoring or agriculture. TII also supports policy makers by bringing valuable advisory expertise from the field.”
A converging vision
While approaches differ — one commercial and market-oriented, the other research and capability-driven — NSG and TII share the goal of enabling the Gulf to harness EO for environmental stewardship, infrastructure resilience, and economic competitiveness.
Both efforts highlight AI as a multiplier in the EO value chain, whether powering NSG’s analytics marketplace or TII’s generative models for environmental monitoring. They also reflect the strategic ambitions of their countries, Saudi Arabia through Vision 2030, and the UAE through its sustainability and space strategies.
As the Gulf’s EO ecosystem matures, the marketplace and machine learning converge, with Saudi Arabia providing the data and the UAE transforming it into insights. This positions the region not just as a consumer of EO technologies, but as a global leader shaping their future.
Add Comment