Geospatial Data Visualization in Science

The core ideas of GIS haven’t changed: data, analysis, visualization, sharing. But how we build and scale them has evolved dramatically. Traditionally, we used traditional desktop GIS for analysis and cartography. Now, many are shifting to Jupyter, GeoPandas, and QGIS. Tools that are open-source, scriptable, and interoperable. They make automation and reproducibility the default. Where ArcGIS Server once hosted services, the modern equivalent is filled by cloud storage (S3, GCS, Azure) combined with STAC catalogs, Iceberg, and other data lakehouse models. Open ecosystems where data and metadata can live side by side and be accessed by any tool. Instead of feature layers, we now use GeoParquet or Iceberg tables. These open formats separate storage from compute, allowing the same dataset to be analyzed in many upstream systems. The old WMS model of rendering images of maps on a server is being replaced by PMTiles, which let maps stream efficiently from object storage and make large datasets browsable instantly with no server backend. Where we once used "Web GIS" for hosting and sharing, today that role is And ArcPy, the scripting workhorse of desktop GIS, has evolved into a broader ecosystem of Python, SQL, Apache Sedona, PySAL, and DuckDB: powerful, lightweight, and built for integration with modern data workflows. What used to be proprietary and monolithic is now open, modular, and cloud-native. Data isn’t published, it’s queried. Maps aren’t hosted, they’re streamed. And analysis isn’t locked into one tool It’s scalable, scriptable, and shareable. The future of GIS isn’t a platform, it’s a stack. 🌎 I'm Matt and I talk about modern GIS, earth observation, AI, and how geospatial is changing. 📬 Want more like this? Join 10k+ others learning from my newsletter → forrest.nyc

Most GIS students never learn this side of Geospatial Science… and that’s EXACTLY why they struggle. A few days ago, I revisited one of the most detailed tutorials ever created on using QGIS for 🌊 marine data analysis - global ocean temperature, salinity, currents, SSH, vector fields, and more. (Prepared by CMEMS Service Desk — brilliant work.) And it hit me again: 👉 Most GIS learners have NEVER been exposed to real scientific or applied geospatial workflows. They only learn: 🔹 How to digitize 🔹 Basic QGIS tools 🔹 A few shapefiles 🔹 A certificate from a random course But the world of geospatial science is MUCH bigger and far more exciting. Look at what this one tutorial teaches (page after page): ✔ Processing ocean temperature & salinity ✔ Working with NetCDF marine datasets ✔ Creating SSH contours ✔ Visualizing ocean currents with vector fields ✔ Sampling rasters with regular grids ✔ Overlaying multi-variable ocean layers ✔ Generating coastal, regional & global maps ✔ Using plugins like VectorFieldCalc, NetCDF Browser, Contour, etc. This is REAL GIS - the kind used in climate science, marine research, weather modelling, and environmental intelligence. And THIS is the kind of skillset that makes you stand out in the job market. 🌍 Why am I sharing this? Because this is exactly why I built Harita Hive- to pull learners out of the digitizing trap and into the world of applied geospatial technology where real opportunities exist: 🛰 Remote sensing 🌦 Climate & environmental analytics 🌊 Marine modelling 📊 Spatial data science 🤖 GeoAI 🗺 WebGIS & automation 🌡 NetCDF, MODIS, CHIRPS, CMEMS, ERA5, etc. Our mission is simple: 👉 Help students build REAL skills with REAL datasets so they can build REAL careers. Not just “learn QGIS”… but learn how GIS is used in the real world - exactly like the tutorial above shows. 📚 For the GIS community If you’ve never worked with marine or climate datasets, this document is worth saving. It’s one of the cleanest walkthroughs for oceanographic mapping I’ve seen. Huge respect to CMEMS Service Desk for making it accessible. 🚀 If you want to learn workflows like this - step-by-step, from scratch: Our flagship program Geospatial Technology Unlocked is starting soon. We teach GIS, RS, Python, WebGIS, GeoAI, automation — from basics → real projects → job-ready skills. If you want to grow beyond “digitizing”… If you want to work with real data, real tools, real science… Drop “GIS” in the comments, and I will send you the curriculum + sample projects + free resources. Join our WhatsApp Community: https://lnkd.in/diwFkgBm Let’s build geospatial talent the world actually needs.

Maps are not the product. Decisions are. In Ag science, we've made significant progress in Earth observation, model quality, and dashboarding. But across many teams, the final mile remains weak: turning spatial signals into action. A workflow I've found useful across projects is simple: Signal → Confidence → Priority zone → Actions - If we skip the confidence layer, we create noise. - If we skip ownership, we create beautiful analytics with no impact on fields. The shift that matters most is moving from "Can we detect it?" to "Can a commercial or agronomy team act on it?" Remote sensing plays an important role here, but it rarely works as a standalone answer in live-season agriculture. It detects spatial variation, it doesn't always diagnose the cause, and timing is often constrained by data availability, cloud cover, and field scale. Its value increases when combined with other data layers, and when outputs are designed for action, not just interpretation. That means: - Explicit uncertainty around recommendations - Clear threshold logic for prioritization - Decision outputs designed for non-technical users Geospatial analytics creates value when it improves timing, targeting, and coordination, not when it only improves map resolution. #AgDataScience #GeospatialAnalytics #DecisionIntelligence

With references I received reaching out to ask about the best approach to visualizing Nigerian population data, I decided to create this practical guide to choropleth mapping! Here is a comprehensive resource on creating meaningful choropleth maps using Python. This isn't just another technical tutorial it's focused on the important question:  "How do classification choices impact how we understand sensitive demographic data?" 📋 Why method selection matters for equitable data representation and how to avoid common visualization pitfalls. 🌍 Why this matters: The questions I received highlighted how many professionals struggle with moving beyond "which map looks best" to "which method tells the most accurate story." This guide provides that bridge. 📊 Featured methods include: ⚖️ Equal Intervals | 📈 Quantiles | 📐 Std Mean | ✂️ Maximum Breaks | 🎯 Head/Tail Breaks | ⭐ Optimal Methods | 🔍 Jenks-Caspall | 🎨 Fisher-Jenks | 🧩 Max-p I'd love to hear what classification methods have worked best in your projects! #DataVisualization #GeospatialAnalysis #Python #ChoroplethMaps #DataScience #SpatialAnalysis #NigeriaData #OpenSource #GIS #CensusData #DataEthics #Mapping #PySAL

Ancient Stone Walls And Power - What Data Science Tools Can Reveal In African Archaeology -- https://lnkd.in/gQzYWyRN <-- shared technical article -- https://lnkd.in/gYbcQuew <-- shared paper -- https://lnkd.in/gNg3j_6j <-- shared paper -- “Visibility has always been important in people's decisions about where to live and how to arrange their spaces. People make connections with what they can see. Being able to see prominent landmarks, such as certain mountain peaks, rivers or ancestral sites, could help reinforce a community's connection to its cultural and spiritual landscape. Some people prefer homes with scenic views, such as apartments overlooking parks or waterfronts, and businesses often choose locations with high visibility to attract customers. In both ancient and modern contexts, visibility plays a key role in how people position themselves in their environment. That's why visibility is a useful concept when studying the past. Archaeologists are interested in what visible and hidden spaces meant to people in long-ago cultures. They have used the idea of visibility to examine things like where settlements were located, socio-political relationships as well as when and where people chose to move. In the past 30 years, they've been helped in these studies by digital tools like geographic information systems (GIS). GIS is a computer system that uses software and data to map, analyze and manage geographic information. But this method is still underutilized in Africa. It has only recently been taken up and very few visibility studies have been conducted on the continent. [The author is] a geospatial data scientist who specializes in uncovering spatial patterns and relationships in archaeological data. [He] work[s] with the Arcreate project, a group of researchers working on mobility, migration, creativity and knowledge transmission in African societies. Recently [he] published a study [link above] of 19th century settlements in the Magaliesberg region in South Africa, using GIS tools to analyze what the visibility of the sites was telling us. Were the settlements designed and positioned to be more visible or less? And did this say something about what mattered to the people who lived there? [He] hope[s] my study serves as a framework for comparative analyses of other African sites in archaeology and sheds some light on what went into the choice of these locations…” #GIS #spatial #mapping #ancient #stonewall #ancienthistory #history #Africa #archaeology #datascience #spatialanalysis #visibility #culture #cultural #landmarks #ancestral #community #village #home #residential #environment #viewshed #visible #hidden #settlements #walls #sociopolitical #spatiotemporal #Arcreate #Magaliesberg #visibilityanalysis #landscape #Kaditshwene #Molokwane #Marothodi #SothoTswana #farming #agriculture #LateFarmingCommunities #LiDAR #household #kraal #homestead

An older animation of an unpublished project of mine just resurfaced - the NDVI time laps of Budapest. As for #NDVI: "𝘕𝘰𝘳𝘮𝘢𝘭𝘪𝘻𝘦𝘥 𝘥𝘪𝘧𝘧𝘦𝘳𝘦𝘯𝘤𝘦 𝘷𝘦𝘨𝘦𝘵𝘢𝘵𝘪𝘰𝘯 𝘪𝘯𝘥𝘦𝘹: 𝘛𝘩𝘦 𝘯𝘰𝘳𝘮𝘢𝘭𝘪𝘻𝘦𝘥 𝘥𝘪𝘧𝘧𝘦𝘳𝘦𝘯𝘤𝘦 𝘷𝘦𝘨𝘦𝘵𝘢𝘵𝘪𝘰𝘯 𝘪𝘯𝘥𝘦𝘹 𝘪𝘴 𝘢 𝘴𝘪𝘮𝘱𝘭𝘦 𝘨𝘳𝘢𝘱𝘩𝘪𝘤𝘢𝘭 𝘪𝘯𝘥𝘪𝘤𝘢𝘵𝘰𝘳 𝘵𝘩𝘢𝘵 𝘤𝘢𝘯 𝘣𝘦 𝘶𝘴𝘦𝘥 𝘵𝘰 𝘢𝘯𝘢𝘭𝘺𝘻𝘦 𝘳𝘦𝘮𝘰𝘵𝘦 𝘴𝘦𝘯𝘴𝘪𝘯𝘨 𝘮𝘦𝘢𝘴𝘶𝘳𝘦𝘮𝘦𝘯𝘵𝘴, 𝘰𝘧𝘵𝘦𝘯 𝘧𝘳𝘰𝘮 𝘢 𝘴𝘱𝘢𝘤𝘦 𝘱𝘭𝘢𝘵𝘧𝘰𝘳𝘮, 𝘢𝘴𝘴𝘦𝘴𝘴𝘪𝘯𝘨 𝘸𝘩𝘦𝘵𝘩𝘦𝘳 𝘰𝘳 𝘯𝘰𝘵 𝘵𝘩𝘦 𝘵𝘢𝘳𝘨𝘦𝘵 𝘣𝘦𝘪𝘯𝘨 𝘰𝘣𝘴𝘦𝘳𝘷𝘦𝘥 𝘤𝘰𝘯𝘵𝘢𝘪𝘯𝘴 𝘭𝘪𝘷𝘦 𝘨𝘳𝘦𝘦𝘯 𝘷𝘦𝘨𝘦𝘵𝘢𝘵𝘪𝘰𝘯." In practice, the NDVI is a +1.0 to -1.0 float-valued variable we can compute by combining a satellite image's red and near-infrared bands. This way, due to the light absorption properties of the green surfaces, the result will be higher values for those pixels that are greener. As I used free #sentinel satellite images, my pixel size here is 10x10m. On the visualization, I used a green-yellow-red color map to illustrate the NDVI values from +1.0 to 0. Negative values typically mean water bodies; near-zero positive values are good references for built-in areas, while values around 0.1-0.5 are sparse and, above that, dense green vegetation. Such maps, and especially the temporal evolution of the data behind them, can be a great tool for environmental monitoring, from detecting deforestation to assessing the green policies and efforts of city developers and governments. #GIS #spatialanalytics #geospatialdata #geospatial #datascience #datavisualization

🚀 Accessing and Visualizing Planet’s Tanager Hyperspectral Data 🌍🛰️ Planet has just released open-access hyperspectral imagery from its Tanager mission — a major step forward for Earth observation and spectral analysis! In this tutorial, I walk you through: ✅ Accessing the Tanager hyperspectral data ✅ Converting the data into a rectangular grid ✅ Visualizing imagery and spectral signatures interactively All powered by the HyperCoast Python package. 🎥 Tutorial video: https://lnkd.in/eXtBTUc2 📓 Notebook example: https://lnkd.in/efYuR9qN 💻 GitHub repo: https://lnkd.in/gkGpJmze 📚 Documentation: https://hypercoast.org #Tanager #Hyperspectral #Geospatial #OpenData #DataScience #Python

What do embeddings mean for geospatial viz? Have you heard about Google DeepMind's AlphaEarth Foundations and are curious about how it will impact core geospatial data as we know it? 🛰️ Here's why AlphaEarth is a game-changer for geospatial visualization: For decades, working with satellite imagery has meant starting from scratch. - Want to do land cover analysis? Time-series comparisons? Train an ML model? Traditionally, as GIS and data analysis professionals, we had to: 🔧 Download raw satellite data ⚒️ Apply cloud masking, atmospheric correction, spectral transforms, and ✂️ Stitch together multi-temporal, multi-modal, and often low-resolution datasets (standard has been 30 m) 🛠️ ...all before you get to your actual analysis. 🌍 And while datasets like DEMs (Digital Elevation Models), DTMs (Digital Terrain Models), and land classification layers exist, they’ve historically been: 🗾 Regionally limited 🕰️ Infrequent or outdated 👾 Inconsistent in resolution or processing methods 😖 Or simply hard to access Even when data was available, making it analysis-ready took specialized tools and was a major time investment (that many can’t afford). 🚀 Now, with AlphaEarth Foundations: A global, 10-meter resolution dataset of analysis-ready satellite embeddings — for every year since 2017. No need for: 🚫 Atmospheric correction 🚫 Cloud masking 🚫 Speckle filtering 🚫 Custom featurization or preprocessing I don't know about you all, but the fact that I can access a high-quality, 10 meter res dataset for the entire world is a GAME CHANGER. The model treats satellite imagery like frames in a video 📽️ — learning across space, time, and modality — to generate rich embeddings that understand terrain, land use, and change over time. 📊 For those of us in geospatial visualization, climate research, urban planning, or environmental storytelling, this changes the game: ✅ You can now create high-impact, insight-rich visualizations ✅ Without heavy preprocessing or remote sensing expertise ✅ With consistent, global coverage at meaningful resolution This opens the door for faster, more accessible, and more equitable Earth analysis — from solo mapmakers to enterprise analysts. We're entering a new chapter: 📌 Earth data that’s ready to use — right out of the box. Now you are not stuck with the choice of which datasets to use, or tasked with finding and layering them. This is composited data churned from analyses of Sentinel-1, Sentinel-2, Landsat 8 and Landsat 9, GEDI Raster Canopy Height metrics, GLO-30 DEM, ERA5-Land Reanalysis Monthly Aggregates, ALOS PALSAR-2 ScanSAR, GRACE monthly mass grids, and several text sources, and will change how compact our data layers are. How do you plan to use this? OR How would you imagine using precomputed, analysis-ready embeddings at a 10-meter resolution? Let us know in the comments! 👇 Google #3D animation #AI #GIS

🚀 Excited to share a new YouTube tutorial I created for students and researchers working with climate and wildfire data! 🌍🔥 Many struggle with extracting information from NetCDF files, especially for GFAS (Global Fire Assimilation System) Wildfire CO₂ data. To help, I’ve made a short and practical video showing how to easily extract data, make maps, and generate graphs in R for the California region. ✨ This video also shows you how to create high-quality maps and graphs that are publication-ready — no extra editing needed! 📊🗺️ 🎥 Watch the video here: https://lnkd.in/ga3bWbHe I hope this will be a useful resource for those working in environmental science, wildfire research, remote sensing, and climate studies. 🙌 💡 Curious to explore how geospatial modeling and data science can help us better understand and mitigate these risks? Let’s connect and discuss! #NetCDF #RStats #GFAS #Wildfire #CO2 #RemoteSensing #ClimateData #DataScience #Geospatial #California #Wildfire #ClimateChange #RStats #DataVisualization #FireEcology #GeospatialAI #RemoteSensing #MachineLearning #ForestMonitoring #EarthObservation #GIS #Geoinformatics #EnvironmentalMonitoring #SustainableForestry #ClimateMonitoring #DataScience #SpatialAnalysis #VegetationMapping #OpenSourceGIS #AIResearch #MLProjects #Research #ForestConservation #SustainableDevelopment #EcoMonitoring #BigData #GeospatialData #SatelliteImagery #SpatialComputing #DeepLearning #ComputerVision #DigitalTwin #LandUseChange #UrbanForestry #ClimateAction #GreenTech #GeoAI #BiodiversityMonitoring #TechForGood #OpenScience #SmartForests #ConservationTech #GeoInnovation #InnovationInGIS #WashU #SLU_STL