home.social
  1. David O’Sullivan shows how spatial autocorrelation makes sampling fundamentally important: even when two surfaces contain the same values, their spatial arrangement means different sampling schemes can “see” very different patterns 🗺️

    URL: geospatialstuff.com/posts/2025

  2. Last month I had the pleasure of attending the Advances in Spatial Machine Learning 2026 workshop.

    It provided an excellent setting for in-depth discussions, shared learning, and exchange of ideas.

    advsml.github.io/2026/

  3. Thank you to everyone who came to discuss my PICO at today 👋

    “Assessing residual spatial autocorrelation in machine learning models”

    Slides & details: jakubnowosad.com/egu2026/

  4. GDAL can fix invalid geometries in a few different ways, depending on what you want to preserve. The main options are a “linework” approach that sticks closely to the original input, and a “structure” approach that prioritises clean, valid polygons 🛠️

    Since GDAL 3.12, you can also run this directly from the command line with `gdal vector make-valid`.

  5. My presentation at (Vienna):

    > Assessing residual spatial autocorrelation in machine learning models

    📅 6 May | ⏰ 16:30 CEST
    📍 PICO2.6, spot 2

    See you there!

  6. Final deadline extension 📢
    Special Issue: Coding Earth: Open Source Solutions in Physical Geography (Progress in Physical Geography: Earth and Environment)

    We already have a dozen or so submissions and look forward to more.

    Submit by 30 June 2026!

    journals.sagepub.com/home/ppg

  7. a5R brings the A5 pentagonal geospatial index to R.

    Equal-area pentagonal cells across 31 resolutions, encoded as 64-bit integers, with millimetre-level precision at the finest scale 🗺️

    R package by Hugh Graham; a5 by Felix Palmer

    github.com/belian-earth/a5R

  8. Geospatial conferences in 2026
    A curated and growing list of events in GIS and spatial data science

    🔗 github.com/Nowosad/conferences

    Which ones are you planning to attend?

  9. We just published a JOSIS paper on what spatial data science languages have in common and what they still need. Insights from across the R, Python & Julia ecosystems.

    URL: doi.org/10.5311/JOSIS.2025.31.

  10. The spcosa package provides an R framework for spatial coverage sampling.

    Explore examples at git.wur.nl/Walvo001/spcosa

  11. A growing list of 2026 geospatial conferences is live 🌍

    URL: github.com/Nowosad/conferences

    If you know of additional GIS or remote-sensing events, please contribute. PRs and suggestions are welcome.

  12. Call for papers: Coding Earth — Open Source Solutions in Physical Geography for Progress in Physical Geography ⚡

    Show how open-source tools, coding workflows, and open science are reshaping physical geography.

    journals.sagepub.com/pb-assets

    Deadline: 1 March 2026.

  13. New R package: rgeomorphon 📦 by Andrew Brown

    Classifies terrain forms using a parallel C++ implementation of the geomorphon algorithm.

    🔗 github.com/brownag/rgeomorphon

  14. Great new resource from Roger Bivand (NHH, June 2024): slides on spatial econometrics and ML for economic & social research.

    URL: rsbivand.github.io/nem24_talk/

  15. 📍 Registration is open for Spatial Data Science across Languages (SDSL) 2025 – Sept 17–18 (+19), Salzburg, Austria.

    Connect R, Python, Julia & more in spatial science.

    🔗 forms.gle/E9fpG88V2VQQKmjk9 -- Apply for on-site by mid-July – limited spots.

  16. 🚨 CFP: Our special issue *Coding Earth: Open Source Solutions in Physical Geography* is now open! 🌍💻

    We’re seeking papers on open-source tools, coding workflows, and critical reflections in open georesearch.

    🗓️ Deadline: Dec 18, 2025
    🔗 journals.sagepub.com/home/ppg

  17. 🚀 New preprint! "Spatial Data Science Languages: Commonalities and Needs" 🌍

    Exploring challenges & insights from & for spatial data handling—geodetic coords, data cubes, and more!

    🔗 Read here: arxiv.org/html/2503.16686v1

  18. Explore the blog series on comparing spatial patterns in raster data using R. 🌍📊

    - Techniques for analyzing continuous and categorical data
    - Handling overlapping and arbitrary regions
    - Advanced methods for comparing spatial patterns

    Find the full series at buff.ly/s35030O

  19. Here’s a useful list of Diamond Open Access Journals by Lorena Abad, covering topics like Geoinformatics, Remote Sensing, Geomorphology, and more.

    Feel free to add suggestions via PR.

    Check it out: buff.ly/3WxYxnT

  20. ✨ The GeoPAT 2 software allows the segmentation/regionalization of large spatial raster data.✨

    Now, thanks to D G Rossiter, it can now be installed on MacOS.

    You can find all of the instructions and other links at buff.ly/3LEzA4b.

  21. 🛰️ A new paper "scikit-eo: A Python package for Remote Sensing Data Analysis" on a tool for analysis with various machine learning and neural networks algorithms.🛰️

    Article: doi.org/10.21105/joss.06692
    Software: yotarazona.github.io/scikit-eo/

  22. New NLCD products for the year 2021 are now available, and starting from 2024, there will be a new land cover product for the conterminous United States at 30-meter spatial resolution and on an annual time step for the years 1985-2023.

    Read more at usgs.gov/centers/eros/news/nlc