iiinova — where your digital self keeps evolving

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 10h

🎥 The recording of my talk, “Rethinking Validation for Spatial Machine Learning,” at Machine Learning for Earth Observation (ML4EO) 2026 in Exeter, UK, is now online! 🌍 Watch it at youtu.be/swWqxLznbP4 #ML4EO #SpatialMachineLearning #EarthObservation

001 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 14/10/2026

🌍 The {terra} documentation now clearly shows which calculations for longitude/latitude data use ellipsoidal algorithms (instead of treating the data as projected). Read it at: rspatial.github.io/terra/reference/… #rstats #GISchat #RSpatial

112 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 07/10/2026

{geodensity} by Andrew Brown provides fast kernel density estimation, calculating density directly on the Earth’s surface using great-circle distances. github.com/brownag/geodensity #rstats #GISchat #RSpatial

012 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 30/09/2026

📊 Visualization for Social Data Science is an open online book by Roger Beecham on using visualization and statistics to explore social science data. (It also includes many great maps!) vis4sds.github.io/vis4sds #rstats #DataViz #GIS #RSpatial

001 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 27/09/2026

⛰️ Want to see geomorphometry in action? Book chapter "Geomorphometry Using Open-Source Programming Languages" comes with a companion collection of code and practical examples in #Rspatial #Geopython #Juliageo. 📖 Chapter […] [Original post on fosstodon.org]

042 4

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 23/09/2026

Want to work with the AlphaEarth Embedding data from R? 🛰️🌍 Felipe Carlos has developed an R package that makes it easier to search and retrieve AlphaEarth embeddings for your area of interest. m3nin0-labs.github.io/alphaearth #rstats #RemoteSensing #EarthObservation #RSpatial

020 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 14/09/2026

🚀 NLMR 1.4.0 is now on GitHub! A major terra migration, new `landscape_classify()` & `catalog_nlm()` helpers, and more. github.com/ropensci/NLMR (⚠️ `nlm_fbm()` & `nlm_gaussianfield()` are currently missing -- still figuring out the best way forward […] [Original post on fosstodon.org]

000 0

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 13/09/2026

Missed the OpenGeoHub Earth Observation Summer School 2026? 🛰️ You can now explore the slides, links, and other course materials from the sessions in one place: github.com/Nowosad/OGH_summer_schoo… #EarthObservation #RemoteSensing #Geospatial #OpenGeoHub

011 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 09/09/2026

Working with OpenStreetMap data in R? 🗺️ Louis Laurian and Timothée Giraud have put together a "Cheat sheet for using OpenStreetMap with R". zenodo.org/records/20842874 #rstats #OpenStreetMap #RSpatial

02619 26

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 08/09/2026

What are the key challenges when using machine learning for spatial prediction? In a new blog post, I summarize our recent paper, where we discuss spatial dependence, validation, sampling, uncertainty, applicability, and reproducibility in spatial ML […] [Original post on fosstodon.org]

014 4

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 02/09/2026

🚀 Exciting developments in blockCV v4.0 by Roozbeh Valavi! A major update for spatial cross-validation in R, adding new tools for kNN distance matching, fold diagnostics, and more. github.com/rvalavi/blockCV#whats-ne… #rstats #rspatial #spatialML #GISchat #MachineLearning

000 0

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 30/08/2026

Notes from UseR! 2026: reflections on geocomputation in R, the evolving spatial ecosystem, and why community, validation, and trustworthy maps matter as much as code. jakubnowosad.com/posts/2026-07-21-u… #rstats #rspatial #geocompx #GISchat #UseR2026

012 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 26/08/2026

Working with spatial data in R? The terra cheat sheet is a quick reference for raster/vector operations, projections, plotting, and more. Check it at rspatial.org/_static/terra_cheatshe… #rstats #GISchat #rspatial

097 9

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 24/08/2026

Last week at #EOschool2026 in Istanbul 🇹🇷, I gave a workshop on spatial visualization with tmap. Slides and related scripts are now available at: jakubnowosad.com/ogh2026/docs/tmap.… #rstats #rspatial #GIS

011 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 20/08/2026

Where can your spatial model be trusted? 🌍 Yesterday at #EOschool2026 in Istanbul, I talked about reliable validation of spatial machine learning: evaluate the map, not just the model. Slides (expanded ML4EO 2026 talk) […] [Original post on fosstodon.org]

000 0

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 19/08/2026

New slides from the Earth Observation Summer School 2026 in Istanbul 🇹🇷 "Building reproducible spatial workflows in R": a tour of functions, iterations, and choosing the right spatial tools -- from sf/terra to wk/geos/gdalraster. Slides + code […] [Original post on fosstodon.org]

011 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 05/08/2026

What's better to think about in the middle of summer than geospatial conferences? ☀️ Whether you're looking for your next event or want to share one with the community -- take a look. github.com/nowosad/geospatial-confe… #rspatial #GISchat #GeoSpatial #Conference #OpenScience

002 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 22/07/2026

A resource from Jérôme Guélat for anyone working with spatial data in R. *Spatial Data Processing with R* combines GIS fundamentals with practical, reusable workflows for analysis and map making. 🗺️ jgrun.github.io/spatial #rspatial #rstats #SpatialAnalysis #OpenSource

131 3

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 15/07/2026

Want to learn geocomputation with R? The materials from our *Geocomputation with R* tutorial at UseR! 2026 are now available: slides, hands-on exercises, datasets, and solutions, covering CRS, vector and raster data, and mapping with **tmap** […] [Original post on fosstodon.org]

001 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 13/07/2026

📢 Call for Papers: The Open-Source Approach to Geospatial Applications (Geocarto International) How open-source software, data, and standards are transforming geospatial science—advancing transparency, reproducibility, and innovation across GIS, remote sensing […] [Original post on fosstodon.org]

033 3

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 12/07/2026

How does landscape shape soil respiration? Baumberger et al. apply interpretable machine learning to high-resolution field data, showing how spatial heterogeneity and land use influence soil respiration over space and time […] [Original post on fosstodon.org]

030 3

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 08/07/2026

A world still to be mapped 🌍 Today at UseR! 2026 in Warsaw, I gave a talk "A world still to be mapped: reflections on geocomputation in R". A look at the #rspatial history, current state, my challenges, and more. Slides: jakubnowosad.com/user2026 #rstats #geocomputation #UseR2026

112 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 05/07/2026

🚂 Heading to useR!2026 in Warsaw (6–9 July 2026). user2026.r-project.org I’ll be around and keen to meet others working on #rspatial and geocomputation in R (etc.). #useR2026 #rstats #rspatial

000 0

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 03/07/2026

⏰️ Still time to sign up for our "Geocomputation with R" workshop at UseR! 2026 (July 6, 9AM). A 3-hour intro to spatial data in R using sf, terra, and tmap by Jannes Muenchow and me. user2026.r-project.org #rspatial #geocompx #RStats #USER2026

012 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 02/07/2026

🚦 New workshop from #ML4EO2026: "Where your models can be trusted" Learn how to evaluate spatial ML using kNN distance matching CV, Area of Applicability (AoA), and more. All with R examples. Materials: slides, code, data, hands-on exercises are at […] [Original post on fosstodon.org]

023 3

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 29/06/2026

🌍 My #ML4EO2026 talk focused on spatial machine learning and prediction-domain adaptive evaluation: defining the prediction domain, adapting validation, and weighting evaluation by deployment location properties. Slides: jakubnowosad.com/ml4eo2026 […] [Original post on fosstodon.org]

011 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 28/06/2026

📘 Geomorphometry: Concepts, Software, Applications (2nd ed.) by Reuter, Grohmann & Lecours is coming this fall! A new edition covering approaches and applications in quantitative terrain analysis and Earth surface processes […] [Original post on fosstodon.org]

000 0

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 24/06/2026

🗺️ New in the development version of the {tmap} R package: blending modes are now available! This makes it much easier to create more expressive terrain visualizations. github.com/r-tmap/tmap/issues/815 #rstats #tmap #GISchat #rspatial

022 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 23/06/2026

⏳ One week remaining to submit to "Coding Earth" in Progress in Physical Geography: Earth and Environment. The issue focuses on open-source tools, geospatial workflows, and code-driven physical geography research. Deadline: 30 June 2026. Url […] [Original post on fosstodon.org]

031 3

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 17/06/2026

📍 Interesting early-stage project: RQGIS Console An R console integrated directly into QGIS -- letting you write and run R code inside QGIS with full access to your active project (layers, CRS, extent, and metadata). github.com/jsmendozap/rqgis #QGIS #RStats #GIS #Geocomputation #RSpatial

046 6

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 14/06/2026

🌍 New paper on key challenges in spatial machine learning! We discuss six major topics shaping the field: model validation, performance metrics, uncertainty quantification, algorithmic diversity, software implementation, and modeling protocols. The paper […] [Original post on fosstodon.org]

135 5

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 10/06/2026

📄 New(-ish) post by David O’Sullivan A clear, hands-on exploration of the Modifiable Areal Unit Problem using simulation and R, showing how scale and aggregation can reshape spatial patterns and interpretation […] [Original post on fosstodon.org]

016 6

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 07/06/2026

🛍️ The geocompx project now has a shop! You can support the project and grab a geocompx t-shirt or mug, spreading open-source spatial data science across R, Python, Julia, and beyond. geocompx.tpopsite.com #geocompx #FOSS4G #GIS #OpenScience

003 3

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 03/06/2026

🌍 Calling R, geospatial & Earth Observation enthusiasts! Take on the R-for-eaRth Data Challenge: methane emission estimation from Sentinel-2 data. Deadline: June 30, 2026 Url: github.com/CredibleAI/R-for-eaRth #RStats #RSpatial #EarthObservation #OpenScience

046 6

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 31/05/2026

📄 New our preprint: Moving Beyond Spatial and Random Cross-Validation in Environmental Modelling: A Call for Prediction-Domain Adaptive Evaluation arxiv.org/abs/2605.13689v1 #SpatialStats #Ecology #MachineLearning

063 6

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 29/05/2026

⏰️ Reminder: submission deadline for the Spatial Data Science Award is soon (May 31). If you’re planning to nominate people contributing to geospatial open science, OSS development, and community building, now is the time to submit […] [Original post on fosstodon.org]

013 3

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 26/05/2026

📍 The David G. Rossiter spatial analysis resources page has moved and hosts a wide set of tutorials and examples on soil geography and geospatial data science. swampstomper.nl/professional/# #spatialanalysis #gischat #rstats #rspatial

011 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 24/05/2026

📢 The geocompx project has a new (and fully reproducible) logo! geocompx is a community-driven initiative for reproducible geocomputation across languages, building open resources for spatial data analysis, modeling, and visualization in R, Python, Julia, and […] [Original post on fosstodon.org]

058 8

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 22/05/2026

⏰ Early-bird registration ends 31 May for the #EOSummerSchool2026 in Istanbul! A week of hands-on training in Earth Observation data science (Python, R, Julia), 17–21 Aug. pretix.eu/opengeohubfoundation/EOSu… #EarthObservation #RSpatial #GISchat

012 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 17/05/2026

Nominations are now open for the Lorena Abad Crespo Award for Innovation in Spatial Data Science! 🎉 Know someone doing outstanding work in spatial data science, GIS, or geocomputation? Nominate them here: spatial-data-science.github.io/awar… […] [Original post on fosstodon.org]

002 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 13/05/2026

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 […] [Original post on fosstodon.org]

100 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 10/05/2026

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 #SpatialMachineLearning #GISchat #MachineLearning

011 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 07/05/2026

Thank you to everyone who came to discuss my PICO at #EGU26 today 👋 “Assessing residual spatial autocorrelation in machine learning models” Slides & details: jakubnowosad.com/egu2026 #MachineLearning #SpatialAnalysis #EGU2026

000 0

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 06/05/2026

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 […] [Original post on fosstodon.org]

001 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 05/05/2026

My presentation at #EGU26 (Vienna): > Assessing residual spatial autocorrelation in machine learning models 📅 6 May | ⏰ 16:30 CEST 📍 PICO2.6, spot 2 See you there! #MachineLearning #SpatialAnalysis #GISchat

000 0

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 01/05/2026

🚂 Looking forward to attending #EGU26 next week and contributing several works on spatial prediction and machine learning. If you are attending, I would be glad to connect and discuss related work -- feel free to reach out. #GISchat #Geocomputation #RSpatial

010 1

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 29/04/2026

Looking for geospatial conferences in 2026? Here’s a growing community-maintained list: github.com/Nowosad/conferences_2026 🗺️ Add missing events via PR. #gischat #GIS #geocomputation

274 7

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 26/04/2026

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! […] [Original post on fosstodon.org]

000 0

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 22/04/2026

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 #rstats #rspatial #dggs

022 2

Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 20/04/2026

R Dev Day @user! 2026 (Warsaw, July 10): a free, hybrid event for contributing to base R. Places are limited; participants selected via self-nomination. Please complete the nomination form by April 24 🛠️ user2026.r-project.org/additional/r… #rstats #useR2026

033 3