Discussion
Loading...

#Tag

  • About
  • Code of conduct
  • Privacy
  • Users
  • Instances
  • About Bonfire
devSJR :python: :rstats: boosted
Eugene
@datastory@mstdn.ca  ·  activity timestamp 7 days ago

🫐 The Blueberry Map Experiment — modelling meets the mountains

In 2022, while living with my family in the Czech Republic, I built a digital map of wild blueberry hotspots in the Jizera Mountains.

At first, it looked like a fun summer project — our neighbors used the map to find the best berry spots and enjoy the landscape.
But behind it was a serious experiment: I tested species distribution modelling (SDM) methods, later adapted for wide-world rare earth element prediction.

Within this “blueberry project” I:
🔹 automated the full spatial workflow in R and QGIS,
🔹 generated geomorphons and other terrain-based predictors,
🔹 built and validated ML models,
🔹 created probability maps and tested them in the field.

✨ What started as a family hobby became a field-tested workflow for predictive geoscience.

#DataScience #MachineLearning #GIS #SpatialModeling #SDM #CzechRepublic #RemoteSensing #Geoscience #RStats #EnvironmentalData #PredictiveMapping #LandscapeEcology #RareEarthElements #CriticalMinerals

Map (Ukrainian text) — Probability map of ripe blueberry occurrence in the Jizera Mountains, generated using species distribution modelling. Green areas show high-probability zones (>0.8).
Map (Ukrainian text) — Probability map of ripe blueberry occurrence in the Jizera Mountains, generated using species distribution modelling. Green areas show high-probability zones (>0.8).
Map (Ukrainian text) — Probability map of ripe blueberry occurrence in the Jizera Mountains, generated using species distribution modelling. Green areas show high-probability zones (>0.8).
Model validation plot (ROC curve) demonstrating high predictive accuracy for the blueberry occurrence model, with AUC near 1.0.
Model validation plot (ROC curve) demonstrating high predictive accuracy for the blueberry occurrence model, with AUC near 1.0.
Model validation plot (ROC curve) demonstrating high predictive accuracy for the blueberry occurrence model, with AUC near 1.0.
Close-up photo of ripe wild blueberries in a mountain forest.
Close-up photo of ripe wild blueberries in a mountain forest.
Close-up photo of ripe wild blueberries in a mountain forest.
Panoramic view of the Jizera Mountains with spruce forest and granite formations under a bright summer sky.
Panoramic view of the Jizera Mountains with spruce forest and granite formations under a bright summer sky.
Panoramic view of the Jizera Mountains with spruce forest and granite formations under a bright summer sky.
  • Copy link
  • Flag this post
  • Block
Eugene
@datastory@mstdn.ca  ·  activity timestamp 7 days ago

🫐 The Blueberry Map Experiment — modelling meets the mountains

In 2022, while living with my family in the Czech Republic, I built a digital map of wild blueberry hotspots in the Jizera Mountains.

At first, it looked like a fun summer project — our neighbors used the map to find the best berry spots and enjoy the landscape.
But behind it was a serious experiment: I tested species distribution modelling (SDM) methods, later adapted for wide-world rare earth element prediction.

Within this “blueberry project” I:
🔹 automated the full spatial workflow in R and QGIS,
🔹 generated geomorphons and other terrain-based predictors,
🔹 built and validated ML models,
🔹 created probability maps and tested them in the field.

✨ What started as a family hobby became a field-tested workflow for predictive geoscience.

#DataScience #MachineLearning #GIS #SpatialModeling #SDM #CzechRepublic #RemoteSensing #Geoscience #RStats #EnvironmentalData #PredictiveMapping #LandscapeEcology #RareEarthElements #CriticalMinerals

Map (Ukrainian text) — Probability map of ripe blueberry occurrence in the Jizera Mountains, generated using species distribution modelling. Green areas show high-probability zones (>0.8).
Map (Ukrainian text) — Probability map of ripe blueberry occurrence in the Jizera Mountains, generated using species distribution modelling. Green areas show high-probability zones (>0.8).
Map (Ukrainian text) — Probability map of ripe blueberry occurrence in the Jizera Mountains, generated using species distribution modelling. Green areas show high-probability zones (>0.8).
Model validation plot (ROC curve) demonstrating high predictive accuracy for the blueberry occurrence model, with AUC near 1.0.
Model validation plot (ROC curve) demonstrating high predictive accuracy for the blueberry occurrence model, with AUC near 1.0.
Model validation plot (ROC curve) demonstrating high predictive accuracy for the blueberry occurrence model, with AUC near 1.0.
Close-up photo of ripe wild blueberries in a mountain forest.
Close-up photo of ripe wild blueberries in a mountain forest.
Close-up photo of ripe wild blueberries in a mountain forest.
Panoramic view of the Jizera Mountains with spruce forest and granite formations under a bright summer sky.
Panoramic view of the Jizera Mountains with spruce forest and granite formations under a bright summer sky.
Panoramic view of the Jizera Mountains with spruce forest and granite formations under a bright summer sky.
  • Copy link
  • Flag this post
  • Block
Log in

bonfire.cafe

A space for Bonfire maintainers and contributors to communicate

bonfire.cafe: About · Code of conduct · Privacy · Users · Instances
Bonfire social · 1.0.0 no JS en
Automatic federation enabled
  • Explore
  • About
  • Members
  • Code of Conduct
Home
Login