<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <atom:link rel="self" type="application/rss+xml" href="https://researchdata.se/en/catalogue/search.rss?subject=imageryBaseMapsEarthCover"/>
    <link>https://researchdata.se/en/catalogue</link>
    <title>Researchdata.se</title>
    <description>Search results</description>
    <language>en</language>
    <item>
      <title>Classifying Ditch and Stream Channels Mapped From High-Resolution Digital Elevation Models Using Machine Learning</title>
      <description>This data contains the digital elevation models with 0.5 m resolution and polyline shapefiles with the location of channels from the 12 study areas used in this study. 

It also has the scripts to generate the datasets used to train the machine learning model to classify channels into ditches and streams, and calculate the hydrological indices. The code to train the model is also included, along with the models obtained. 

For the ground truth data, the channels were mapped differently based on their type: ditches were manually digitized based on the visual analysis of some topographic indices and orthophotos obtained from the DEM. Streams were mapped by initially detecting all natural channel heads, then tracing the downstream channels, and finally manually editing them based on orthophotos. 

We recommend using Docker to set the environment.</description>
      <pubDate>Tue, 23 Jun 2026 08:38:56 GMT</pubDate>
      <link>https://researchdata.se/en/catalogue/dataset/2025-170</link>
      <guid>https://researchdata.se/en/catalogue/dataset/2025-170</guid>
      <dc:publisher>Swedish University of Agricultural Sciences</dc:publisher>
      <dc:creator>Mariana Busarello</dc:creator>
      <dc:creator>William Lidberg</dc:creator>
      <dc:creator>Anneli Ågren</dc:creator>
      <dc:creator>Florian Westphal</dc:creator>
    </item>
    <item>
      <title>SLU Forage Availability Maps</title>
      <description>This dataset contains the SLU Forage Availability Map.
This dataset contains 12 maps, distributed as .tif files. 

They show the predicted abundance of the pine (Pinus sylvestris), rowan (Sorbus aucuparia), oak (Quercus spp.), willow (Salix caprea), aspen (Populus tremula) and birch (Betula spp.) trees from 10 cm height to 4 cm diameter at breast height across the country using data from the Swedish National Forest Inventory and remote sensing datasets. Map products are aggregated to a 1ha and a 1km2 scale and show the average amount of trees estimated per spatial unit. 

For more information on the development of this dataset, please see the article: 
https://doi.org/10.1016/j.jag.2025.104850

For access to the 10x10m maps, contact the authors.</description>
      <pubDate>Tue, 04 Nov 2025 15:22:50 GMT</pubDate>
      <link>https://researchdata.se/en/catalogue/dataset/2025-259</link>
      <guid>https://researchdata.se/en/catalogue/dataset/2025-259</guid>
      <dc:publisher>Swedish University of Agricultural Sciences</dc:publisher>
      <dc:creator>Lukas Graf</dc:creator>
      <dc:creator>Inka Bohlin</dc:creator>
      <dc:creator>Per-Ola Hedwall</dc:creator>
      <dc:creator>Jonas Dahlgren</dc:creator>
      <dc:creator>Annika M. Felton</dc:creator>
    </item>
    <item>
      <title>GIS-based Time model. Gothenburg, 1960-2015</title>
      <description>The GIS-based Time model of Gothenburg aims to map the process of urban development in Gothenburg since 1960 and in particular to document the changes in the spatial form of the city - streets, buildings and plots - through time. Major steps have in recent decades been taken when it comes to understanding how cities work. Essential is the change from understanding cities as locations to understanding them as flows (Batty 2013)1. In principle this means that we need to understand locations (or places) as defined by flows (or different forms of traffic), rather than locations only served by flows. This implies that we need to understand the built form and spatial structure of cities as a system, that by shaping flows creates a series of places with very specific relations to all other places in the city, which also give them very specific performative potentials. It also implies the rather fascinating notion that what happens in one place is dependent on its relation to all other places (Hillier 1996)2. Hence, to understand the individual place, we need a model of the city as a whole.

Extensive research in this direction has taken place in recent years, that has also spilled over to urban design practice, not least in Sweden, where the idea that to understand the part you need to understand the whole is starting to be established. With the GIS-based Time model for Gothenburg that we present here, we address the next challenge. Place is not only something defined by its spatial relation to all other places in its system, but also by its history, or its evolution over time. Since the built form of the city changes over time, often by cities growing but at times also by cities shrinking, the spatial relation between places changes over time. If cities tend to grow, and most often by extending their periphery, it means that most places get a more central location over time. If this is a general tendency, it does not mean that all places increase their centrality to an equal degree. Depending on the structure of the individual city’s spatial form, different places become more centrally located to different degrees as well as their relative distance to other places changes to different degrees. The even more fascinating notion then becomes apparent; places move over time! To capture, study and understand this, we need a "time model".

The GIS-based time model of Gothenburg consists of:
• 12 GIS-layers of the street network, from 1960 to 2015, in 5-year intervals 
• 12 GIS-layers of the buildings from 1960 to 2015, in 5-year intervals -  Please note that this dataset has been moved to a separate catalog post (https://doi.org/10.5878/t8s9-6y15) and unpublished due to licensing restrictions on its source dataset.
• 12 GIS- layers of the plots from1960 to 2015, in 5-year intervals

In the  GIS-based Time model, for every time-frame, the combination of the three fundamental components of spatial form, that is streets, plots and buildings, provides a consistent description of the built environment at that particular time. The evolution of three components can be studied individually, where one could for example analyze the changing patterns of street centrality over time by focusing on the street network; or, the densification processes by focusing on the buildings; or, the expansion of the city by way of occupying more buildable land, by focusing on plots. The combined snapshots of street centrality, density and land division can provide insightful observations about the spatial form of the city at each time-frame; for example, the patterns of spatial segregation, the distribution of urban density or the patterns of sprawl. The observation of how the interrelated layers of spatial form together evolved and transformed through time can provide a more complete image of the patterns of urban growth in the city. 

The Time model was created following the principles of the model of spatial form of the city, as developed by the Spatial Morphology Group (SMoG) at Chalmers University of Technology, within the three-year research project ‘International Spatial Morphology Lab (SMoL)’.

The project is funded by Älvstranden Utveckling AB in the framework of a larger cooperation project called Fusion Point Gothenburg. The data is shared via SND to create a research infrastructure that is open to new study
initiatives.

1. Batty, M. (2013), The New Science of Cities, Cambridge: MIT Press. 
2. Hillier, B., (1996), Space Is the Machine. Cambridge: University of Cambridge

- 12 GIS-layers of the street network in Gothenburg, from 1960 to 2015, in 5-year intervals. File format: shapefile (.shp), MapinfoTAB (.TAB). The coordinate system used is SWEREF 99TM, EPSG:3006.

- 12 GIS-layers of plots in Gothenburg, from 1960 to 2015, in 5-year intervals. Only built upon plots (plots with buildings) are included. File format: shapefile (.shp), MapinfoTAB (.TAB). The coordinate system used is SWEREF 99TM, EPSG:3006.

See the attached Technical Documentation for the description and further details on the production of the datasets. 
See the attached Report for the description of the related research project.</description>
      <pubDate>Fri, 12 Sep 2025 08:58:29 GMT</pubDate>
      <link>https://researchdata.se/en/catalogue/dataset/snd1154-1</link>
      <guid>https://researchdata.se/en/catalogue/dataset/snd1154-1</guid>
      <dc:publisher>Chalmers University of Technology</dc:publisher>
      <dc:creator>Ioanna Stavroulaki</dc:creator>
      <dc:creator>Lars Marcus</dc:creator>
      <dc:creator>Meta Berghauser Pont</dc:creator>
      <dc:creator>Ehsan Abshirini</dc:creator>
      <dc:creator>Jan Sahlberg</dc:creator>
      <dc:creator>Alice Örnö Ax</dc:creator>
    </item>
    <item>
      <title>Automatic Detection of Ditches and Natural Streams from Digital Elevation Models Using Deep Learning</title>
      <description>This data contains the digital elevation models and polyline shapefiles with the location of channels from the 12 study areas used in this study. It also has the code to generate the datasets used to train the deep learning models to detect channels, ditches, and streams, and calculate the topographic indices. The code to train the models is also included, along with the models with the highest performance in 0.5 m resolution. The channels were mapped differently based on their type: ditches were manually digitized based on the visual analysis of some topographic indices and orthophotos obtained from the DEM. Streams were mapped by initially detecting all natural channel heads, then tracing the downstream channels, and finally manually editing them based on orthophotos.</description>
      <pubDate>Fri, 15 Mar 2024 15:10:56 GMT</pubDate>
      <link>https://researchdata.se/en/catalogue/dataset/2024-57</link>
      <guid>https://researchdata.se/en/catalogue/dataset/2024-57</guid>
      <dc:publisher>Swedish University of Agricultural Sciences</dc:publisher>
      <dc:creator>Mariana dos Santos Toledo Busarello</dc:creator>
      <dc:creator>William Lidberg</dc:creator>
      <dc:creator>Anneli Ågren</dc:creator>
      <dc:creator>Florian Westphal</dc:creator>
    </item>
    <item>
      <title>Synthetic images of corals (Desmophyllum pertusum) with object detection models</title>
      <description>Two object detection models using Darknet/YOLOv4 were trained on images of the coral Desmophyllum pertusum from the Kosterhavet National Park. In one of the models, the training image data was amplified using StyleGAN2 generative modeling.
The dataset contains 2266 synthetic images with labels and 409 original images of corals used for training the ML model. Included is also the YOLOv4 models and the StyleGAN2 network.

The still images were extracted from raw video data collected using a remotely operated underwater vehicle. 
409 JPEG images from the raw video data are provided in 720x576 resolution. In certain images, coordinates visible in the OSD have been cropped.
The synthetic images are PNG files in 512x512 resolution.
The StyleGAN2 network is included as a serialized pickle file (*.pkl).
The object detection models are provided in the .weights format used by the Darknet/YOLOv4 package. Two files are included (trained on original images only, trained on original + synthetic images).

The machine learning software packages used is currently (2022) available on Github:
StyleGAN2: https://github.com/NVlabs/stylegan2
YOLOv4: https://github.com/AlexeyAB/darknet</description>
      <pubDate>Wed, 12 Apr 2023 09:35:40 GMT</pubDate>
      <link>https://researchdata.se/en/catalogue/dataset/2022-98-1</link>
      <guid>https://researchdata.se/en/catalogue/dataset/2022-98-1</guid>
      <dc:publisher>University of Gothenburg</dc:publisher>
      <dc:creator>Matthias Obst</dc:creator>
      <dc:creator>Sarah Al-Khateeb</dc:creator>
      <dc:creator>Victor Anton</dc:creator>
      <dc:creator>Jannes Germishuys</dc:creator>
    </item>
    <item>
      <title>Sentinel-2 Data set for Uganda</title>
      <description>Optical top of the atmosphere reflectance data from the Earth resource satellite Sentinel-2 for Uganda (2018-2020) was downloaded and processed in a systematic way to bottom of the atmosphere reflectance with the open software FORCE (Framework for Operational Radiometric Correction for Environmental monitoring, see https://force-eo.readthedocs.io/en/latest/). For each year (2018-2020), the number of clear sky observations were calculated and a best available pixel mosaic were created. For each quarter (12 in total), for two vegetation indices (EVI = Enhanced Vegetation Index and NDVI = Normalized Difference Vegetation Index) a number of relevant measures were calculated (minimum, maximum, mean, standard deviation, 5%, 50% (median) and 95% quantiles). The data have 10-meter spatial resolution and are stored in GeoTIFF format. 
Data are suitable for studies of vegetation and vegetation changes. More information and details in article published in Data (https://www.mdpi.com/journal/data), 'submitted' 2021-02-15</description>
      <pubDate>Mon, 29 Mar 2021 15:00:09 GMT</pubDate>
      <link>https://researchdata.se/en/catalogue/dataset/2021-50-1</link>
      <guid>https://researchdata.se/en/catalogue/dataset/2021-50-1</guid>
      <dc:publisher>Lund University</dc:publisher>
      <dc:creator>Jonas Ardö</dc:creator>
    </item>
    <item>
      <title>Two maps over the agrarian landscape of the Vadstena plains ca 1640 -- from the dissertation "Ostergotland "Västanstång": studies of the older agrarian landscape and its genesis"</title>
      <description>This study is an investigation of the agrarian landscape in the western part of Ostergotland County (in Sweden). The study was published in the form of a dissertation in Human Geography written (in German) by Steffan Helmfrid, and defended in 1962. Two maps were produced as a part of the dissertation, based on historical material. These maps show the agrarian landscape as it was around 1640. Helmfrid's analogue process for producing these maps was unique for the 1960s. These two maps are made available here.

Two maps are made available here:

1) Name: No Title (Working version)
scale: 1:50,000
coordinate system: no coordinates
file format: tiff (.tif)
file size: 198 MB
file name: helmfrid_orginal_photoshop.tif

2) Name: "The Agrarian Landscape on the Vadstena plains around the year 1640"
scale: 1:100,000
coordinate system: RT 90 2.5 gon V (EPSG 3021)
file format: geotiff (.tif)
file size: 40 MB
file name: helmfrid_rectify.tif

The first map is an unpublished working version that was used as the basis for the second map which is the published version contained in Helmfrid's dissertation. The working version has a larger-scale, which means it is more detailed than the second map. 

The second map was digitized and rectified in 2009 by Johan Berg, also from Stockholm University. This was done as part of a project on land relations during the Younger Iron Age in western Ostergotland. (See link to separate data , 2019-102, in SND's catalog). The working version (the first map) was digitized in 2019 when it was decided to publish Helmfrid's maps via the Swedish National Data Service.

The working version was based on a compilation of 400 smaller village and farm maps, all from the 1640s. These sources can be found in the following historical property maps (in Swedish: geometriska jordeböcker): D5, D6, D7, D8 and D10 (See https://riksarkivet.se/visa-kartsamlingar -- "D" refers to maps from Ostergotland county). All the original historical maps were signed by one surveyor -- Larson Groth. 

The rectified map can be opened in most Geographic Information System (GIS) programs. The map can also be opened in image editing software, but without coordinates. The working version (the first map) is also a raster file, but because the working version was never rectified, it has been saved as a normal tiff file (.tif), not a geotiff. It can be opened in image editing programs like Photoshop or opensource Gimp. 

More information on how Helmfrid produced these maps and a link and reference to his dissertation can be found (in Swedish) in the attached documentation file.</description>
      <pubDate>Wed, 25 Sep 2019 00:00:00 GMT</pubDate>
      <link>https://researchdata.se/en/catalogue/dataset/snd1122-1</link>
      <guid>https://researchdata.se/en/catalogue/dataset/snd1122-1</guid>
      <dc:publisher>Stockholm University</dc:publisher>
      <dc:creator>Staffan Helmfrid</dc:creator>
    </item>
    <item>
      <title>Occurrence patterns of L. pulmonaria and its host trees (aspen and goat willow), together with 400-year fire history data, Teerisuo-Lososuo, Finland</title>
      <description>This data archive contains data used in the study by Fabritius et al. (2019): Estimation of metapopulation colonization rates from disturbance history and occurrence pattern data, Ecology. All data have been previously used also in other studies (Gu et al. 2001, Snäll et al. 2005, Lankia et al. 2012, Ronnås et al. 2017). 

The data include two-time-point occurrence pattern data of an endangered lichen L. pulmonaria, empirical snapshot data on aspen (P. tremula) and goat willow (S. caprea) counts, and 400-year data of forest fires. All data have been collected in the Teerisuo-Lososuo nature reserve, Kuhmo, Finland (2 086 ha; 63°88'N, 29°18'E) and projected in the Finnish YKJ projection.

Ilkka Hanski initiated the surveys in the late 1990’s and Gergely Várkonyi, Matti Komulainen, Jari Huotari, Pirjo Appelgren and Mikko Heikura assisted with field work.

The data include two-time-point occurrence pattern data of an endangered lichen L. pulmonaria, empirical snapshot data on aspen (P. tremula) and goat willow (S. caprea) counts, and 400-year data of forest fires. All data have been collected in the Teerisuo-Lososuo nature reserve, Kuhmo, Finland (2 086 ha; 63°88'N, 29°18'E) and projected in the Finnish YKJ projection, EPSG:2393.</description>
      <pubDate>Tue, 28 May 2019 00:00:00 GMT</pubDate>
      <link>https://researchdata.se/en/catalogue/dataset/snd1104-1</link>
      <guid>https://researchdata.se/en/catalogue/dataset/snd1104-1</guid>
      <dc:publisher>Swedish University of Agricultural Sciences</dc:publisher>
      <dc:creator>Henna Fabritius</dc:creator>
      <dc:creator>Heini Lankia</dc:creator>
      <dc:creator>Tuomo Wallenius</dc:creator>
      <dc:creator>Gergely Várkonyi</dc:creator>
      <dc:creator>Juho Pennanen</dc:creator>
      <dc:creator>Tord Snäll</dc:creator>
    </item>
  </channel>
</rss>