Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Mapping potential location for bilberry picking with remote sensing, local field data andphone application
SLU Swedish University of Agricultural Sciences, Sweden.
RISE Research Institutes of Sweden.
RISE Research Institutes of Sweden.
RISE Research Institutes of Sweden.
Show others and affiliations
2023 (English)Conference paper, Oral presentation with published abstract (Other academic)
Abstract [en]

The aim of the study was to create a practical method for identifying potential locations for bilberry picking with help of remote sensing, local field data and phone application to support the development of the local berry value chain. Local field data w as collected 2021 and 2022 and consisted 503 and 525 plots from a study area of circa 25x45km in Västerbotten, Sweden. The potential for bilberry production was evaluated by measuring the shrub cover and amount of raw berries. Wall to wall remote sensing d ata included a Sentinel 2 image from same summer, airborne laser scanning data from 2020 and other map products. We created classification models for bilberry shrub and yield using both logistic regression (2 classes) and ordinal regression (3 classes) mod els using 2021 data, and validated and calibrated models with 2022 data. Predictor variables consisted of spectral metrics from satellite data; structural metrics from laser data; existing raster maps of tree species, stand attributes, site index, soil moi sture and land use classes. The 2 class models performed better than three class models, delivering the AUC 0.73, overall accuracy 0.83 and kappa value 0.51 for best bilberry shrub model and 0.75, 0.77 and 0.50 respectively for best bilberry yield model. T he best models included both laser based structural metrics describing e.g canopy closure and spectral metrics, but also e.g. volume of pine, soil moisture and site index were found significant predictor variables. Calibration of the models improved annual predictions and the validation of the 2021 raster maps with 2022 data produced similar AUC, OA, and kappa values for bilberry yield (0.73, 0.74 and 0.46), but lower for bilberry shrub (0.61, 0.68 and 0.24). A dedicated phone application was developed duri ng the project, which was used both for collecting the field data and for presenting the potential locations of berry yields. Local berry maps can help berry pickers easier to find the berries in forest landscape and therefore support local berry value cha in. This study is part of the FAIRCHAIN project, which has received funding from the European Union’s funding programme H2020 research and innovation programme under grand agreement 101000723.

Place, publisher, year, edition, pages
2023.
National Category
Environmental Engineering
Identifiers
URN: urn:nbn:se:ri:diva-72590OAI: oai:DiVA.org:ri-72590DiVA, id: diva2:1851535
Conference
Nordic Wildberry Conference. Umeå, Sweden. 7-8 September, 2023
Available from: 2024-04-15 Created: 2024-04-15 Last updated: 2024-04-15Bibliographically approved

Open Access in DiVA

No full text in DiVA

Authority records

Karlsson, Anna-KarinÖstergren, Karin

Search in DiVA

By author/editor
Karlsson, Anna-KarinÖstergren, Karin
By organisation
RISE Research Institutes of SwedenAgriculture and Food
Environmental Engineering

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 10 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf