• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 2
  • 1
  • Tagged with
  • 3
  • 2
  • 2
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

En modell som förutsäger naturlig skogstyp utifrån geografik information / A modell that predicts naturual forest types based on geographic information

Iinatti, Sara January 2014 (has links)
Sveriges yta täcks idag av närmare 60 % skog, varav större delen utnyttjas för skogsbruk. Hur skogsbruk ska bedrivas är omdiskuterat, och alternativa metoder till det dominerande trakthyggesbruket har utvecklats. Vissa av dessa metoder, t.ex. Lübeckmodellen, utgår från naturliga förutsättningar för att producera en skog som lämpar sig på en given plats. Syftet med denna studie är att utveckla en modell som med enkla GIS-data översiktligt kan prediktera naturliga skogstyper utifrån andra variabler än befintlig skog, då den skog som finns idag ofta är planterad och därför inte alltid avslöjar en plats naturliga skogstyp. Förhoppningen är att modellen ska kunna vara användbar vid skötsel av och omställning till naturlig skog. Studien baserades på skogsområden vars vegetation antogs vara naturlig (naturreservat, nyckelbiotoper o.dyl.) i Askersunds kommun, Örebro län. En additiv Generalized linear model (GZLM) har använts, och prediktionen baseras på jordmån, berggrund, höjddata och grundvattendata. Resultatet av första testet, baserat på de träningspunkter som utgjorde grunden för modellen, och av valideringen med nya, oberoende punkter från andra skogsområden, skiljer sig åt. Det första testet indikerar att prediktionen fungerar väl för flera av skogstyperna, medan valideringen ger motsatt bild och visar att prediktionen i praktiken inte fungerar. Om modellen vidareutvecklas och kan fungera skulle den kunna underlätta och effektivisera skötsel av skog, t.ex. vid kontinuerligt skogsbruk anpassat efter naturlig skogstyp. Fältarbetet skulle kunna koncentreras till områden som predikterats till lämplig skogstyp. / Sweden is covered by 60 % forestland, and most of it is used in forestry. There is a debate about the forestry methods in Sweden, where clear felling is the most commonly used practice, and alternate methods are being developed. Some of these methods, for example the Lübeck model, is focused on producing the best suited forest for a site, based on the environmental conditions. That forest would probably be of the natural forest type of that location. With this study a model is developed to predict natural forest types based on easily available GIS-data and without using remote sensing. Remote sensing is not considered to be useful in this particular case, since only the already existing vegetation can be analyzed using that method. Existing forests are likely to have been planted, and does not necessarily correspond to the type of vegetation that would develop, were there no human impact. The study was conducted by using forested areas that could be assumed to be natural (within nature reserves and other protected areas) in Askersund municipality, Örebro county, Sweden. An additive Generalized Linear Model (GZLM) has been used, and the prediction is based on soil, bedrock, elevation data and ground water data. The results from the first test, based on the training points on which the model was built, and the validation based on new, independent points from different forest areas, differ. The first test indicate that prediction works well for several of the forest types, while the validation gives the opposite results and shows that the prediction does not actually work. If the model was further developed and would become a working tool, it could make forest management easier and more effective, for example when continuous forestry is carried out. Field work could be concentrated to areas where the forest has been predicted to be of desired type.
2

Forest Growth And Volume Estimation Using Machine Learning

Dahmén, Gustav, Strand, Erica January 2022 (has links)
Estimation of forest parameters using remote sensing information could streamline the forest industry from a time and economic perspective. This thesis utilizes object detection and semantic segmentation to detect and classify individual trees from images over 3D models reconstructed from satellite images. This thesis investigated two methods that showed different strengths in detecting and classifying trees in deciduous, evergreen, or mixed forests. These methods are not just valuable for forest inventory but can be greatly useful for telecommunication companies and in defense and intelligence applications. This thesis also presents methods for estimating tree volume and estimating tree growth in 3D models. The results from the methods show the potential to be used in forest management. Finally, this thesis shows several benefits of managing a digitalized forest, economically, environmentally, and socially.
3

The diversity of fungi in coniferous forests and mixed forests / Svampars diversitet i barr- och blandskog

Asplund, Ida January 2023 (has links)
The fungal domain has one of the highest biodiversities among eukaryotes and species within the domain fill important ecological roles, such as mutualistic mycorrhiza, decomposers, parasites and pathogens. The development of forest ecosystems and their related processes has not only been linked to fungal diversity but the composition and abundance of fungi benefits the abundance and diversity of other species as well. In this study three questions concerning the differences in the fungal communities between mixed forests and coniferous forests were considered. Inventories were done in coniferous forest and mixed forest areas classified as nature reserves and later statistically analyzed. The study could show that while the probability of finding ectomycorrhiza was significantly higher in mixed forests than coniferous forests, the probability of finding saprotrophs was significantly higher in coniferous forests than in mixed forests. Results in contrast to other studies were also found. The study revealed that more is needed to be done on the topic of forest fungi in mixed forest and coniferous forest areas. / Svampdomänen har en av de högsta biologiska mångfalderna bland eukaryoter och domänens arter innehar viktiga ekologiska roller, såsom mutualistisk mykorrhiza, nedbrytare, parasiter och patogener. Utvecklingen av skogsekosystem och deras relaterade processer har inte bara kopplats till svampdiversitet utan svamparnas sammansättning och förekomst gynnar också förekomst av och mångfald hos andra arter. I denna studie behandlades tre frågor om skillnaderna i svampsamhällen mellan blandskogar och barrskogar. Inventeringar gjordes i barrskogs- och blandskogsområden klassificerade som naturreservat och analyserades senare statistiskt. Studien kunde visa att även om sannolikheten för att hitta ektomykorrhiza var signifikant högre i blandskogar än barrskogar, var sannolikheten för att hitta saprotrofer signifikant högre i barrskogar än i blandskogar. Resultat i motsats till andra studier hittades också. Studien visade att det behövs mer forskning om skogssvampar i blandskog och barrskogsområden.

Page generated in 0.0227 seconds