Webbläsaren som du använder stöds inte av denna webbplats. Alla versioner av Internet Explorer stöds inte längre, av oss eller Microsoft (läs mer här: * https://www.microsoft.com/en-us/microsoft-365/windows/end-of-ie-support).

Var god och använd en modern webbläsare för att ta del av denna webbplats, som t.ex. nyaste versioner av Edge, Chrome, Firefox eller Safari osv.

 Behnaz Pirzamanbein . Foto

Behnaz Pirzamanbin

Biträdande universitetslektor

 Behnaz Pirzamanbein . Foto

Inferring North American Holocene Land Cover Change using Paleoecological Evidence

Författare

  • Andria Elizabeth Dawson
  • John W Williams
  • Marie-José Gaillard-Lemdahl
  • Behnaz Pirzamanbein
  • Johan Lindström

Summary, in English

Climate-vegetation feedbacks (CVFs) can amplify or mitigate climate variations. These feedbacks operate at multiple timescales. Fast processes such as leaf-scale carbon fluxes operate at sub-annual timescales, while slow processes, driven by changes in forest composition and structure, occur over decadal and longer timescales. Slow processes are rarely directly observed from instrumental data; yet, they are critical to predictions of the terrestrial biosphere including CVFs over the coming decades. Networks of paleoecological data offer a strong observational constraint on long-term vegetation dynamics.
Fossil pollen has been widely used to reconstruct past environments. However, the relationship between pollen and vegetation is complex; it is governed by processes that include differential pollen production, dispersal, and sedimentation. REVEALS is a commonly applied pollen-vegetation model used to produce regional-scale quantitative vegetation reconstructions from fossil pollen by accounting for some of these key processes. These reconstructions can then be subsequently aggregated to land cover types. However, REVEALS is not explicitly spatial, so is not able to estimate vegetation competition for locations and times for which there is no pollen count data. Spatially comprehensive land cover estimates can be inferred using a post-hoc Bayesian spatial model.

Using a network of fossil pollen data, we reconstruct Holocene land cover for North America. First, we use the REVEALS pollen-vegetation model to estimate past vegetation from fossil pollen. Then, we translate these estimates of vegetation to land cover type. We use a Bayesian hierarchical approach to spatially interpolate the point-based land cover reconstructions and generate spatially comprehensive estimates of North American Holocene land cover. We find that there have been large changes within forests, in particular between evergreen and deciduous forests. These shifts impact seasonal energy budgets. We quantify the spatio-temporal extent and magnitude of these land cover shifts, which provides insight about natural variability, climatic response, and CVFs over long timescales.

Avdelning/ar

  • BECC: Biodiversity and Ecosystem services in a Changing Climate
  • Statistiska institutionen
  • MERGE: ModElling the Regional and Global Earth system
  • eSSENCE: The e-Science Collaboration
  • LTH profilområde: Aerosoler
  • Matematisk statistik

Publiceringsår

2022-12

Språk

Engelska

Dokumenttyp

Konferensbidrag: abstract

Ämne

  • Probability Theory and Statistics

Conference name

AGU fall meeting 2022

Conference date

2022-12-12 - 2022-12-16

Conference place

Chicago, United States

Aktiv

Published