Responses of Northern Vegetation Productivity and Phenology to Climate Variability: A Remote Sensing Perspective
Marsh, Hanna LU
(2026)
- Abstract
- Rapid warming in the Arctic is transforming northern ecosystems and altering their role in the global carbon cycle. Rising temperatures are driving shifts in vegetation composition, phenology, and ecosystem functioning, including earlier growing seasons and the expansion of shrubs across tundra landscapes. These changes influence surface energy balance, carbon uptake, and climate feedbacks, yet substantial uncertainty remains in how northern ecosystems respond to environmental variability and long-term climate change. Improving the monitoring and quantification of vegetation productivity and phenology is therefore critical for understanding carbon–climate interactions. This thesis investigates gross primary productivity (GPP) and... (More)
- Rapid warming in the Arctic is transforming northern ecosystems and altering their role in the global carbon cycle. Rising temperatures are driving shifts in vegetation composition, phenology, and ecosystem functioning, including earlier growing seasons and the expansion of shrubs across tundra landscapes. These changes influence surface energy balance, carbon uptake, and climate feedbacks, yet substantial uncertainty remains in how northern ecosystems respond to environmental variability and long-term climate change. Improving the monitoring and quantification of vegetation productivity and phenology is therefore critical for understanding carbon–climate interactions. This thesis investigates gross primary productivity (GPP) and vegetation phenology in northern ecosystems by combining satellite remote sensing, eddy covariance observations, and high-resolution unmanned aerial vehicle (UAV) measurements. By linking observations across spatial scales, the work aims to improve the representation and interpretation of vegetation productivity in rapidly changing Arctic and sub-Arctic environments. Paper I examine lagged effects of spring temperature anomalies on GPP across northern latitudes using multiple satellite-derived products. Paper II evaluates the Plant Phenology Index (PPI) as a proxy for ecosystem productivity and develops a new high-resolution PPI-based GPP dataset for latitudes north of 45° N (2001–2021). Paper III investigates large-scale changes in growing season phenology over the past two decades by comparing satellite-derived estimates with ecosystem model simulations and their climatic drivers. Paper IV complements these analyses with a detailed field study of a High Arctic peatland, using UAV imagery and in situ flux measurements to map vegetation and assess the scaling of greenhouse gas fluxes across heterogeneous landscapes. The results show that northern vegetation is highly responsive to climatic warming. Spring temperature anomalies can enhance early-season GPP but may also induce negative legacy effects that reduce productivity later in the growing season, influencing the annual carbon balance. Among the evaluated vegetation indices, PPI proved to be the most robust proxy for GPP in northern ecosystems based on a Monte Carlo framework testing multiple eddy covariance input–output combinations. Upscaling yields a PPI-based GPP estimate of ~22 Pg C yr⁻¹ for northern latitudes, consistent with established satellite products. Comparisons of phenology reveal substantial variability among ecosystem models, but both models and satellite data indicate a trend toward longer growing seasons, primarily driven by earlier spring onset. In contrast, the timing of the end of the growing season remains highly uncertain and poorly represented in models. The Svalbard field study highlights challenges related to spatial heterogeneity, showing that biases in sampling relative to vegetation distribution can influence ecosystem-scale flux estimates. These findings suggest that future changes in soil thermal conditions and vegetation composition will strongly affect carbon dynamics in High Arctic tundra. Together, the studies demonstrate that integrating multi-scale observations ranging from eddy covariance and UAV data to satellite observations and ecosystem models is essential for accurately monitoring vegetation productivity and phenology in northern ecosystems and for advancing our understanding of Arctic carbon cycling. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/c57ae91b-5dd1-4940-860b-b26e955ad5cf
- author
- Marsh, Hanna
LU
- supervisor
-
- Zheng Duan LU
- Wenxin Zhang LU
- Hongxiao Jin LU
- opponent
-
- Docent Reese, Heather, Göteborgs universitet
- organization
- alternative title
- Klimatresponser hos Nordlig ekosystemproduktivitet och fenologi: Ett fjärranalysperspektiv
- publishing date
- 2026-04
- type
- Thesis
- publication status
- published
- subject
- keywords
- GPP, Phenology, MODIS, PPI, Greenhouse flux dynamics, UAV, Northern vegetation, Arctic tundra
- pages
- 50 pages
- publisher
- Lund University
- defense location
- Världen Lecture Hall, Geocentrum I, Sölvegatan 10, Lund.
- defense date
- 2026-05-22 09:00:00
- ISBN
- 978-91-8104-941-1
- 978-91-8104-942-8
- language
- English
- LU publication?
- yes
- id
- c57ae91b-5dd1-4940-860b-b26e955ad5cf
- date added to LUP
- 2026-04-21 17:58:18
- date last changed
- 2026-04-29 08:48:50
@phdthesis{c57ae91b-5dd1-4940-860b-b26e955ad5cf,
abstract = {{Rapid warming in the Arctic is transforming northern ecosystems and altering their role in the global carbon cycle. Rising temperatures are driving shifts in vegetation composition, phenology, and ecosystem functioning, including earlier growing seasons and the expansion of shrubs across tundra landscapes. These changes influence surface energy balance, carbon uptake, and climate feedbacks, yet substantial uncertainty remains in how northern ecosystems respond to environmental variability and long-term climate change. Improving the monitoring and quantification of vegetation productivity and phenology is therefore critical for understanding carbon–climate interactions. This thesis investigates gross primary productivity (GPP) and vegetation phenology in northern ecosystems by combining satellite remote sensing, eddy covariance observations, and high-resolution unmanned aerial vehicle (UAV) measurements. By linking observations across spatial scales, the work aims to improve the representation and interpretation of vegetation productivity in rapidly changing Arctic and sub-Arctic environments. Paper I examine lagged effects of spring temperature anomalies on GPP across northern latitudes using multiple satellite-derived products. Paper II evaluates the Plant Phenology Index (PPI) as a proxy for ecosystem productivity and develops a new high-resolution PPI-based GPP dataset for latitudes north of 45° N (2001–2021). Paper III investigates large-scale changes in growing season phenology over the past two decades by comparing satellite-derived estimates with ecosystem model simulations and their climatic drivers. Paper IV complements these analyses with a detailed field study of a High Arctic peatland, using UAV imagery and in situ flux measurements to map vegetation and assess the scaling of greenhouse gas fluxes across heterogeneous landscapes. The results show that northern vegetation is highly responsive to climatic warming. Spring temperature anomalies can enhance early-season GPP but may also induce negative legacy effects that reduce productivity later in the growing season, influencing the annual carbon balance. Among the evaluated vegetation indices, PPI proved to be the most robust proxy for GPP in northern ecosystems based on a Monte Carlo framework testing multiple eddy covariance input–output combinations. Upscaling yields a PPI-based GPP estimate of ~22 Pg C yr⁻¹ for northern latitudes, consistent with established satellite products. Comparisons of phenology reveal substantial variability among ecosystem models, but both models and satellite data indicate a trend toward longer growing seasons, primarily driven by earlier spring onset. In contrast, the timing of the end of the growing season remains highly uncertain and poorly represented in models. The Svalbard field study highlights challenges related to spatial heterogeneity, showing that biases in sampling relative to vegetation distribution can influence ecosystem-scale flux estimates. These findings suggest that future changes in soil thermal conditions and vegetation composition will strongly affect carbon dynamics in High Arctic tundra. Together, the studies demonstrate that integrating multi-scale observations ranging from eddy covariance and UAV data to satellite observations and ecosystem models is essential for accurately monitoring vegetation productivity and phenology in northern ecosystems and for advancing our understanding of Arctic carbon cycling.}},
author = {{Marsh, Hanna}},
isbn = {{978-91-8104-941-1}},
keywords = {{GPP; Phenology; MODIS; PPI; Greenhouse flux dynamics; UAV; Northern vegetation; Arctic tundra}},
language = {{eng}},
publisher = {{Lund University}},
school = {{Lund University}},
title = {{Responses of Northern Vegetation Productivity and Phenology to Climate Variability: A Remote Sensing Perspective}},
url = {{https://lup.lub.lu.se/search/files/248225895/Hanna_Marsh_-_WEBB.pdf}},
year = {{2026}},
}