Halftime review

Adrian Gustafson
Department of Physical geography and Ecosystem Analysis
Centre for Environmental and Climate Research

Bridging the gap between global climate change

and its local and regional impacts in the Arctic

Introduction
Climate change in the Arctic
Global anthropogenic climate change is also affecting the northern high latitudes. In fact while the
global land surface temperature have warmed 0.78°C since 1950 (IPCC 2014), the Arctic have
warmed twice as fast. In the warmest places over temperature have increased by 2°C over the
last 50 years (AMAP 2017). This change is expected to continue and climate models are
projecting that the Arctic will have warmed with 3-5°C by 2050 and with 5-9°C by the late century
under a RCP4.5-scenario (AMAP 2017). Like temperature, climate models are projecting
increased precipitation. A larger part of which is projected to fall as rain instead of snow in winter
(AMAP 2017; Bintanja and Andry 2017).

This phenomenon with stronger climatic change in the Arctic than for the global average is termed
Arctic amplification (Serreze and Barry 2011). It is caused by a number of amplifying feedbacks
such as declining sea ice extent and changed albedo. These feedbacks are complex as they are
acting in different seasons and are interacting with each other.

Ice acts as an insulator between a warmer Arctic ocean and a colder atmosphere in winter. If this
insulation is weakened more long wave radiation and latent heat flux will be released to the
atmosphere with a warming effect. A dark, ice free surface also absorbs more incoming solar
radiation which further heats the water and weakens the ice (Serreze and Barry 2011). Similarly, a
reduction of albedo from earlier snowmelt in spring will absorb a larger amount of incoming solar
radiation. As a consequence the land surface will heat further and increase the sensible and latent
heat fluxes to the atmosphere.

An increased latent heat flux might have a cooling effect on the land surface (Zhang et al. 2014).
However, a greater amount of water vapour in the atmosphere might also increase the cloud
cover. Depending on the type and state of the cloud this could be a negative feedback to the
system as high cloud albedo will reflect incoming shortwave radiation. Contrary, it might also act
as a reinforcing feedback as it will augment the downward long wave radiation.

These biogeophysical feedbacks will also influence the biogeochemical cycles through a thawing
permafrost (permanently frozen soils). Frozen soils, peatlands and sediments store very large
amounts of carbon. The exact amount is difficult to estimate and the estimates have continuously
grown in size from 192 Pg to 1466 Pg (van Huissteden and Dolman 2012). Hugelius et al.
estimated Arctic carbon stocks in top 3 meters to be approximately 1300 Pg with an approximate
uncertainty of ±200 Pg (Hugelius et al. 2014). Out of these 1300 Pg approximately 500 Pg is not
frozen in the ground. As permafrost thaws from a warming Arctic, a substantial carbon pool is
thus at risk of becoming subject to decomposition and release CO2 or CH4 to the atmosphere
(van Huissteden and Dolman 2012).

1
Climate induced vegetation change
Climate change also induce change in the biosphere. Plants are responding to increase in CO2-
concentration, temperature and precipitation (Elmendorf et al. 2012; Myers-Smith et al. 2015).
Generally a greening of the Arctic have been observed as a response to increased winter
temperatures and longer growing seasons (Xu et al. 2013). One observed change is an increase of
shrubs (Myers-Smith et al. 2011). Shrubs are woody, many-stemmed vascular plants with diverse
growth forms. They can range in hight from tall 3-4 meter high, e.g., Salix spp. and Alnus spp., to
a creeping prostrate growth, e.g., S. arctica or Saxifraga oppositifolia. Common erect dwarf
shrubs with a height of ~0.5 meter is Betula nana and Empetrum nigrum ssp. hermaphroditum.

Shrub increase, or ’shrubification’, have been observed in a number of different ways. First,
infilling of shrubs in open patches patches or a densification of shrubs in a patch. Second, there
can be a change of stature in existing shrub communities where shrubs grow taller or the
community changes towards a larger proportion of taller shrubs. Third, shrubs could expand
beyond their current range, either in elevational gradients in mountainous areas or towards higher
latitudes. Modelling have found Arctic shrubs to be a slight sink at present with a potential to grow
to a larger sunk until the mid of this century (Zhang et al. 2013).

It has been suggested that a shift from graminoid tundra to a more shrub dominated tundra could
amplify the terrestrial biogeophysical feedbacks (Chapin et al. 2005). This would mainly be the
case during snowmelt as a taller vegetation would faster protrude through the snow and thus
lower the albedo. This has been shown to be the case in modelling studies, although by coupling
a regional climate model to a dynamic vegetation model Zhang et al. (2014) found that this would
also enhance the carbon uptake.

An increased shrub cover could potentially have an implication on the nitrogen cycling through the
snow-shrub feedback (Sturm et al. 2001; Sturm 2005). In winter, snow would be trapped by the
shrubs and thus not be compacted by wind in the same extent as snow on the shrub free tundra.
This in turn would increase the soil temperatures as a less dense snow cover have a lower heat
conductivity, thus insulating the ground. In summer however, shrubs would provide a larger shade
from the sun and thus reducing the ground temperature. Studies have shown that soils under
shrubs can be ~2°C cooler than surrounding soils in summer and ~5°C warmer in winter (Myers-
Smith et al. 2011). This could then speed up temperature dependent nitrogen mineralisation rates.
A greater rate of mineralisation would provide nitrogen limited shrubs with more nutrients, thus
enhancing the feedback.

Herbivory and grazing by reindeers and ungulate browsers have the possibility to offset
shrubification (Myers-Smith et al. 2011; Christie et al. 2015). Exclosure experiments have shown
an increase in height and abundance of vascular plants where deciduous plants are increasing as
reindeers are excluded (Vowles et al. 2017). This has then led to the hypothesis that herbivory and
grazing could change competition balance between evergreens and deciduous. A held back
expansion of taller shrubs through herbivory could offset the albedo reduction, and thus limit the
albedo feedback to the arctic system (te Beest et al. 2016). Furthermore, an evergreen, lower
stature vegetation could potentially slow down mineralisation rates through their recalcitrant litter
(Vowles and Björk 2018).

Aims and hypotheses

The Arctic system is complex and the feedbacks are many. Models provide good tools to
disentangle these complex feedbacks. Although change is pan-arctic, it also happens locally and
affects local residents. Earth system models are good tools to inform stakeholders that act on a
global level, but are in general not detailed enough to provide useful information to local
stakeholders.
2
In this thesis I will further investigate the implications of a warming Arctic and expanding shrubs
and forests. I will research:

* Shrub expansion and its biogeophysical feedbacks.

* Arctic biogeochemistry and carbon budget.

* Implications of nutrient limitations for ecosystems, feedbacks and climate

* The influence of reindeer grazing in limiting shrub expansion at local scale

Methods
My main tool for answering the research questions posed above is the dynamic vegetation model
LPJ-GUESS. However, in order to investigate the feedbacks between vegetation and climate I will
also couple LPJ-GUESS to the regional climate model, RCA, developed at the Rossby Centre.

LPJ-GUESS
The Lund-Potsdam-Jena General Ecosystem Simulator, LPJ-GUESS (Smith, Prentice, and Sykes
2001; Sitch et al. 2003; Smith et al. 2014) is a dynamic vegetation model that is optimised for
regional to global use. The basic unit of the model is a grid cell. Within each grid cell a number of
replicate patched are used to simulate ecosystem carbon and vegetation dynamics on a daily
time step given climate (temperature, precipitation, radiation), atmospheric CO2-concentration,
nitrogen deposition and soil texture. A number of plant functional types, PFTs, represent the
vegetation in a grid cell. These are distinguished through their allometry, phenology, life strategies,
need for nutrients, water and sunshine and bioclimatic limits. Vegetation composition and
dynamics is the emergent outcome of competition for light, water and resources by the simulated
PFTs. Within each patch disturbance in the form of fire can occur. The probability of a fire is
calculated from the amount and flammability of plant litter in each patch and the soil water
content in the uppermost soil layer.

To simulate soil biogeochemical processes LPJ-GUESS uses a CENTURY-based (Parton et al.

1993) soil module. In this soil-module plant leaf and root litter as well as coarse woody debris are
inputs. Each simulation day the plant litter is decomposed and partitioned into different pools of
soil organic matter, SOM. These pools are distinguished mainly by their turnover rates. A fixed
fraction of the carbon that is transferred from one pool to another is lost as CO2 to the
atmosphere.

LPJ-GUESS also have the ability to simulate land use and land-use change (Lindeskog et al.
2013). This is done through the division of each grid cell into stands of a particular land cover
type, e.g., pasture, cropland, forest, peatland or natural. Managed stands only have one patch
while peatland and natural stands have a variable number of replicate patches specified by the
user.

RCA-GUESS
Plant and ecosystem functioning and composition is strongly dependent on the climate. In turn
the exchange of carbon and water between the land surface and atmosphere in turn affect the
climate patterns (Pitman 2003). To further investigate these feedbacks I will couple LPJ-GUESS to
the Rossby Centre Regional Climate model, RCA4 (Smith et al. 2011). This coupling has been
used before in the Arctic by Miller and Smith (2012) and Zhang et al (2014), over Europe by
Wramneby et al. (2010), and over the tropics by Wu et al. (2016). These schemes however have
only implemented couplings between older versions of LPJ-GUESS where nitrogen feedbacks are
not included. These feedbacks are of great importance in the high latitudes (Wårlind et al. 2014).
3
 Table 1 - Courses
Name Credits Completed/planned
(ECTS)

Introduction Course for new PhD Students at the Faculty if Science 0,5 Yes

Scientific Writing - How to structure and publish a scientific paper 1,5 Yes

Introduction for PhD students, Dept of Geology Lund University 3 Yes

General Physical Geography 4,5 Yes

Greenhouse Gases Biogeochemistry and Measuring techniques in 5 Yes

Ecosystem Landscapes

Biogenic Volatiles - Exchange at Different Scales and interactions 4 Yes

with Ecosystem Processes

Efficient Programming of Modern HPC strucures 7,5 Almost…

Reproducible and Interactive Data Analysis and Modeling using 4 Yes

Jupyter Notebooks

International Arctic Field School, Universite Laval, Canada 6 Yes

Bayesian Analysis and Decision Theory 5 Almost…

Teaching and Learning in Higher Education 3 Yes

Plant Ecological Physiology 3 Yes

Environmental challenges and the role of environmental science: past 3 Planned for spring 2019
and present

Science communication and role in society 3 Planned for autumn

2019

Research ethics 2 Planned for spring 2019

Interdisciplinarity in environmental research 3 Planned for autumn

2019

Arctic in a Changing climate 5 Spring 2019

First Steps in Biosphere-Atmosphere Modelling 5 Summer 2019

Introductory project work 8 Winter 2019

ClimBEco research school 2 Finished

Sum of ECTS 78

The coupling between LPJ-GUESS and RCA4 is done through the Land surface scheme (LSS) of
the atmospheric model (Samuelsson et al. 2015). The LSS of RCA4 consists of 1-3 tiles
determined through different land-use classes and the presence of snow. These vary in size but
open land is always present with at least 1% of the grid area. The open land tile consists of a
varying fraction of herbaceous plants while the forest tile is divided into broadleaved and
needleleaved forest. LPJ-GUESS required as described above daily surface temperature,
precipitation and incoming radiation, which is provided by the atmospheric model each simulation
day. Albedo and LAI are then calculated within the vegetation model and then reported to the
LSS. The fractional covers of the two forest types and open land is updated annually. This
provides a fully coupled biogeophysical feedback-loop. Biogeochemical processes such as

4
carbon uptake by the plants consists in an ’open loop’ and are not reported back to the
atmospheric model.

Progress
During my PhD-studies I have done mostly model development to this point. This also includes a
big part of evaluation of the new implementations. In addition to the planned papers of my PhD I
have also joined a few side projects that are described below. Regarding courses I have at this
point completed 47 ECTS out of the required 60 ECTS. The last courses are planned to be
performed during 2019 and I estimate to have 18 ECTS more than required when finished. The
courses I have planned or completed are listed in table 1.

I have attended several conferences and workshops, some where I have held presentations on my
work. Conferences that I have continuously visited are the spring and/or autumn meetings of both
the strategic research areas BECC and MERGE. In addition I have been at attending two
conferences of the polar CORDEX network that I am a part of. On both these two conferences I
have presented my future and/or current work. One study that I have presented is described
below under side projects. I have also visited the SOMPA workshop in Helsinki where I presented
our recent work on the methane model.

Below I will briefly describe the new processes that we have implemented in LPJ-GUESS during
my PhD. I will then describe the progress that I have made on each paper under the publication
plan.

Model development
During my PhD I have aided in the development of LPJ-GUESS. Currently there is a new version
of the model about to be released, version 4.1. This will among a lot of other things include new
features that were previously only a part of the branch Wetland Hydrology and Methane, LPJ-
GUESS-WHyMe. These will now become a standard part of the model and although they have
been implemented before, a lot of new coding and testing was needed to meet the coding-
standards of the main version of the model. The new features could be divided into three parts i)
soil physics and permafrost ii) a new methane module and iii) a set of Arctic PFTs.

Soil physics and permafrost

The first improvement of already existing features in LPJ-GUESS consists of new representations
of physical soil processes in the model. The old soil scheme for LPJ-GUESS consists of two soil
layers that have a total depth of 1.5 meters. Soil temperature is calculated for the top soil layer
and is assumed to be representative of the whole soil column. In the new soil scheme adopted
from Wania et al. (2009a) we have divided the soil column into an array of 15 soil layers. Each soil
layer have its own temperature and water content. The temperature in each layer is simulated by
solving the heat diffusion equation,

( ∂z )
∂T ∂ ∂T
= D(z)
∂t ∂z
using a Crank-Nicolson finite difference scheme (Wania, Ross, and Prentice 2009). In this
equation T represents soil temperature, t time, z soil depth and D the thermal diffusivity of the soil
layer. Water in the layer is allowed to freeze if the temperature drops below zero. The water or ice
content of the soil layer changes the heat conductivity of the soil layer and thus have an effect on
the temperature. Other soil properties that affects the thermal diffusivity, e.g., soil texture and
carbon content, is shared for all soil layers. This new routine that now will become a standard part
of LPJ-GUESS version 4.1 enables us to study the impact of climate change on permafrost.

5
 Figure 1. Artificially
drained or flooded
peatlands at Degerö.
Numbers on x-axis
represent mean summer
water table position
(above) and amount of
water removed or added
each day in mm (below).
Methane production is
enhanced when the
peatland is flooded due
to lower oxidation rates
and a larger part of plant
transportation from
lower soil layers to the
atmosphere.

We have evaluated this new soil scheme against site data of soil temperatures across the Arctic.
One thing that proved very important for the winter soil temperatures was the snow thickness and
the implementation of a multi-layered soil scheme. This highlights the importance of snow for the
Arctic soil processes. Although the model is currently developed to include a few new snow
processes such as compaction, this work will not be a part of my PhD thesis.

Methane module
The methane module was developed first by Wania et al (2010). This module includes a new
hydrology scheme than that for Gerten et al. (2004). This hydrology scheme is particular to
peatland stands and assume that soils are organic peat soils. The soil column is further divided
into an acrotelm (sequentially inundated) and a catotelm (permanently inundated). The acrotelm
thus have a fluctuating water table that is below the surface when it has negative values and
above the surface when it has positive values. The water table position also determines the
amount of oxygen/air that is in the soil.

In peatlands the decomposed carbon that is produced in the CENTURY-based soil module each
simulation day is transferred to the methane module. In this module the substrates are partitioned
to either CO2 or CH4. While the CO2 escapes directly to the atmosphere a fraction of the CH4 is
first oxidized and turned into CO2. The oxygen diffusion into the soil is solved with the same
numerical solver that is used for temperature, however using diffusivities for gas instead of heat.
The amount of CH4 being oxidized in each soil layer depends on the oxygenation level in that
layer. If the level of oxygenation in the layer is sufficient to satisfy the stoichiometric relationship,

CH4 + 2O2 → CO2 + 2H2O ,

the CH4 is converted to CO2. The fraction that is not oxidized can reach the atmosphere through
three different pathways. Firstly, CH4 can be transported by vascular plants with aerenchyma, very
porous root tissue which is an adaptation in vascular plants that thrive in poorly oxygenated
habitats (Larcher 2003). This transport is then driven mostly by the abundance and size of
aerenchymatous plants, their phenology and rooting depth (Wania, Ross, and Prentice 2010).

The second pathway that CH4 can escape to the atmosphere is through diffusion through the soil
column. This diffusion is driven by the gradient between the soil layer and the atmosphere and
solved with the same numerical solver as for the heat and oxygen diffusion. A maximum quantity
6
of CH4 is allowed in the soil layer. This limit is set by a temperature dependent solubility of CH4 in
water. If the concentration of CH4 in a layer rises above this limit the third and final pathway is
activated, ebullition, where CH4 escapes directly to the atmosphere in the form of bubbles.

The amount of CH4 produced and released to the atmosphere is highly dependent on the type of
vegetation inhabiting the peatland. We have implemented a switch in the model where we can
artificially remove or add a fixed amount of water to the peatland each day. This allows for testing
of draining or flooding a peatland in the model. Figure 1 shows a test that has been done where
we artificially remove or add water from a site in Degerö. As the water table is lowered, a larger
number of shrubs starts to populate the peatland, while only mosses survives when the peatland
is flooded. The relative importance of each pathway is also shown and as the water table
increases, so does the plant transport.

A set of PFTs that are allowed to grow on peatlands have been created for this purpose. The
different PFTs are separated by their drought tolerance, where woody plants have a lower
tolerance than herbaceous plants and mosses. Some PFTs also have aerenchyma which allows
them to transport CH4 from the lower soil layers to the atmosphere. A peatland moss is also
added to the set of wetland PFTs. This PFT is representing the commonly spread Sphagnum spp.
mosses in the model.

Arctic PFTs
We have besides the new peatland forms of some PFTs also adopted a new set of arctic PFTs,
first used by Wolf et al (2007). These PFTs, with some modifications, have also been used by
Miller and Smith (2012), Zhang et al., (2013) and Tang et al. (2015, 2016). They have now been
adapted to become a standard part of the Arctic simulations with LPJ-GUESS. These include a
number of shrub functional types, one new Arctic grass/forb, and a PFT to represent low stature
tundra species and lichens. Below I will describe the shrub PFTs a bit closer.

Firstly we have evergreen and deciduous prostrate dwarf shrubs. These are very low stature
woody PFTs with a lateral growth along the ground surface. Role model species for the
evergreens could be Saxifraga oppositifolia, Dryas Octopetala or Cassiope tetragona, for the
summer greens this could be Salix arctica, Arctostaphylos alpinus and Salix reticulata. They are
not particularly strong competitors when it comes to light but thrive in the open lands without
higher vegetation. To survive these PFTs require a small amount of insulating snow cover.

Secondly we have erect dwarf shrubs, both evergreen and deciduous. These represent the
important species Empetrum nigrum ssp. hermaphroditum and other Erica spp. (evergreen) as
well as Betula nana, Vaccinium myrtillus, V. uliginosum (deciduous). These thrive in the field layer
of the tundra and are likely very important for the carbon cycling in the Arctic.

Lastly we have the tall shrubs like Juniperus communis (evergeen) and Salix spp. (deciduous).
These could potentially have a larger effect on albedo reduction as they could protrude through
the snow cover in winter and spring.

Side projects
Here follows a short and non-exhaustive description of the side projects that I have undertaken
during my PhD and a plan for how to proceed with them.

The Analysis of the CENTURY

The first side project we humbly call ’the analysis of the CENTURY’. It is a cooperation between
the LPJ-GUESS developers group and the Microbial Ecology group lead by Johannes Rousk at
the ecology department of Lund University. Johannes and his team (where I also did my master
7
thesis) specializes in measuring microbial activity in soils using stable isotopes. They have
collected soils from a large number of sites with different temperature regimes. From this they
have made a model of the temperature adaptation of microbial communities. The most important
thing with this is that the carbon use efficiency, CUE, varies with the climate (Berglund/Rousk et
al., in prep). As the CUE in the CENTURY model is a fixed ratio that does not vary this new feature
will have implications for the carbon sequestration potential of soils. We expect this new feature in
the model to lead to larger simulated carbon pools in the Arctic. These carbon pools might
however be more sensitive to warming as the microbial community will change.

This project has been ongoing for some time and is currently paused. My part of this project is to
conduct site level evaluation at Harvard forest from where we have good evaluation data. I have
also been part in the planning of the project from the beginning and will do some small part of the
writing as well.

Local methane emissions, land use change and policy implications

The second side project is my participation within a BECC Action group. In this project we will try
to model the site Skogaryd (58º23’N, 12º09’E). This research station is situated on a drained
peatland with planted spruce forest. The forest was planted in 1952 and will be cut in 2019. A
group of three models, LPJ-GUESS, Coup (Jansson and Moon 2001), and ForSAFE (Wallman et
al. 2005), have been invited to simulate this succession. This modeling experiment was done for a
historic period of the years 1951-2011 by He et al. (2016). The project will help us develop local
methane processes connected to land use and land use change for our models. Historically,
draining peatlands was an important way of gaining agricultural land or increasing productivity in
forests. With concerns of climate change these peatlands have been put under study as they
could potentially be large sources of greenhouse gases, both CO2 and CH4. The question is what
the best action from a climate perspective would be, should we restore these present forests back
to wetlands again when the biomass is harvested or should we replant the forest? Do we rist that
these wetlands start to leach DOC when we harvest? Policy relevant questions like this could be
answered with our models.

The action group will soon be dissolved but the project will most likely continue in one way or the
other. How is still unclear. There will however be at least one paper released from the action group
which I will co-author. My part in that paper (which to a large extent is already done) is to
implement land-use management on peatlands in LPJ-GUESS, conduct the simulations for the
modeling experiment and analyze the results.

Robustness of the climate feedback from dynamic vegetation in regional climate models
As a third side project, I was invited by Heidrun Matthes and Annette Rinke from the Alfred
Wegener Institute, AWI, to lead a paper on robustness of the climate signal from coupled dynamic
vegetation and climate models. Most climate models have a static land-surface. Static here
means that vegetation does not migrate or establish in new places with a changed climate. Some
climate models solve this problem by reading in and updating a map. That is for instance the case
with both RCA and . The purpose was to compare the climate signal from a land surface with
dynamic vegetation calculated for a variable as:

Va rproj,static − Va rproj,dynamic
sig n a l =
Va rhist,static − Va rhist,dynamic

The variables that we intended to investigate were latent and sensible heat flux, surface air
temperature, precipitation, atmospheric pressure and so on. I was going to use the data from

8
Zhang et al., 2013 and they also had some already finished simulations. The work with this was
started and I presented some of the preliminary results on the polar CORDEX meeting in
Cambridge in October, 2017. I later decided to postpone this paper since we lacked data from the
historic RCA simulation with static vegetation. If I have time towards the end of the PhD-project I
might take up this paper again since it is a quite interesting topic.

Publication plan
Following is the publication plan and a short introduction to all the papers. Paper 1 is the currently
in progress and will thus be described more in detail.

Paper 1 - Modeling of recent and future impacts of climate change on

ecosystem functioning and stability in the Swedish sub-arctic
This paper is what I am currently working on. Thus the analysis and evaluation is still in the
making will not be exhaustive. The results are only preliminary. The simulations for the historic
period are more or less finished. We will add simulations for projected climate. However, I feel that
there is a large uncertainty in the PFT parameters, especially for parameters affecting the
competition for nitrogen. Wramneby et al. (2008) performed a nice analysis of parameter
uncertainties in LPJ-GUESS, In this study she did not find large impacts of parameter uncertainty
for the balance between evergreen and deciduous plants or shade tolerant/intolerant for
ecosystem functioning. If one plant was declining another took over and NEE would not change
drastically. However, this was before the implementation of the nitrogen cycle in the model (Smith
et al. 2014). The competition between PFTs parameterised to global averages on the standard
LPJ-GUESS grid size of 0.5°x0.5° seem to be stable. However, adding new PFTs and focusing on
cold environments only, the ecosystem functioning have proven very sensitive to the
parameterisation of the PFTs.

Motivation and hypotheses

The Arctic is changing on all scales, regional (Walker et al. 2006), landscape (Van Bogaert et al.
2011) and on the community levels (Bjorkman et al. 2018). Changes in traits of individual plants
and communities have been observed over the whole Arctic. Although change on the regional and
global scales will affect local residents of Arctic communities (Arctic resilience report?) few studies
of local climate impacts have been made. Models of climate induced vegetation change on the
landscape levels would be helpful for local stakeholders in adapting to a future climate (AMAP
2017).

Methods
For this study we used LPJ-GUESS over the Abisko region. The model was forced with finely
gridded (50mx50m) climate data between the years 1913 to 2000 (Yang, Hanna, and Callaghan
2011). This dataset was produced from a local climate model, TopoModel. Within the model
elevational effects of temperature (lapse rate) was taken into account as were local effects on
temperature such as shading of the mountain. The model was constructed and parameterised
from measurements of temperature along elevational transects in the Torneträsk region. The
distance of each grid cell to lake Torneträsk was also found to have an effect on local temperature
and was included in this local climate model.

Nitrogen wet and dry deposition was derived from the standard LPJ-GUESS nitrogen input data
(Lamarque et al. 2013) from the grid cell in the dataset closest to Abisko. Soil parameters required

9
Figure 2. Updated LPJ-GUESS phenology parameter for heat requirement (Growing degree
days, GDD) for mountain birch to reach full leaf-out. Original implementation used GDD5 as heat
requirements while the new calibrated implementation uses GDD0. Calibration was made against
MODIS derived NDVI (right axis).

by LPJ-GUESS were derived from the closest grid point in the WISE soil data set (Batjes 2005) on
a 0.5°x0.5° resolution.

A subset of the standard LPJ-GUESS PFTs were used for this study plus seven additional Arctic
PFTs not used in the standard setup of LPJ-GUESS, in total 11 PFTs were used (Jag bore ha en
table här…). The shade-intolerant boreal summer green (IBS) functional type usually represents
Betula spp. in global simulations and have been parameterised for early-successional forests in
the transition between temperate and boreal climates. In the Torneträsk region however, Betula
spp. are generally the tallest species. This makes re-parameterisation of this functional group
necessary.

As a basis for this study, PFT parameters from Wolf et al. (2010) were used. These mainly differ
from the standard model settings in the allometry with a shift to shorter vegetation with a lower
leaf to sapwood ratio. Wramneby et al. (2008) found that the two most important parameter for
the competitive balance between shade tolerant/intolerant PFTs was the turnover of sapwood and
minimum light requirement on the forest floor for establishment. A higher number of the former
parameter would shift the competition in favour of shade-tolerant functional types. A study by
Karlsson et al. (2001) found this parameter to be 1% per year for mountain birch. This parameter
is set to 5% for shade-tolerant types in standard LPJ-GUESS settings. The updated value for this
parameter would thus shift the competitive balance for IBS to become a late successional
species, which would not be an unreasonable assumption in this region. Karlsson et al. (2001)
also found turnover of fine roots to range between 60-80% per year which agrees well with the
standard formulation for trees in LPJ-GUESS at 70% per year.

A common way to model the heat requirement for different phenological events is through
accumulation of growing degree days, GDD. To model the heat requirements to go from zero to
full leaf cover, LPJ-GUESS in its standard formulation use a PFT-specific parameter of
accumulated GDD5 (days with mean temperature above 5 degrees). This parameter was evaluated
against MODIS derived Normalised difference vegetation index (NDVI) over the mountain birch
forest. It was found that the model was sensitive to this parameter and that it did not capture the
10
Figure 3. The extent and carbon mass of IBS (mountain birch) in the Abisko region. Since the
beginning of the century mountain birch forests have become denser and more abundant higher
up on the mountain sides.

timing of full leaf cover well (Figure 2). Plants growing in cold environments with short growing
seasons will likely be adapted to these environments (Larcher 2003). These plants could
commence bud break as soon as temperatures go above zero. We thus changed the model
formulation of heat requirements of the PFTs to require GDD0 instead of GDD5. This yielded a
spring phenology more in agreement with observed data (Figure 2)

In a similar way that arctic plants have heat requirements to reach full leaf cover, they also have
heat requirements for establishment. Wolf et al. (2010) introduced a minimum limit of GDD0 for a
PFT to establish. These limits were also utilised in this study.

Analysis
We compared 10-year averages between the years 1915 and 1925 (referred to as 1920) and
between the years 1985 to 1995 (referred to as 1990) for the historic period. Variables investigated
were the biomass, height, density and leaf area index (LAI) for each PFT functional groups of
evergreens and deciduous plants, and the community as a whole.

Preliminary results
From the beginning of the century IBS have significantly increased its biomass from 1.09 kgC m-2
for the years around 1920 to 1.42 kgC m-2 for the years around 1990. This increase is due to a
densification of the birch forest (Figure 3). These results are in line with Callaghan et al. (2013) who
also found that birch forests in Abisko have become denser. Van Bogaert et al. (2011) found birch
forests to have extended their elevational range at several sites in the Torneträsk region.

11
GPP in the mountain birch forest increased from approx 118 gC m-2 yr-1 around 1920 to 155 gC
m-2 yr-1. This is a large underestimation compared to Olsson et al. (2017) who used eddy
covariance data to estimate GPP in the birch forest. They found values of 440 gC m-2 yr-1 for five
years between 2007 and 2014 (Olsson et al. 2017). These were year without outbreaks of
defoliating moths.

This analysis is as can be seen not finished. I will further evaluate the results against observed
values from e.g., Callaghan et al. and Van Bogaert et al. 2011. There are other studies as well that
could aid in the evaluation of for instance shrub increase. Furthermore I will evaluate the
elevational increase IBS abundance and link it to temperature (winter and summer), nitrogen and
water availability.

Paper 2 - Implications of nitrogen cycling on coupled vegetation and climate

for a warming Arctic
In this paper I will investigate the the implications of the nitrogen feedbacks and shrub expansion
within RCA-GUESS. It will be a comparison with the paper by Zhang et al. (2014) where nitrogen
was not included. This paper found large reductions in albedo from a changed vegetation which
further amplified the warming, especially during spring (Zhang et al. 2014).

In preparation for this paper I will need to rebuild the interface between RCA and LPJ-GUESS.
The previous version on RCA-GUESS builds on an old code structure of LPJ-GUESS and does
not include nitrogen cycling or the current land use change scheme introduced by Lindeskog et
al. (2013). To facilitate maintenance and future development of the model, the interface will need
to be updated to match the new code structure. Luckily, LPJ-GUESS have been coupled to other
atmospheric models, e.g., in EC-EARTH (Alessandri et al. 2016). I will draw from this experience
when remaking the interface. Furthermore there is a group of developers forming around this
problem, as more climate models are being coupled to LPJ-GUESS. I will be a part of that group.

Hypotheses of this paper will be that shrubification of the Arctic drive biogeophysical feedbacks
mainly in the form of reductions of albedo. However, by not including shrub and other specialised
Arctic PFTs, previous modelling studies have underestimated ecosystem productivity. Thus the
biogeochemical feedback might also be stronger than previously thought.

The data needed for this study is first of all forcing data for the boundary conditions of the
regional climate mode. This will come from the new version of the earth system model EC-EARTH
where LPJ-GUESS constitutes the dynamic vegetation submodel. Furthermore we will need CO2
concentration and nitrogen deposition for the future climate scenarios as as inputs.

To evaluate the results for the historic period the Circumpolar Arctic Vegetation Map (Walker et al.
2005) will be very useful. We could also use MODIS derived NDVI to evaluate phenological
responses to warming.

Paper 3 - The influence of reindeers grazing on local and regional vegetation

shifts in the Arctic
In this paper I will implement a probabilistic grazing model to use with LPJ-GUESS in the Abisko
region. We will use the same small-scale climate data used in the previous Abisko-study and the
same region. This time however the focus will be on how large ungulate browsers shape the
ecosystem, e.g., treelike advance, and carbon balance at the local scale and in northern Sweden
where grazing pressure could be considered high. Grazing have been suggested to alter the
competitive balance between deciduous and evergreen plants through the reindeer preference

12
(Vowles et al. 2017). As evergreen plants have more recalcitrant leaf litter, this could have
implications for the carbon cycling.

Reindeer grazing is a continuous disturbance to the plants. Reindeer distribution within the
landscape will be modelled probabilistically. The probability of biomass loss of a PFT will be
dependent on the density of reindeers in the area at the time, the reindeer need for energy and the
preference of that PFT. This would require some literature search and possible expert opinions to
parameterise the preferences of reindeers of certain plants. The conceptual idea of the new
grazing module is still under construction and will be deal more with when the time comes.

The paper will not only implement a new grazing module but also investigate how grazing by
reindeers shape the forest-tundra ecotone and the balance between evergreens and deciduous
plants.

Paper 4 - Integration of permafrost and peat soil dynamics in regional dynamic

vegetation model for Arctic environments
In this paper we will explore the pan-arctic implications of permafrost degradation and methane
dynamics in the Arctic. As the climate is becoming warmer permafrost starts to thaw. This will
bring further vegetation change as the environment becomes more suitable for taller plants. It has
been hypothesised that plant community change could drive changes in CH4 emissions as plants
with larger leaves will transpire more water and thus have the potential to dry out soils (Merbold et
al. 2009). A larger number of shrubs on peatlands could also limit CH4 emissions as these do not
have aerenchyma that transport CH4 from lower soil layers to the atmosphere. On the other hand,
more productive ecosystems, a thawing tundra and plant community change could lead to more
CH4 emissions. This paper will delve deeper into those feedbacks. It will also contain a
description of the processes now implemented in the standard version of LPJ-GUESS. A few
preliminary results for the site-level evaluation of the methane model can be seen in figure 4 and
for the simulated permafrost in figure 5.

Plan for the remaining work

Below is a plan for when the manuscripts are to be submitted.

Paper 1
I have already done simulations and model setup for the historic period of this paper and begun
the writing process. Depending on how novel we find the results in this study I might also do
some simulations with future projections of this area. I expect the manuscript to be ready and
handed to co-authors for review in March 2019.

Paper 2
Technical work and model setup can be commenced during spring 2019. I would like to present
(preliminary) results at the Arctic CORDEX conference in Copenhagen October 2019. My goal is to
have the manuscript ready by October-November 2019.

Paper 3
I will start the development of the reindeer module after paper 2 is finished or if time should arise
in between model runs of paper 2. I expect this technical development to take 1-2 months. My
goal is to have the manuscript ready in February 2020

13
Figure 4. Site level evaluation of the CH4 scheme adopted from Wania et al. (2010).

A)

Figure 5. Simulated permafrost

B) area in LPJ-GUESS. A) global
permafrost extent and active layer
thickness down to 1.5 meter. B)
Simulated permafrost extent
between the years 1900 to 2015.

14
Paper 4
This paper will be co-authored with Paul Miller with Paul as the lead author. The branch is ready
for integration to trunk and will be so during spring. I will help with the simulations and analysis of
this paper and perhaps some writing. I expect the manuscript to be submitted during 2019.

15
References
Alessandri, Andrea, Franco Catalano, Matteo De Felice, Bart Van Den Hurk, Francisco Doblas
Reyes, Souhail Boussetta, Gianpaolo Balsamo, and Paul A. Miller. 2016. 'Multi-scale
enhancement of climate prediction over land by increasing the model sensitivity to vegetation
variability in EC-Earth', Climate Dynamics, 49: 1215-37.

AMAP. 2017. Snow, Water, Ice and Permafrost in the Arctic (Arctic Monitoring and Assessment
Programme (AMAP): Oslo, Norway).

Batjes, N. H. 2005. "ISRIC-WISE global data ser of derived soil properties on a 0.5 by 0.5 degree
grid (version 3.0)." In.: ISRIC – World Soil Information, Wageningen.

Bintanja, R., and O. Andry. 2017. 'Towards a rain-dominated Arctic', Nature Climate Change, 7:
263-67.

Bjorkman, A. D., I. H. Myers-Smith, S. C. Elmendorf, S. Normand, N. Ruger, P. S. A. Beck, A.

Blach-Overgaard, D. Blok, J. H. C. Cornelissen, B. C. Forbes, D. Georges, S. J. Goetz, K. C. Guay,
G. H. R. Henry, J. HilleRisLambers, R. D. Hollister, D. N. Karger, J. Kattge, P. Manning, J. S.
Prevey, C. Rixen, G. Schaepman-Strub, H. J. D. Thomas, M. Vellend, M. Wilmking, S. Wipf, M.
Carbognani, L. Hermanutz, E. Levesque, U. Molau, A. Petraglia, N. A. Soudzilovskaia, M. J.
Spasojevic, M. Tomaselli, T. Vowles, J. M. Alatalo, H. D. Alexander, A. Anadon-Rosell, S. Angers-
Blondin, M. T. Beest, L. Berner, R. G. Bjork, A. Buchwal, A. Buras, K. Christie, E. J. Cooper, S.
Dullinger, B. Elberling, A. Eskelinen, E. R. Frei, O. Grau, P. Grogan, M. Hallinger, K. A. Harper,
Mmpd Heijmans, J. Hudson, K. Hulber, M. Iturrate-Garcia, C. M. Iversen, F. Jaroszynska, J. F.
Johnstone, R. H. Jorgensen, E. Kaarlejarvi, R. Klady, S. Kuleza, A. Kulonen, L. J. Lamarque, T.
Lantz, C. J. Little, J. D. M. Speed, A. Michelsen, A. Milbau, J. Nabe-Nielsen, S. S. Nielsen, J. M.
Ninot, S. F. Oberbauer, J. Olofsson, V. G. Onipchenko, S. B. Rumpf, P. Semenchuk, R. Shetti, L. S.
Collier, L. E. Street, K. N. Suding, K. D. Tape, A. Trant, U. A. Treier, J. P. Tremblay, M. Tremblay, S.
Venn, S. Weijers, T. Zamin, N. Boulanger-Lapointe, W. A. Gould, D. S. Hik, A. Hofgaard, I. S.
Jonsdottir, J. Jorgenson, J. Klein, B. Magnusson, C. Tweedie, P. A. Wookey, M. Bahn, B. Blonder,
P. M. van Bodegom, B. Bond-Lamberty, G. Campetella, B. E. L. Cerabolini, F. S. Chapin, 3rd, W.
K. Cornwell, J. Craine, M. Dainese, F. T. de Vries, S. Diaz, B. J. Enquist, W. Green, R. Milla, U.
Niinemets, Y. Onoda, J. C. Ordonez, W. A. Ozinga, J. Penuelas, H. Poorter, P. Poschlod, P. B.
Reich, B. Sandel, B. Schamp, S. Sheremetev, and E. Weiher. 2018. 'Plant functional trait change
across a warming tundra biome', Nature, 562: 57-62.

Callaghan, T. V., C. Jonasson, T. Thierfelder, Z. Yang, H. Hedenas, M. Johansson, U. Molau, R.

Van Bogaert, A. Michelsen, J. Olofsson, D. Gwynn-Jones, S. Bokhorst, G. Phoenix, J. W. Bjerke,
H. Tommervik, T. R. Christensen, E. Hanna, E. K. Koller, and V. L. Sloan. 2013. 'Ecosystem change
and stability over multiple decades in the Swedish subarctic: complex processes and multiple
drivers', Philos Trans R Soc Lond B Biol Sci, 368: 20120488.

Chapin, F. S., 3rd, M. Sturm, M. C. Serreze, J. P. McFadden, J. R. Key, A. H. Lloyd, A. D. McGuire,

T. S. Rupp, A. H. Lynch, J. P. Schimel, J. Beringer, W. L. Chapman, H. E. Epstein, E. S. Euskirchen,
L. D. Hinzman, G. Jia, C. L. Ping, K. D. Tape, C. D. Thompson, D. A. Walker, and J. M. Welker.
2005. 'Role of land-surface changes in arctic summer warming', Science, 310: 657-60.

Christie, Katie S., John P. Bryant, Laura Gough, Virve T. Ravolainen, Roger W. Ruess, and Ken D.
Tape. 2015. 'The Role of Vertebrate Herbivores in Regulating Shrub Expansion in the Arctic: A
Synthesis', BioScience.

Elmendorf, Sarah C., Gregory H. R. Henry, Robert D. Hollister, Robert G. Björk, Noémie
Boulanger-Lapointe, Elisabeth J. Cooper, Johannes H. C. Cornelissen, Thomas A. Day, Ellen
16
Dorrepaal, Tatiana G. Elumeeva, Mike Gill, William A. Gould, John Harte, David S. Hik, Annika
Hofgaard, David R. Johnson, Jill F. Johnstone, Ingibjörg Svala Jónsdóttir, Janet C. Jorgenson, Kari
Klanderud, Julia A. Klein, Saewan Koh, Gaku Kudo, Mark Lara, Esther Lévesque, Borgthor
Magnússon, Jeremy L. May, Joel A. Mercado-Dı´az, Anders Michelsen, Ulf Molau, Isla H. Myers-
Smith, Steven F. Oberbauer, Vladimir G. Onipchenko, Christian Rixen, Niels Martin Schmidt, Gaius
R. Shaver, Marko J. Spasojevic, Þóra Ellen Þórhallsdóttir, Anne Tolvanen, Tiffany Troxler, Craig E.
Tweedie, Sandra Villareal, Carl-Henrik Wahren, Xanthe Walker, Patrick J. Webber, Jeffrey M.
Welker, and Sonja Wipf. 2012. 'Plot-scale evidence of tundra vegetation change and links to
recent summer warming', Nature Climate Change, 2: 453-57.

Gerten, Dieter, Sibyll Schaphoff, Uwe Haberlandt, Wolfgang Lucht, and Stephen Sitch. 2004.
'Terrestrial vegetation and water balance—hydrological evaluation of a dynamic global vegetation
model', Journal of Hydrology, 286: 249-70.

He, Hongxing, Astrid Meyer, Per-Erik Jansson, Magnus Svensson, Tobias Rütting, and Leif
Klemedtsson. 2018. 'Simulating ectomycorrhiza in boreal forests: implementing ectomycorrhizal
fungi model MYCOFON in CoupModel (v5)', Geoscientific Model Development, 11: 725-51.

Hugelius, G., J. Strauss, S. Zubrzycki, J. W. Harden, E. A. G. Schuur, C. L. Ping, L. Schirrmeister,

G. Grosse, G. J. Michaelson, C. D. Koven, amp, apos, J. A. Donnell, B. Elberling, U. Mishra, P.
Camill, Z. Yu, J. Palmtag, and P. Kuhry. 2014. 'Estimated stocks of circumpolar permafrost carbon
with quantified uncertainty ranges and identified data gaps', Biogeosciences, 11: 6573-93.

IPCC. 2014. "Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III
to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change." In, edited by
Core Writing Team, R.K. Pachauri and L.A. Meyer. Geneva, Switzerland: IPCC.

Jansson, Per-Erik, and David S. Moon. 2001. 'A coupled model of water, heat and mass transfer
using object orientation to improve flexibility and functionality', Environmental Modelling &
Software, 16: 37-46.

Karlsson, P. S., K.O. Nordell, and B. Sveinbjörnsson. 2001. 'Biomass and Nitrogen Turnover and
Nutrient Use Characteristics in Subarctic Mountain Birch Trees.' in F. E. Wielgolaski (ed.), Nordic
Mountain Birch Ecosystems (United Nations Educational, Scientific and Cultural Organisation
(UNESCO): Oslo, Norway).

Lamarque, J. F., F. Dentener, J. McConnell, C. U. Ro, M. Shaw, R. Vet, D. Bergmann, P. Cameron-

Smith, S. Dalsoren, R. Doherty, G. Faluvegi, S. J. Ghan, B. Josse, Y. H. Lee, I. A. MacKenzie, D.
Plummer, D. T. Shindell, R. B. Skeie, D. S. Stevenson, S. Strode, G. Zeng, M. Curran, D. Dahl-
Jensen, S. Das, D. Fritzsche, and M. Nolan. 2013. 'Multi-model mean nitrogen and sulfur
deposition from the Atmospheric Chemistry and Climate Model Intercomparison Project
(ACCMIP): evaluation of historical and projected future changes', Atmospheric Chemistry and
Physics, 13: 7997-8018.

Larcher, Walter. 2003. Physiological Plant Ecology (Springer-Verlag Berlin and Heidelberg GmbH &
Co. K: Stuttgart).

Lindeskog, M., A. Arneth, A. Bondeau, K. Waha, J. Seaquist, S. Olin, and B. Smith. 2013.
'Implications of accounting for land use in simulations of ecosystem carbon cycling in Africa',
Earth System Dynamics, 4: 385-407.

Merbold, L., W. L. Kutsch, C. Corradi, O. Kolle, C. Rebmann, P. C. Stoy, S. A. Zimov, and E. D.

Schulze. 2009. 'Artificial drainage and associated carbon fluxes (CO2/CH4) in a tundra
ecosystem', Global Change Biology, 15: 2599-614.
17
Miller, P. A., and B. Smith. 2012. 'Modelling tundra vegetation response to recent arctic warming',
Ambio, 41 Suppl 3: 281-91.

Myers-Smith, Isla H., Sarah C. Elmendorf, Pieter S. A. Beck, Martin Wilmking, Martin Hallinger,
Daan Blok, Ken D. Tape, Shelly A. Rayback, Marc Macias-Fauria, Bruce C. Forbes, James D. M.
Speed, Noémie Boulanger-Lapointe, Christian Rixen, Esther Lévesque, Niels Martin Schmidt,
Claudia Baittinger, Andrew J. Trant, Luise Hermanutz, Laura Siegwart Collier, Melissa A. Dawes,
Trevor C. Lantz, Stef Weijers, Rasmus Halfdan Jørgensen, Agata Buchwal, Allan Buras, Adam T.
Naito, Virve Ravolainen, Gabriela Schaepman-Strub, Julia A. Wheeler, Sonja Wipf, Kevin C. Guay,
David S. Hik, and Mark Vellend. 2015. 'Climate sensitivity of shrub growth across the tundra
biome', Nature Climate Change, 5: 887-91.

Myers-Smith, Isla H., Bruce C. Forbes, Martin Wilmking, Martin Hallinger, Trevor Lantz, Daan Blok,
Ken D. Tape, Marc Macias-Fauria, Ute Sass-Klaassen, Esther Lévesque, Stéphane Boudreau,
Pascale Ropars, Luise Hermanutz, Andrew Trant, Laura Siegwart Collier, Stef Weijers, Jelte
Rozema, Shelly A. Rayback, Niels Martin Schmidt, Gabriela Schaepman-Strub, Sonja Wipf,
Christian Rixen, Cécile B. Ménard, Susanna Venn, Scott Goetz, Laia Andreu-Hayles, Sarah
Elmendorf, Virve Ravolainen, Jeffrey Welker, Paul Grogan, Howard E. Epstein, and David S. Hik.
2011. 'Shrub expansion in tundra ecosystems: dynamics, impacts and research priorities',
Environmental Research Letters, 6.

Olsson, Per-Ola, Michal Heliasz, Hongxiao Jin, and Lars Eklundh. 2017. 'Mapping the reduction in
gross primary productivity in subarctic birch forests due to insect outbreaks', Biogeosciences, 14:
1703-19.

Parton, W. J., J. M. O. Scurlock, D. S. Ojima, T. G. Gilmanov, R. J. Scholes, D. S. Schimel, T.

Kirschner, J.-C. Menaut, T. Seastedt, E Garcia Moya, Apinan Kamnalrut, and J. I. Kinyamario.
1993. 'Observations and modeling of biomass and soil organic matter dynamics for the grassland
biome worldwide', Global Biogeochemical Cycles, 7: 785-809.

Pitman, A. J. 2003. 'The evolution of, and revolution in, land surface schemes designed for
climate models', International Journal of Climatology, 23: 479-510.

Samuelsson, Patrick, Stefan Gollvik, Marco Kupiainen, Ekaterina Kourzeneva, and Willem Jan
van de Berg. 2015. "The surface processes of the Rossby Centre regional atmospheric climate
model (RCA4)." In Meteorologi. SMHI.

Serreze, Mark C., and Roger G. Barry. 2011. 'Processes and impacts of Arctic amplification: A
research synthesis', Global and Planetary Change, 77: 85-96.

Sitch, S., B. Smith, I. C. Prentice, A. Arneth, A. Bondeau, W. Cramer, J. O. Kaplan, S. Levis, W.

Lucht, M. T. Sykes, K. Thonicke, and S. Venevsky. 2003. 'Evaluation of ecosystem dynamics,
plant geography and terrestrial carbon cycling in the LPJ dynamic global vegetation model',
Global Change Biology, 9: 161-85.

Smith, B., D. Wårlind, A. Arneth, T. Hickler, P. Leadley, J. Siltberg, and S. Zaehle. 2014.
'Implications of incorporating N cycling and N limitations on primary production in an individual-
based dynamic vegetation model', Biogeosciences, 11: 2027-54.

Smith, Benjamin, I. Colin Prentice, and Martin T. Sykes. 2001. 'Representation of vegetation
dynamics in the modelling of terrestrial ecosystems: comparing two contrasting approaches
within European climate space', Global Ecology and Biogeography, 10: 621-37.

18
Smith, Benjamin, Patrick Samuelsson, Anna Wramneby, and Markku Rummukainen. 2011. 'A
model of the coupled dynamics of climate, vegetation and terrestrial ecosystem biogeochemistry
for regional applications', Tellus A: Dynamic Meteorology and Oceanography, 63: 87-106.

Sturm, Matthew. 2005. 'Changing snow and shrub conditions affect albedo with global
implications', Journal of Geophysical Research, 110.

Sturm, Matthew, Jon Holmgren, Joseph P. McFadden, Glen E. Liston, F. Stuart Chapin, and
Charles H. Racine. 2001. 'Snow–Shrub Interactions in Arctic Tundra: A Hypothesis with Climatic
Implications', Journal of Climate, 14: 336-44.

Tang, J., P. A. Miller, A. Persson, D. Olefeldt, P. Pilesjö, M. Heliasz, M. Jackowicz-Korczynski, Z.

Yang, B. Smith, T. V. Callaghan, and T. R. Christensen. 2015. 'Carbon budget estimation of a
subarctic catchment using a dynamic ecosystem model at high spatial resolution',
Biogeosciences, 12: 2791-808.

Tang, Jing, Guy Schurgers, Hanna Valolahti, Patrick Faubert, Päivi Tiiva, Anders Michelsen, and
Riikka Rinnan. 2016. 'What are the challenges for modelling isoprene and monoterpene emission
dynamics of subarctic plants?', Biogeosciences Discussions: 1-29.

te Beest, Mariska, Judith Sitters, Cécile B. Ménard, and Johan Olofsson. 2016. 'Reindeer grazing
increases summer albedo by reducing shrub abundance in Arctic tundra', Environmental Research
Letters, 11.

Van Bogaert, Rik, Kristof Haneca, Jan Hoogesteger, Christer Jonasson, Morgan De Dapper, and
Terry V. Callaghan. 2011. 'A century of tree line changes in sub-Arctic Sweden shows local and
regional variability and only a minor influence of 20th century climate warming', Journal of
Biogeography, 38: 907-21.

van Huissteden, J., and A. J. Dolman. 2012. 'Soil carbon in the Arctic and the permafrost carbon
feedback', Current Opinion in Environmental Sustainability, 4: 545-51.

Vowles, Tage, and Robert G. Björk. 2018. 'Implications of evergreen shrub expansion in the
Arctic', Journal of Ecology, 107: 650-55.

Vowles, Tage, Cajsa Lovehav, Ulf Molau, and Robert G. Björk. 2017. 'Contrasting impacts of
reindeer grazing in two tundra grasslands', Environmental Research Letters, 12.

Walker, Donald A., Martha K. Raynolds, Fred J. A. Daniëls, Eythor Einarsson, Arve Elvebakk,
William A. Gould, Adrian E. Katenin, Sergei S. Kholod, Carl J. Markon, Evgeny S. Melnikov, Natalia
G. Moskalenko, Stephen S. Talbot, Boris A. † Yurtsev, and Cavm Team The other members of the.
2005. 'The Circumpolar Arctic vegetation map', Journal of Vegetation Science, 16: 267-82.

Walker, M. D., C. H. Wahren, R. D. Hollister, G. H. R. Henry, L. E. Ahlquist, J. M. Alatalo, M. S.

Bret-Harte, M. P. Calef, T. V. Callaghan, A. B. Carroll, H. E. Epstein, I. S. Jonsdottir, J. A. Klein, B.
Magnusson, U. Molau, S. F. Oberbauer, S. P. Rewa, C. H. Robinson, G. R. Shaver, K. N. Suding, C.
C. Thompson, A. Tolvanen, O. Totland, P. L. Turner, C. E. Tweedie, P. J. Webber, and P. A. Wookey.
2006. 'From The Cover: Plant community responses to experimental warming across the tundra
biome', Proceedings of the National Academy of Sciences, 103: 1342-46.

Wallman, Patrik, Mats G. E. Svensson, Harald Sverdrup, and Salim Belyazid. 2005. 'ForSAFE—an
integrated process-oriented forest model for long-term sustainability assessments', Forest
Ecology and Management, 207: 19-36.

19
Wania, R., I. Ross, and I. C. Prentice. 2009a. 'Integrating peatlands and permafrost into a dynamic
global vegetation model: 1. Evaluation and sensitivity of physical land surface processes', Global
Biogeochemical Cycles, 23: n/a-n/a.

———. 2009b. 'Integrating peatlands and permafrost into a dynamic global vegetation model: 2.
Evaluation and sensitivity of vegetation and carbon cycle processes', Global Biogeochemical
Cycles, 23: n/a-n/a.

———. 2010. 'Implementation and evaluation of a new methane model within a dynamic global
vegetation model: LPJ-WHyMe v1.3.1', Geoscientific Model Development, 3: 565-84.

Wolf, Annett, Terry V. Callaghan, and Karin Larson. 2007. 'Future changes in vegetation and
ecosystem function of the Barents Region', Climatic Change, 87: 51-73.

Wramneby, Anna, Benjamin Smith, and Patrick Samuelsson. 2010. 'Hot spots of vegetation-
climate feedbacks under future greenhouse forcing in Europe', Journal of Geophysical Research,
115.

Wramneby, Anna, Benjamin Smith, Sönke Zaehle, and Martin T. Sykes. 2008. 'Parameter
uncertainties in the modelling of vegetation dynamics—Effects on tree community structure and
ecosystem functioning in European forest biomes', Ecological Modelling, 216: 277-90.

Wu, Minchao, Guy Schurgers, Markku Rummukainen, Benjamin Smith, Patrick Samuelsson,
Christer Jansson, Joe Siltberg, and Wilhelm May. 2016. 'Vegetation–climate feedbacks modulate
rainfall patterns in Africa under future climate change', Earth System Dynamics, 7: 627-47.

Wårlind, D., B. Smith, T. Hickler, and A. Arneth. 2014. 'Nitrogen feedbacks increase future
terrestrial ecosystem carbon uptake in an individual-based dynamic vegetation model',
Biogeosciences, 11: 6131-46.

Xu, L., R. B. Myneni, F. S. Chapin Iii, T. V. Callaghan, J. E. Pinzon, C. J. Tucker, Z. Zhu, J. Bi, P.
Ciais, H. Tømmervik, E. S. Euskirchen, B. C. Forbes, S. L. Piao, B. T. Anderson, S. Ganguly, R. R.
Nemani, S. J. Goetz, P. S. A. Beck, A. G. Bunn, C. Cao, and J. C. Stroeve. 2013. 'Temperature and
vegetation seasonality diminishment over northern lands', Nature Climate Change, 3: 581-86.

Yang, Zhenlin, Edward Hanna, and Terry V. Callaghan. 2011. 'Modelling surface‐air‐temperature
variation over complex terrain around abisko, swedish lapland: uncertainties of measurements
and models at different scales', Geografiska Annaler: Series A, Physical Geography, 93: 89-112.

Zhang, W., C. Jansson, P. A. Miller, B. Smith, and P. Samuelsson. 2014. 'Biogeophysical

feedbacks enhance the Arctic terrestrial carbon sink in regional Earth system dynamics',
Biogeosciences, 11: 5503-19.

Zhang, Wenxin, Paul A. Miller, Benjamin Smith, Rita Wania, Torben Koenigk, and Ralf Döscher.
2013. 'Tundra shrubification and tree-line advance amplify arctic climate warming: results from an
individual-based dynamic vegetation model', Environmental Research Letters, 8.

20