2  Biogeography

2.1 A global view of vegetation

We start with the observation that the global distribution of plants and characteristic assemblies of different plant species is not random, but relates to climate. Around the beginning of the 19th century, Alexander von Humboldt pursued expeditions to different places on Earth and connected the multitude of his local observations into a global view of biogeography. He was the first to realize that vegetation zones along elevational gradients were shifted across different mountain ranges on different continents and located at different latitudes, and that the elevational and latitudinal patterns in vegetation were related to climate - in much the same way across the Earth. This finding is visualised by the schematic scientific drawing in Figure 2.1. The figure illustrates, for example, that the treeline increases from the high northern latitudes of Lapland at 68\(^\circ\)N to the Alps and Pyrenees at 42-46\(^\circ\)N and to the Himalaya at 29-32\(^\circ\)N. Or that the elevation at which Betula alba (birch) grows varies from 1000-2000 ft in Lapland, to 4000-5000 ft in the Alps, and about 15,000 ft in the Himalaya. Figure 2.1 also illustrates that the treeline doesn’t increase monotonically with decreasing latitude. Although not spelled out in the figure, this is a hint at the importance of water, along with temperature, for controlling vegetation. We’ll revisit this observation later in this book. Humboldt’s integration of globally distributed observations into his global view of climate as a driver of vegetation was pioneering and forms the basis of our modern approach to understanding and modelling global vegetation patterns and how they are influenced by climate change.

Figure 2.1: Vegetation zones along elevation and latitude. Source: Wikimedia.

2.2 Biomes

Today’s geodata provides information on climate and vegetation properties with global coverage. This enables a global mapping of vegetation, filling in the blank spots between the local observations made by Alexander von Humboldt 200 years ago. Global patterns in the distribution of vegetation become apparent when grouping plants and plant functional types (see Section 2.3) into biomes or broad vegetation classes. A global map of biomes following the definition and global delineation by Olson et al. (2001) is given in Figure 2.2. The exact delineation and naming of biomes varies between different sources, but the main distinctions are shared among all of them.

Figure 2.2: Fourteen biomes and the locations of sites representative for each biome (used below).

Each biome is characterised by the types of dominant plants (plant functional types), their characteristics (traits), ecosystem structural and functional properties, and landscape processes (e.g., fire). Underlying all of this is the characteristic climate. The following biome classification (based on Bonan (2015) and slightly deviating from Olson et al. (2001)) can be made.

Note

Click on the biome names below to open a tab that lists some key characteristics and provides a Walter-Lieth climate diagram for a representative site, located in that biome. The sites are locations where ecosystem water and carbon fluxes, along with meteorological variables, are measured. We will revisit data from these same sites in later chapters, then with a focus on water and carbon fluxes and the phenology. The locations of the sites are indicated on the map in Figure 2.2.

Walter-Lieth climate diagrams show monthly climatologies of average temperature and precipitation on the same axis and scaled such that months where the blue curve (precipitation, mm month^-1) is above the red curve (temperature, \(^\circ\)C) indicates humid periods, while the opposite relative curve position indicates dry periods. Dry periods are roughly characterized by a potential evapotranspiration higher than precipitation (see Chapter 7). Blue bars along the x-axis indicate months with likely frost. Annotations on the top of the graph indicate the observation period (left), the mean annual temperature (center), and the mean annual precipitation (right). The average temperatures of the warmest and coldest months are given to the left of the temperature axis.

Tropical moist broadleaved (evergreen) forest

• warm year-round
• no or little precipitation seasonality
• tall vegetation height, high productivity, high total leaf area per unit ground area (leaf area index, LAI), fast decomposition
• large, thin leaves

Figure 2.3: Climate diagram of the site BR-Sa3, located in the ropical moist broadleaved (evergreen) forest biome.

Tropical deciduous forest (xeric woodland)

• warm year-round
• dry season
• leaves are (partly) shed during the dry season as a water saving strategy
• shorter stature and lower LAI than tropical moist broadleaved (evergreen) forests
• small, thick leaves

Figure 2.4: Climate diagram of the site ZM-Mon, located in the Tropical deciduous forest biome.

Tropical savannah

• warm year-round
• pronounced dry season
• widely spaced trees, mixed with grasses
• low plant biomass, deep roots
• regular fires

Figure 2.5: Climate diagram of the site AU-How, located in the Tropical savannah biome.

Grasslands

• tropical and temperate dry climates
• C₃ and C₄ grasses (see below), distribution governed by temperature
• temperate grasslands in dry, seasonally hot, semiarid climates with annual precipitation less than about 1000 mm
• transitional between forests and desert

Figure 2.6: Climate diagram of the site US-Tw1, located in the Grasslands biome.

Shrublands and deserts

• less than 250 mm annual precipitation
• short, widely spaced desert scrub and shrubland
• germination after intense rain, leaf shedding during dry period
• cacti and succulents store water and use the CAM photosynthetic pathway to minimize water loss during photosynthesis

Figure 2.7: Climate diagram of the site US-SRM, located in the Shrublands and deserts biome.

Temperate forests

• more than 1000 mm annual precipitation
• deciduous and evergreen trees, sometimes mixed
• low winter light and temperature limits productivity
• tall trees

Figure 2.8: Climate diagram of the site DE-Hai, located in the Temperate forests biome.

Mediterranean forests

• additional productivity limitation in summer due to water limitation
• less limited productivity by temperature and light in winter than temperate forests
• mild and moist winters
• short stature trees, dense shrubs
• thick, waxy leaves

Figure 2.9: Climate diagram of the site FR-Pue, located in the Mediterranean forests biome.

Boreal forests

• very cold winters, cool summers
• winter light and temperature limits productivity
• mostly needle-leaved evergreen
• shorter and more open than temperate forests
• in Siberia: extreme winter low temperatures favour deciduousness of larch (larix decidua)

Figure 2.10: Climate diagram of the site FI-Hyy, located in the Boreal forests biome.

Tundra

• tree-less, grass-like sedges, dwarf shrubs, lichens, and mosses
• very short vegetation period
• frozen ground, seasonal thawing of top ~50 cm (permafrost)

Figure 2.11: Climate diagram of the site US-ICh, located in the Tundra biome.

2.2.1 Biomes in climate space

Figure 2.2 shows biomes in geographical space - as a global map. Acknowledging their intricate relationship with climate, biomes can also be delineated in climate space - in a diagram defined by mean annual temperature (MAT) along one axis and mean annual precipitation (MAP) along the other axis. This yields a plot following Whittaker (1975). Note that the biome definition shown in Figure 2.12 does not exactly match the definition of biomes given above as it is mainly aimed at separating with respect to MAT and MAP. For example, Mediterranean forests are not separated in the “Whittaker-plot”. This is because their characteristic climate is not distinguished by annual means, but by the seasonality of temperature, solar radiation, and precipitation.

Code

# library(ggplot2)
# library(devtools)
# devtools::install_github("valentinitnelav/plotbiomes")
# library(plotbiomes)
# plotbiomes::whittaker_base_plot() +
#   theme_classic()
# ggsave(here::here("book/images/biomes_whittaker.png"), width = 7, height = 4)

knitr::include_graphics("images/biomes_whittaker.png")

Figure 2.12: Biomes in climate space, defined by mean annual temperature and precipitation, following Whittaker (1975).

2.3 Plant functional types

Each biome is characterised by a typical assembly of plant functional types (PFT, Figure 2.13). PFTs are a grouping of plant species based on their key physiological, morphological, and life history characteristics. That is, we can distinguish between annual (mostly grasses and herbs) and perennial (mostly trees and shrubs) plants, between needle-leaved and broadleaved trees, and between deciduous and evergreen trees. A key physiological distinction can further be made between grasses following the C₃ vs. the C₄ photosynthetic pathway (see Chapter 4). Thereby, the bewildering diversity of the plant kingdom can be reduced to a small set of PFT. A common categorization distinguishes the following PFTs:

1. needle-leaved evergreen trees
2. broadleaved evergreen trees
3. needle-leaved deciduous trees
4. broadleaved deciduous trees
5. broadleaved annuals (herbs)
6. C₃ grasses
7. C₄ grasses

Global vegetation models use PFTs as their basic unit for distinguishing plants. The exact delineation of PFTs implemented in such models may vary from the list given above. Further distinctions may be made and are relevant in a global vegetation and carbon cycle modelling context. For example, only a relatively small subset of plants is known to associate with symbiotic nitrogen (N)-fixing bacteria that live in root nodules of the host plant (“N-fixing plants”, see also Chapter 11). This association is highly relevant for the N economy of the plant and its productivity and competitiveness under different levels of N availability.

Plants can also be distinguished into the botanical classification of angiosperms (flowering plants) and gymnosperms (seed-producing plants that include conifers, cycads, and Ginkgo). The distinction between angiosperms and gymnosperms largely aligns with the distinction between needle-leaved and broadleaved plants (but see Ginkgo). The two groups are not only distinguished by their phylogenetic heritage, but also by essential characteristics that relate to the efficiency by which they photosynthesise and transpire water. Angiosperm leaves typically exhibit higher photosynthesis and transpiration rates and are thinner and shorter-lived than leaves of gymnosperms. These differences relate to differences in how the water transport system (plant hydraulics) is built. A larger number and a wider diameter of water transport organs in angiosperms enable a higher water conductivity - essential for sustaining higher photosynthetic rates than in gymnosperms.

Figure 2.13: Assembly of plant functional types as a basis for biome classification

2.4 Traits

The physiological, morphological, and life history characteristics of different plants determine their productivity and competitiveness in a given climate. Such characteristics are referred to as plant functional traits, or often just traits. Plant species can be described by a set of traits and a subset of certain traits yields the distinction into PFTs described above: leaf habit (deciduous vs. evergreen), leaf form (needle-leaved vs. broadleaved), and the life history strategy distinguishing annual vs. perennial. A range of additional traits are commonly described and investigated scientifically. Here, we will not consider additional ones. The concept of a plant functional trait is that it describes a largely immutable characteristic of a plant species that determines metabolic rates (photosynthesis, respiration) and their relationship to the abiotic environment (e.g., temperature), nutrient and water demand, and ultimately its demographic rates (growth, fecundity, mortality) and thus competitiveness.

Today’s global distribution of species and biomes (assuming no intervention by human land use and forest management) is the outcome of competition and thus reflects the combination of plant functional traits that optimises competitiveness of a species under the present-day climate. The distribution of biomes and PFTs is thus a direct reflection of the climate.

For example, whether a region is dominated by deciduous or by evergreen trees and forests is determined by the benefits and costs for a plant of maintaining leaves year-round. Evergreen trees benefit from the ability to photosynthesise and gain carbon around the year. For example, in Mediterranean regions, although light levels are lower in winter than in summer, ample moisture and non-negligible light enables evergreen trees to assimilate carbon also during winter months. However, leaves and needles of evergreen plants have to be built for lasting several years. Such leaves are typically much thicker than those of deciduous trees and thus require more carbon per unit leaf area for their construction. The leaf mass per unit leaf area is commonly referred to as the leaf mass per area, LMA, and is an important additional plant trait as it is directly linked to a plant’s carbon balance. Over the leaf lifespan, the initial high construction costs of high-LMA leaves are outweighed by the additional carbon assimilation during periods when deciduous trees shed their leaves.

Leaf-shedding of deciduous trees, in contrast, is a strategy to avoid having to build costly long-lasting leaves and maintaining them year-round (which also incurs an additional respiration cost, also in the form of carbon). Leaves are shed during periods when the climate is unfavourable for photosynthesis - during cold and dark winter months, or during excessively dry periods.

Note that plant functional traits are often not entirely immutable. Instead, traits may vary also within a species, and these variations are often driven by the environment. This is called acclimation. Such variations can even arise over the course of a season. For example, photosynthetic traits can rapidly acclimate to the large changes in light availability over the course of a year. Some traits are more plastic than others. For example, a tree is either needle-leaved or broadleaved. There is no continuum between the two leaf forms. In contrast, the nitrogen content per unit leaf mass is relatively plastic within a species.

A changing environment changes the competitiveness of a given species, i.e., of a given trait combination. As a result, some traits may acclimate to some extent within weeks to years. Over longer time scales, the altered demographic rates in a new climate affect the competitiveness of a species (even after some of its traits may have acclimated to a new climate) and ultimately shift demographic rates and the community composition. In grasslands, where the demographic cycle is short, such community composition changes may unfold over time scales of a few years. In forests, the longevity of an individual tree is on the order of decades to centuries and community composition changes unfold on a correspondingly long time scale.

2.5 Global vegetation patterns

The biome classification is a way to discretize vegetation based on several characteristics (e.g., tree cover fraction). However, many of these characteristics describe observable variables that vary more or less gradually across environmental gradients, and each of these variables can be mapped across the globe thanks to Earth observation data. These global patterns of different observable variables reflect how the climate influences vegetation structure and functioning across the globe (independent of a classification into biomes). In much of the remainder of this course, we will investigate these vegetation-climate relationships without considering the biome classification. These relationships are informative for understanding how different processes of terrestrial ecology, plant physiology, the carbon cycle, and land-climate interactions are driven by the environment and vary across the globe.

Vegetation height

Figure 2.14: The global distribution of vegetation height from Simard et al. (2011).

Figure 2.15: Distribution of vegetation height by biome. Data from Simard et al. (2011). Note that the vegetation height data has gaps in areas with sparse or generally short-statured vegetation cover (deserts, grasslands, Tundra, see Figure 2.14). Therefore, the pixels for which data is used for visualising the vegetation height per biome is biased towards particularly tall vegetation in these areas. Consequently, the distribution of vegetation heights in the biome ‘Tundra’ appears implausibly high.

Leaf area index

Figure 2.16: Annual maximum leaf area index, averaged over years 2000-2008. Data from the MODIS MOD15A2 (C006) product.

Figure 2.17: Multi-year average of the annual maximum leaf area index by biome.

Biomass

Figure 2.18: Aboveground biomass carbon, mean over years 1993-2012. Data from Liu et al. (2015).

Figure 2.19: Aboveground biomass carbon by biome. Data from Liu et al. (2015).

Tree cover fraction

Figure 2.20: Tree cover fraction. Data from the MODIS MOD44B (v061) product.

Figure 2.21: Tree cover fraction by biome.

Note that the global vegetation patterns shown above are based on Earth observation data from satellite remote sensing. These reflect the actual state of the land surface and thus the influence of human land use and land use change. In contrast, biomes are defined for a potential natural vegetation. Therefore, you see, for example, that the tree cover fraction has very wide distributions, particularly in the biome “Temperate broadleaved and mixed forests” (Figure 2.21).

2.6 Vegetation types

Note that the categorisation of vegetation into biomes is different from a categorisation of vegetation into vegetation types or land cover classes. A widely used distinction of vegetation types/land cover classes is that of the International Geosphere Biosphere Programme (IGBP, see Table 2.1). Vegetation types distinguish what can be seen from space and are thus amenable to remote-sensing based land cover classification. However, for example, different types of evergreen broadleaved forests are not distinguished (e.g., moist tropical, Mediterranean, and temperate evergreen broadleaved forests). Also, vegetation types/land cover classes distinguish land cover that is dominated by human activities (agricultural and urban), while biomes distinguish naturally occurring (i.e., without human influence) vegetation zones. In other words, biomes reflect the potential natural vegetation.

Table 2.1: IGBP vegetation types/land cover classes. Source: https://fluxnet.org/data/badm-data-templates/igbp-classification/

Code	Name	Description
BSV	Barren Sparse Vegetation	Lands exposed soil, sand, or rocks and has less than 10% vegetative cover during any time of the year.
CRO	Croplands	Lands covered with temporary crops followed by harvest and a bare soil period (e.g., single and multiple cropping systems). Note that perennial woody crops will be classified as the appropriate forest or shrub land cover type.
CSH	Closed Shrublands	Lands with woody vegetation less than 2 meters tall and with shrub canopy cover >60%. The shrub foliage can be either evergreen or deciduous.
CVM	Cropland/Natural Vegetation Mosaics	Lands with a mosaic of croplands, forest, shrublands, and grasslands in which no one component comprises more than 60% of the landscape.
DBF	Deciduous Broadleaf Forests	Lands dominated by woody vegetation with a percent cover >60% and height exceeding 2 meters. Consists of broadleaf tree communities with an annual cycle of leaf-on and leaf-off periods.
DNF	Deciduous Needleleaf Forests	Lands dominated by woody vegetation with a percent cover >60% and height exceeding 2 meters. Consists of seasonal needleleaf tree communities with an annual cycle of leaf-on and leaf-off periods.
EBF	Evergreen Broadleaf Forests	Lands dominated by woody vegetation with a percent cover >60% and height exceeding 2 meters. Almost all trees and shrubs remain green year round. Canopy is never without green foliage.
ENF	Evergreen Needleleaf Forests	Lands dominated by woody vegetation with a percent cover >60% and height exceeding 2 meters. Almost all trees remain green all year. Canopy is never without green foliage.
GRA	Grasslands	Lands with herbaceous types of cover. Tree and shrub cover is less than 10%. Permanent wetlands lands with a permanent mixture of water and herbaceous or woody vegetation. The vegetation can be present in either salt, brackish, or fresh water.
MF	Mixed Forests	Lands dominated by trees with a percent cover >60% and height exceeding 2 meters. Consists of tree communities with interspersed mixtures or mosaics of the other four forest types. None of the forest types exceeds 60% of landscape.
OSH	Open Shrublands	Lands with woody vegetation less than 2 meters tall and with shrub canopy cover between 10-60%. The shrub foliage can be either evergreen or deciduous.
SAV	Savannas	Lands with herbaceous and other understory systems, and with forest canopy cover between 10-30%. The forest cover height exceeds 2 meters.
SNO	Snow and Ice	Lands under snow/ice cover most of the year.
URB	Urban and Built-Up Lands	Land covered by buildings and other man-made structures.
WAT	Water Bodies.
WET	Permanent Wetlands	Lands with a permanent mixture of water and herbaceous or woody vegetation that cover extensive areas. The vegetation can be present in either salt, brackish, or fresh water.
WSA	Woody Savannas	Lands with herbaceous and other understory systems, and with forest canopy cover between 30-60%. The forest cover height exceeds 2 meters.

Code

library(ggplot2)
library(terra)
library(tidyterra)
library(rnaturalearth)
library(rnaturalearthdata)

# Most color codes used here are from external data
# load(here::here("data/worldgrids_pal.rda")) # obtained from https://github.com/Envirometrix/plotKML/blob/master/data/worldgrids_pal.rda

# load land cover data, raster data
raster_vegtype <- terra::rast(
  here::here("data/modis_landcover__LPDAAC__v5.1__0.1deg__2010.nc")
  ) |> 
  tidyterra::mutate(Band1 = as.factor(Band1)) |> 
  tidyterra::rename(landcover = Band1)
  
# load coast outline, vector data
coast <- rnaturalearth::ne_coastline(scale = 110, returnclass = "sf")

# create map
ggplot() +
  tidyterra::geom_spatraster(data = raster_vegtype) +
  geom_sf(data = coast,
          colour = 'black',
          linewidth = 0.1) +
  scale_fill_manual(
    values = c(
      "1" = "azure", # worldgrids_pal$IGBP["Water " ] , #"water" ;
      "2" = "dodgerblue4" , #"evergreen needleleaf forest" ;
      "3" = "forestgreen" , #"evergreen broadleaf forest" ;
      "4" = "#9ED003", # worldgrids_pal$IGBP["Deciduous Needleleaf forest "] , #"deciduous needleleaf forest" ;
      "5" =  "springgreen3" , #"deciduous broadleaf forest" ;
      "6" =  "#31A278", # worldgrids_pal$IGBP["Mixed forest "] #"mixed forest" ;
      "7" =  "#A84974", #worldgrids_pal$IGBP["Closed shrublands " ] , #"closed shrublands" ;
      "8" =  "#FFD3A0" , #"open shrublands" ;
      "9" =  "#CAFE8F", #worldgrids_pal$IGBP["Woody savannas "] , #"woody savannas" ;
      "10" = "#F7D600", #worldgrids_pal$IGBP["Savannas " ] , # "savannas" ;
      "11" = "khaki3" , # "grasslands" ;
      "12" = "violetred" , # "permanent wetlands" ;
      "13" = "#FAFE03", # worldgrids_pal$IGBP["Croplands " ] , # "croplands" ;
      "14" = "#FF2A00", #worldgrids_pal$IGBP["Urban and built-up "] , # "urban and built-up" ;
      "15" = "#8A9000" , #worldgrids_pal$IGBP["Cropland/Natural vegetation mosaic "] , # "cropland (natural vegetation mosaic)" ;
      "16" = "#FFFFFF", #worldgrids_pal$IGBP["Snow and ice "] , # "snow and ice" ;
      "17" = "#AAAAAA", # worldgrids_pal$IGBP["Barren or sparsely vegetated "] , # "barren or sparsely vegetated" ;
      "18" = "#424242" # worldgrids_pal$IGBP["Unclassified "] # "unclassified or fill values" ;      
    ),
    labels = c(
      "1" = "water",
      "2" = "evergreen needleleaf forest",
      "3" = "evergreen broadleaf forest",
      "4" = "deciduous needleleaf forest",
      "5" = "deciduous broadleaf forest",
      "6" = "mixed forest",
      "7" = "closed shrublands",
      "8" = "open shrublands",
      "9" = "woody savannas",
      "10" = "savannas",
      "11" = "grasslands",
      "12" = "permanent wetlands",
      "13" = "croplands",
      "14" = "urban and built-up",
      "15" = "cropland (natural vegetation mosaic)",
      "16" = "snow and ice",
      "17" = "barren or sparsely vegetated",
      "18" = "unclassified or fill values"      
    )
  ) + 
  scale_x_continuous(expand = c(0, 0)) +  # avoid gap between plotting area and axis
  scale_y_continuous(expand = c(0, 0), limits = c(-60, 85)) +
  labs(title = "Vegetation types") + 
  theme(legend.position = "bottom",
        legend.text = element_text(size = 6),
        legend.title = element_blank())

2.7 Climate zones

While the biome categorisation is based on plant functional types and reflects a mix of vegetation types and climate, the categorisation into vegetation types is independent of climate, but purely based on vegetation and land cover properties that can be sensed from space. In contrast to this approach, we may focus exclusively on the climate and distinguish climate zones. The Köppen-Geiger climate classification (Köppen 1884) is based on a hierarchical classification that separates climate based on the seasonality of monthly air temperature and precipitation. The climate zones are chosen such that they largely align with the biomes. For this, water availability during the main vegetation periods and temperature minima during winter and maxima during summer are recognized to be particularly influential for the composition of the vegetation and are considered for the Köppen-Geiger climate classification.

Figure 2.22: Overview of the Köppen-Geiger climate classes including the defining criteria. Taken from Beck et al. (2018) and adapted from Peel, Finlayson, and McMahon (2007). Variable definitions: MAT = mean annual air temperature (°C); \(T_{cold} =\) the air temperature of the coldest month (°C); \(T_{hot} =\) the air temperature of the warmest month (°C); \(T_{mon10} =\) the number of months with air temperature >10 °C (unitless); MAP = mean annual precipitation (mm y\(^{−1}\)); \(P_{dry} =\) precipitation in the driest month (mm month\(^{−1}\)); \(P_{sdry} =\) precipitation in the driest month in summer (mm month\(^{−1}\)); Pwdry = precipitation in the driest month in winter (mm month\(^{−1}\)); \(P_{swet} =\) precipitation in the wettest month in summer (mm month\(^{−1}\)); \(P_{wwet} =\) precipitation in the wettest month in winter (mm month\(^{−1}\)); \(P_{threshold}\) = 2×MAT if >70% of precipitation falls in winter, \(P_{threshold}\) = 2×MAT+28 if >70% of precipitation falls in summer, otherwise \(P_{threshold}\) = 2×MAT+14. Summer (winter) is the six-month period that is warmer (colder) between April-September and October-March.

Based on the classification scheme described in Figure 2.22 and data of monthly climatologies, the climate zones can be mapped across the globe (Figure 2.23).

Figure 2.23: Köppen-Geiger climate zones for a present-day climate (1980-2016). See Figure 2.22 for a definition of the codes (names). Figure from Beck et al. (2018).

2.8 Hillslope-scale heterogeneity

Topography shapes microclimates and drives small-scale variations in hydrology. In mountain regions, vegetation may thus vary strongly along small spatial scales - on the order of 10-10³ m. This scale - from the river channel to the ridge - is referred to as the hillslope scale. In much the same way that climate drives vegetation across biomes, so it does across the hillslope scale. The incident solar radiation and - as a consequence of that - near-surface air temperatures are affected by the local slope, aspect, and shading by the surrounding topography. Soil moisture and the groundwater table depth are affected by lateral subsurface flow of water, driven by gradients in water potentials along topographical gradients. Subsurface water flow converges in depressions and concave terrain (e.g., in valley bottoms) and diverges in convex terrain. (e.g., on ridges and hilltops). As a consequence, the water table is shallow in valley bottoms and deep under ridges. Radiation and hydrology thus create microclimates and plant water availability conditions that are shaped by topography, from which result variations in vegetation.

These hillslope-scale variations “intersect” with the background climate (average climate across larger spatial scales, on the order of 10⁴-10⁵ m). A shallow groundwater table and moist soils in valley bottoms can promote plant productivity in arid regions and seasons. In contrast, in a humid climate, a very shallow water table in valley bottoms inhibits plant productivity due to anaerobic conditions in waterlogged soils and the inability of roots to penetrate into permanently water-saturated soil. Similarly, the influence of radiation in promoting versus inhibiting plant productivity depends on the background climate. In cold climates of the high northern latitudes, the solar zenith angle is relatively also, even in mid-summer. This creates a strong influence of the local slope and aspect. In the northern hemisphere, south-facing slopes receive more radiation - a difference to north-facing slopes that can be critical for sustaining tree growth in high northern regions. In contrast, elevated incident solar radiation in south-facing slopes can aggravate water limitation and suppress vegetation productivity in more arid climates of the subtropics and mid-latitudes of the northern hemisphere.

Figure 2.24: Hillslope effects on vegetation in different background climate zones. Top-left: Hydrology promoting vegetation in valley bottoms in a semi-arid climate in Arizona, USA (Source: wikipedia.org). Top-right: Hydrology inhibiting vegetation in topographic depressions in a humid climate due to seasonal waterlogging in the Pantanal region (Source: wikimapia.org). Bottom-left: Radiation inhibiting vegetation on south-facing (sun-exposed) slopes in Texas, USA (Source: thecommonmilkweed.blogspot.com). Bottom-right: Radiation promoting vegetation on south-facing (sun-exposed) slopes in southwest Yukon, Canada (Source: photo by Ryan Danby, Queen’s University, Canada). Figure extracted from Fan et al. (2019).

Climate also varies along elevation due to thermodynamics and radiative transfer in the atmosphere - independent of the slope and aspect of the local terrain. Temperatures decline following the lapse rate in the troposphere, leading to a shift of vegetation zones along large elevational gradients. This phenomenon is depicted also in Humboldt’s drawings (Figure 2.1). Radiation levels are elevated at high altitudes due to a shorter path length of solar radiation travelling across the atmosphere and, consequently, a reduced attenuation of the radiation intensity. Atmospheric pressure and thus the partial pressure of oxygen declines with elevation. This implies reduced respiration and an increased efficiency of photosynthesis.

Small-scale variations in vegetation may also be driven my variations in soil characteristics and bedrock lithology which influences drainage and soil chemistry (pH, nutrient availability). Where the ground is seasonally snow-covered, very small-scale terrain features (10⁰-10² m) and dominant wind directions drive snow accumulation and dispersion, leading to substantial small-scale variations in maximum snow depth and the duration of seasonal snow cover - with implications for plants (Körner and Hiltbrunner 2021).

2.9 Species distribution modelling and ecological niche

The notion that climate drives vegetation and the distribution of plant species enables a prediction and mapping of the species occurrence across geographical space. Data on species presence and absence can be associated with multiple variables characterizing the abiotic environment (mostly climate) recorded for the same location. The resulting empirical patterns can be used for spatial modelling and predicting whether a species is likely to occur at a new location or even a novel climate Figure 2.25.

Figure 2.25: Methodological approach to species distribution modelling and predicting to a novel climate. Figure from Svenning et al. (2011).

The approach to species distribution modelling is fundamentally rooted in the ecological niche concept. As formulated by Hutchinson (1957; Araújo and Guisan 2006), each species has a fundamental niche that can be conceived as a domain in a multidimensional space defined by the environmental conditions. Environmental filtering yields the fundamental niche which is generally a subset of the Earth’s climate space (Figure 2.26). The actual, or realized niche is smaller as it is additionally constrained by biotic interactions with other species. Natural selection yields the realized niche. Several challenges for species distribution modelling exist (Zimmermann et al. 2010). For example, the species distribution is hardly ever in equilibrium with the environment, given that climate has been constantly changing (although not as rapidly as today for millions of years); topographic heterogeneity, climatic history, and ecological refugia affect the species distribution and how it varies under climate change; genetic adaptation can lead to populations within species that exhibit different responses to the environment; and niche stability over long time scales may be undermined not only by genetic adaptation, but also by physiological effects of CO₂ which fundamentally alters photosynthesis and transpiration and will therefore shift water availability-related niche limits observed today.

Figure 2.26: Environmental filtering yields the fundamental niche. Biotic interactions yield the realized niche.

2.10 Temporal variations of vegetation

Spatial patterns in vegetation are influenced by how climate varies across different areas. Similarly, changes in vegetation are influenced by how climate changes over time. Vegetation change at temporal scales of thousands of years is recorded, for example, by pollen deposited in layered sediments. Given that they are preserved and that the sediment layers can be dated, the vegetation composition over time can be reconstructed. An example is given for the Moossee, near Bern in Figure 2.27. The pollen diagram documents how vegetation on the Swiss Plateau underwent several major changes since the last Glacial. At around 19 kyr BP, a steppe tundra established, replaced around 16 kyr BP by a shrub tundra. At around 15 kyr BP, a boreal forest established before it was replaced by a temperate mixed oak forest at the beginning of the Holocene - the current warm period at 11,600 yr BP. A beech-dominated temperate forest, as it is common for this region today, established around 8200 yr BP. Figure 2.27 also shows that these biome replacements coincided with rapid climatic shifts and that the vegetation change was largely instantaneous. This instantaneous response of species composition is interpreted as a reflection of the importance of glacial refugia from which a species may rapidly expand upon climate change (Rey et al. 2020).

Figure 2.27: Pollen diagram from the Moossee, Switzerland. The diagram shows the percentage of pollen recorded over time (green areas). Cultural indicators and charcoal influx indicate the presence of humans in the region. Parallel variations observed from other data sources are also shown for interpretation: Millennial-scale variations in solar radiation, lake levels, and temperature reconstructions. Figure from Rey et al. (2020).

Multiple vegetation reconstructions for the Last Glacial Maximum (LGM) obtained from globally distributed locations provide a picture of how biomes were distributed across the Earth around 20 ka BP (Figure 2.29, bottom panel). The difference in climate and CO₂ (global mean temperatures were around 6°C lower than at present; CO₂ was at around 180 ppm, today it’s at over 420 ppm) caused forest biomes to recede and grasslands to expand across most of the globe. The upper panel of Figure 2.29 shows the simulated biome distribution. These simulation results are not based on a species distribution model, but on a Dynamic Global Vegetation Model that accounts for both climate and CO₂ effects and simulates the distribution of PFTs from which biomes were derived here. The simulation of vegetation distribution at the level of PFTs instead of species is less affected by biotic interactions that are hard to simulate at global scale.

Figure 2.28: The global distribution of biomes at the Mid-Holocene (6000 kyr BP). Figure from I. Colin Prentice, Jolly, and Participants (2000)

Figure 2.29: The global distribution of biomes at the Last Glacial Maximum from a model simulation (top) and from a biome reconstruction based on pollen records. Figure from I. C. Prentice, Harrison, and Bartlein (2011).

The mapping of vegetation for past climates is enabled by the biomization of paleo-ecological records (I. Prentice et al. 1996). Classifying vegetation into biomes is useful for distilling ecological information beyond the anthropogenic impact. Biomes reflect a potential natural vegetation. The use of biomes is therefore important for designing conservation efforts suited for a specific climate (Olson et al. 2001), and is instructive for understanding environmental controls on ecosystem types as reflected by the agreement of biomes and climate zones (Section 2.7). Yet, to understand principles and mechanisms of how plant physiology, biogeochemical cycling, and land-climate interactions are shaped by the environment, we often don’t need to rely on a classification of vegetation. As we will learn in later chapters, the abiotic environment drives vegetation functioning in much the same way (bar some exceptions) through generally applicable physical laws across the whole terrestrial biosphere.

Exercise

1. Explain the difference between biomes, as defined for example by Olson et al. (2001), and the IGBP vegetation classes.
2. Give a good reason for why precipitation and temperature are plotted both along the y-axis in the Walter-Lieth climate diagrams. What is the meaning of the blue hashed area and the red dotted area?
3. Compare the seasonality of temperature and precipitation of the example site given for the Mediterranean forest biome (FR-Pue) and the boreal forest biome (FI-Hyy). Which site do you expect to have a stronger seasonal variation in runoff?
4. For the seven biomes for which climate diagrams are shown above, establish a ranking with respect to the growing season length.
5. Consider an ecosystem that is characterised with a fractional plant cover of trees of 10% and a fractional plant cover of grasses of 90%. What biome does it belong to?
6. In a changing climate, where do you expect vegetation composition shifts to unfold faster: in a large plain or in a mountainous landscape?

Araújo, Miguel B., and Antoine Guisan. 2006. “Five (or so) Challenges for Species Distribution Modelling.” Journal of Biogeography 33 (10): 1677–88. https://doi.org/10.1111/j.1365-2699.2006.01584.x.

Beck, Hylke E., Niklaus E. Zimmermann, Tim R. McVicar, Noemi Vergopolan, Alexis Berg, and Eric F. Wood. 2018. “Present and Future Köppen-Geiger Climate Classification Maps at 1-Km Resolution.” Scientific Data 5 (1): 180214. https://doi.org/10.1038/sdata.2018.214.

Bonan, Gordon. 2015. Ecological Climatology: Concepts and Applications. 3rd ed. Cambridge University Press.

Fan, Y., M. Clark, D. M. Lawrence, S. Swenson, L. E. Band, S. L. Brantley, P. D. Brooks, et al. 2019. “Hillslope Hydrology in Global Change Research and Earth System Modeling.” Water Resources Research 55 (2): 1737–72. https://doi.org/https://doi.org/10.1029/2018WR023903.

Hutchinson, G. Evelyn. 1957. “Concluding Remarks.” Cold Spring Harbor Symposia on Quantitative Biology 22 (January): 415–27. https://doi.org/10.1101/SQB.1957.022.01.039.

Köppen, Wladimir. 1884. “Die Wärmezonen Der Erde, Nach Der Dauer Der Heissen, Gemässigten Und Kalten Zeit Und Nach Der Wirkung Der Wärme Auf Die Organische Welt Betrachtet (the Thermal Zones of the Earth According to the Duration of Hot, Moderate and Cold Periods and to the Impact of Heat on the Organic World).” Meteorologische Zeitschrift 20 (3): 351–60. https://doi.org/10.1127/0941-2948/2011/105.

Körner, Christian, and Erika Hiltbrunner. 2021. “Why Is the Alpine Flora Comparatively Robust Against Climatic Warming?” Diversity 13 (8): 383. https://doi.org/10.3390/d13080383.

Liu, Yi Y., Albert I. J. M. van Dijk, Richard A. M. de Jeu, Josep G. Canadell, Matthew F. McCabe, Jason P. Evans, and Guojie Wang. 2015. “Recent Reversal in Loss of Global Terrestrial Biomass.” Nature Climate Change 5 (5): 470–74. https://doi.org/10.1038/nclimate2581.

Olson, David M., Eric Dinerstein, Eric D. Wikramanayake, Neil D. Burgess, George V. N. Powell, Emma C. Underwood, Jennifer A. D’amico, et al. 2001. “Terrestrial Ecoregions of the World: A New Map of Life on Earth: A New Global Map of Terrestrial Ecoregions Provides an Innovative Tool for Conserving Biodiversity.” BioScience 51 (11): 933–38. https://doi.org/10.1641/0006-3568(2001)051[0933:TEOTWA]2.0.CO;2.

Peel, M. C., B. L. Finlayson, and T. A. McMahon. 2007. “Updated World Map of the Köppen-Geiger Climate Classification.” Hydrology and Earth System Sciences 11 (5): 1633–44. https://doi.org/10.5194/hess-11-1633-2007.

Prentice, I Colin, D. Jolly, and BIOME 6000 Participants. 2000. “Mid-Holocene and Glacial-Maximum Vegetation Geography of the Northern Continents and Africa.” Journal of Biogeography 27 (3): 507–19. http://www.jstor.org/stable/2656208.

Prentice, I. C., S. P. Harrison, and P. J. Bartlein. 2011. “Global Vegetation and Terrestrial Carbon Cycle Changes After the Last Ice Age.” New Phytologist 189 (4): 988–98. https://doi.org/10.1111/j.1469-8137.2010.03620.x.

Prentice, Iain, Guiot Joel, Brian Huntley, Jolly D., and Rachid Cheddadi. 1996. “Reconstructing Biomes from Palaeoecological Data: A General Method and Its Application to European Pollen Data at 0 and 6 Ka.” Cd 12 (February): 185. https://doi.org/10.1007/s003820050102.

Rey, Fabian, Erika Gobet, Christoph Schwörer, Albert Hafner, Sönke Szidat, and Willy Tinner. 2020. “Climate Impacts on Vegetation and Fire Dynamics Since the Last Deglaciation at Moossee (Switzerland).” Climate of the Past 16 (4): 1347–67. https://doi.org/10.5194/cp-16-1347-2020.

Simard, Marc, Naiara Pinto, Joshua B. Fisher, and Alessandro Baccini. 2011. “Mapping Forest Canopy Height Globally with Spaceborne Lidar.” Journal of Geophysical Research: Biogeosciences 116 (G4). https://doi.org/10.1029/2011JG001708.

Svenning, Jens-Christian, Camilla Fløjgaard, Katharine A. Marske, David Nógues-Bravo, and Signe Normand. 2011. “Applications of Species Distribution Modeling to Paleobiology.” Quaternary Science Reviews 30 (21): 2930–47. https://doi.org/10.1016/j.quascirev.2011.06.012.

Whittaker, Robert Harding. 1975. Communities and Ecosystems. 3rd ed. Macmillan.

Zimmermann, Niklaus E., Thomas C. Edwards Jr, Catherine H. Graham, Peter B. Pearman, and Jens-Christian Svenning. 2010. “New Trends in Species Distribution Modelling.” Ecography 33 (6): 985–89. https://doi.org/10.1111/j.1600-0587.2010.06953.x.