Lodgepole pine is native to the western part of North America, occurring from SW Alaska and Yukon to Utah, Colorado, and the Mexican state of Baja California (Karl 1993). The species has also been intentionally introduced worldwide. In New Zealand, lodgepole pine was introduced for commercial purposes and for erosion control (Richardson 1998). In northern Europe, lodgepole pine was widely planted because it presented a higher yield per ha and faster growth than the native Scots pine (Pinus sylvestris L.) (Karlman 1981). In Iceland, the species has been repeatedly imported for forestry since the first half of the 20^th century. Trees originating from the Skagway region are now the most widespread in Icelandic plantations (Sigurgeirsson 1988). The Icelandic plants are likely to be hybrids between P. contorta Douglas ex Loudon subsp. contorta and P. contorta subsp. latifolia Engelm. ex S. Watson (Rudolf and Lapp 1987).

Lodgepole pine has a wide ecological amplitude, and is well adapted to survive and reproduce in harsh environments (Wheeler and Guries 1982). Within its native range, it grows from near sea level to an altitude of 3,350 m a.s.l., and from the mild but cool and rainy Pacific coast to the cold and continental interior of the northern Rocky Mountains (Critchfield 1957). In its native range, lodgepole pine grows in a wide variety of topographic settings from flat plains to steep slopes and rocky ridges (Pfister and Daubenmire 1975). It tolerates a wide spectrum of soil conditions including both dry and wet, fertile and poor soils and even bare gravel (Despain 2001; Elfving et al. 2001).

The lodgepole pine’s ability to thrive across diverse ecological conditions, regenerate post-fire, and rapidly mature early in its life cycle are essential factors enabling it to play a wide array of successional roles (Elfving et al. 2001). On poor soils, the species can become dominant and represent the final climax stage, forming extensive monotypic stands (Timber Management Research Forest Service 1979).

Lodgepole pine has several life history traits that make it potentially highly invasive. These include small seed mass (<50 mg), short juvenile period (<10 years) and short interval between large seed crops. Small seed mass allows larger numbers of seeds produced, better dispersal, higher initial germinability, and shorter chilling period needed to overcome dormancy, whereas a short juvenile period and short interval between large seed crops translate into early and high recruitment (Richardson and Rejmánek 2004). These advantages facilitated the invasion of lodgepole pine in many countries where it had been introduced by humans (Ledgard 2001; Langdon et al. 2010; Jacobson and Hannerz 2020).

Steinadalur is a short (2 km) but relatively wide (1.7 km) valley about 3 km inland from the coast in SE Iceland. The valley (ca. 40 m a.s.l.) is open to the east but otherwise surrounded by mountains reaching up to 600 m a.s.l. The bottom of the valley is flat and has been filled with sediment (gravel and stones) by the glacial river Kaldakvísl. The surrounding mountains are largely covered by birch woodland (B. pubescens subsp. tortuosa) to an elevation of about 200 m a.s.l. The higher parts of the slopes are mostly dominated by heath and grassland vegetation, which is also patchily present in the lower parts of the valley (Fig. 1).

Figure 1.

Heath vegetation already colonised by lodgepole pine (plantation can be seen in the distance) (A) and young lodgepole pines (marked with arrows) colonising moss heath in the valley mouth (B) and a birch woodland with dense vegetation cover (C).

The plantation in Steinadalur consisting of lodgepole pine and Sitka spruce (Picea sitchensis) was initiated in 1959 and expanded to ca. 0.02 km² in 1961 (Guðmundsdóttir 2012). The first records of lodgepole pine spread beyond the original plantation date from 1985 (Guðmundsdóttir 2012).

There is no weather station in Steinadalur, but climate stations are located 45 km to the SW (Fagurhólsmýri) and 40 km to the NE (Höfn) and the similarity of their climate suggests that they are a good proxy for Steinadalur. The regional climate is highly oceanic with annual precipitation well over 1,500 mm and exceeding 100 mm in most months (unpublished data from the Icelandic Met Office). The average annual temperature (2000–2020) was 5.3 °C and 5.5 °C at Höfn and Fagurhólsmýri respectively, mean temperature of the warmest month was 13 °C and 14 °C, while that of the coldest month was -1.5 °C and -1.9 °C, respectively (see Suppl. material 1). The frost-free period for both stations was ca. 4 months, from June to September. It is almost certain that due to specific orographic conditions in Steinadalur, microclimatic conditions differ from Fagurhólsmýri and Höfn, but we do not expect these differences to be pronounced. The prevailing winds in the Steinadalur valley are from the northwest (Icelandic Met Office 2022).

We assessed the relationship between the species’ total colonised area and time, and the total number of trees in the transects over time by fitting linear and non-linear models to our observations. Considering the nature of the process (plant invasion) and well-documented spread patterns, the exponential function was likely the most suitable choice. For fitting both linear and exponential models, we employed the nonlinear least squares regression using the nls function in R 4.4.1 (R Core Team 2024).

The linear model assumed a constant growth rate over time expressed as:

A(t) = a + b × (t-1985)

where:

A(t) is the area occupied by the lodgepole pine (or tree count) at time t

a is the y-intercept, representing the initial area in 1985,

b is the slope of the line, representing the rate of change of the plant area over time

t represents the calendar year.

Whereas the exponential model was:

A(t) = A₀ × e^r×(t-1985)

where:

A(t) is the area occupied by the lodgepole pine (or tree count) at time t

A₀ is the initial area of spread in 1985

e represents Euler’s number

r is the growth rate

and t represents the calendar year.

The fitting process involved optimising the model parameters to minimise differences between predicted and observed values. Model comparison was performed using three metrics, i.e. residual standard error, variance explained and the Akaike Information Criterion (AIC), providing insights into the goodness of fit and model complexity.

Community level (alpha) species richness and diversity were compared for the three different vegetation types (uninvaded birch woodland, uninvaded heath and lodgepole pine plantation). Statistically significant differences between vegetation types were assessed using the Kruskal-Wallis test and pairwise comparisons using Dunn´s test with Bonferroni correction for p-values, at α = 0.05.

The extent of occurrence (EOO) of lodgepole pine in Steinadalur expanded nearly tenfold over just a decade, growing from 0.25 km² in 2010 to 2.39 km² in 2021 (Fig. 2). Over the same period, the number of lodgepole pine individuals recorded across 22 transects increased dramatically, rising from 429 in 2010 to 3,315 in 2021—an almost eightfold growth. Similarly, the average density of lodgepole pine across all transects rose by over sevenfold, from 0.06 plants/m² in 2010 to 0.46 plants/m² in 2021. Original data and fitted curves can be found in Suppl. material 1.

Figure 2.

The extent of occurrence (EOO) of lodgepole pine (P. contorta) in Steinadalur (1985–2021).

The extent of occurrence (EOO) and indices of population growth of lodgepole pine (measured by tree count and mean tree density in transects) were analysed using both linear and exponential models. Exponential models consistently outperformed linear ones, showing lower residual standard error, higher explained variance, and lower Akaike Information Criterion (AIC) values (Table 1). Consequently, the observed trends are best represented by an exponential growth model.

The mean annual spread rate of lodgepole pine increased significantly, from 8.5 ± 2.4 m/year during 1985–2010 to 61.6 ± 40.2 m/year between field studies (2010–2021). Local rates of spread also shifted, with minimum rates rising from 3.4 m/year in 2010 to 8.3 m/year in 2021, and maximum rates increasing from 13.4 m/year to 119.3 m/year over the same period (Fig. 3A). Elevational spread also progressed, with the highest recorded elevation rising from 70 m in 1985 to 116 m in 2010, and 170 m in 2021. Meanwhile, the lowest recorded elevations shifted downward, from 44 m in 1985 to 38 m in 2010, and 25 m in 2021 (Fig. 3B).

Figure 3.

The rate of horizontal (A) and vertical (B) spread of lodgepole pine in Steinadalur (SE Iceland).

Lodgepole pine density exhibited clear spatial gradients in both 2010 and 2021, with the highest densities near the original plantation and decreasing with distance. In 2010, peak densities of 0.5–0.6 plants/m² were observed primarily within 100 m of the plantation edge (Fig. 4). By 2021, these densities had increased by an order of magnitude, with plants recorded farther from the edge, often in areas previously uncolonised.

Figure 4.

Changes in lodgepole pine densities along the 22 transects in Steinadalur (SE Iceland).

Lodgepole pine colonisation occurred across various native habitats in Steinadalur, including dwarf shrub and Carex bigelowii heathland, mossy Racomitrium grass heath, birch forest, and early-succession open habitats formed by unconsolidated fluvial sediments.

Vascular plant species richness was lowest in lodgepole pine plantations, with both birch woodlands and heathlands supporting significantly more species (Fig. 5A). Statistical analysis confirmed these differences (Kruskal-Wallis rank sum test: H = 24.31, df = 2, p = 5.27 × 10⁻⁶). Pairwise comparisons using Dunn’s test revealed significant differences between birch and pine (p = 2.73 × 10⁻⁶) and between heath and pine (p = 0.02), but not between birch and heath (p = 0.10).

Figure 5.

Violin plots showing the number of vascular plant species recorded in vegetation plots (A) and values of Shannon diversity index (B) in three different vegetation types: birch woodland, heath and lodgepole pine plantation. Points denote values of direct measurements, diamonds denote median value for each vegetation type.

Similar patterns were observed in Shannon diversity index values, which were lowest in lodgepole pine plantations (Fig. 5B). Differences between vegetation types were statistically significant (Kruskal-Wallis rank sum test: H = 25.46, df = 2, p = 2.96 × 10⁻⁶). Dunn’s test showed significant differences between all vegetation types: birch vs. heath (p = 0.03), birch vs. pine (p = 1.35 × 10⁻⁶), and heath vs. pine (p = 0.049).

The invasion process generally follows a predictable trajectory, irrespective of taxonomic identity of the species (Shigesada and Kawasaki 2001). Following establishment, there is typically a lag period with slow spread, after which the invader enters an exponential growth phase. This expansion continues until available space is saturated, at which point the spread rate levels off (Shigesada and Kawasaki 2001; Arim et al. 2006).

Metrics for lodgepole pine in Steinadalur reflect an accelerating spread, particularly over the last decade. The mean spread rate increased nearly eightfold, from 8.5 m/year over the first 25 years (1985–2010) to 61.6 m (2010-2021). Occupied area expanded almost tenfold, and tree density in belt transects increased nearly eightfold between 2010 and 2021. These rates align with models of exponential, not linear, growth, strongly suggesting that lodgepole pine in Steinadalur has entered the exponential growth phase.

This raises the question: are these patterns primarily driven by Malthusian population growth in unsaturated environment, or do environmental changes, such as climate warming, play a role? Across Europe, the Normalized Difference Vegetation Index (NDVI) has shown a positive trend over the last 30 years (Eisfelder et al. 2023). In Iceland, this trend is pronounced in the west and north but weaker in the east and southeast. These changes are attributed to higher temperatures, increased rainfall, and reduced summer grazing by free-range sheep (Raynolds et al. 2015), although disentangling causal factors remains challenging. Since the 1980s, Iceland’s mean temperatures have risen by 0.47 °C per decade, about three times the global average (Hanna et al. 2004; Björnsson et al. 2018).

Steinadalur, located in the southeast of the country, benefits from a milder climate, longer growing season, and higher rainfall than other regions (unpublished data from the Icelandic Met Office). Warm temperatures likely facilitated the lodgepole pine’s growth and expansion, and ongoing warming trends are expected to favour it further. Sheep graze in Steinadalur during summer but their impact on the pine has not been documented. The species’ altitudinal range, reaching 170 m by 2021, indicates that temperature constraints are unlikely to limit its spread.

The Steinadalur case study illustrates that lodgepole pine can colonise not only eroded or sparsely vegetated land but also areas with closed vegetation. The lodgepole pine plantation in Steinadalur was established in areas previously occupied by heath or birch woodland. In Iceland, birch woodlands and forests represent the most structurally complex native vegetation. These birch ecosystems vary from old forest fragments with tall, monocormic trees and dense ground layers of graminoids and broad-leaved dicots to open woodlands dominated by polycormic shrub-like birch and dwarf shrubs (Ottósson et al. 2016). According to our results, lodgepole pine is likely to invade most low-stature vegetation types, particularly in warmer lowland regions of Iceland.

Native lodgepole pine forests in North America are characterised by limited undergrowth (Eyre 1980; Perry et al. 2008). When planted outside their native range, lodgepole pine consistently reduces the species richness and diversity of native vegetation (Ledgard and Paul 2008; Urrutia et al. 2013). In New Zealand, vascular plant species richness in grasslands invaded by lodgepole pine declined from 38 to seven species within 20 years (Ledgard and Paul 2008), with none of the remaining species being native. Similarly, in British Columbia, the oldest lodgepole pine forests exhibited the lowest plant species diversity (Sullivan 2004). At the ecosystem level, lodgepole pine invasion constitutes a major state shift. Dense, fast-growing stands drastically reduce sunlight penetration, gradually eliminating shade-intolerant species associated with subarctic heathlands, grasslands, and birch woodlands. In Steinadalur, although based on a small sample size, comparisons of vegetation types strongly suggest significant impacts on native species composition and vascular plant species richness, likely due to light limitation as dense canopies develop, eliminating low-growing and light-demanding species. The significant reductions in species richness observed are clear indications of profound ecosystem changes.

A specific concern in Iceland is the impact of lodgepole pine on native bird populations, particularly wading birds. Iceland is a critical breeding area for waders in Europe (Gunnarsson 2020). Pálsdóttir et al. (2022) documented significant declines in wader densities near plantation edges, likely indicating population losses rather than shifts in habitat use. According to their study, estimated losses from plantations may already amount to tens of thousands of birds, underscoring the far-reaching impacts of tree plantations on Iceland’s ecosystems.

In addition to biodiversity loss, lodgepole pine invasion can result in reduced surface streamflow, heightened fire risks, soil erosion following clearcutting, destabilized riverbanks, and a decline in recreational opportunities and grazing land for livestock (De Wit et al. 2001). Managing invasive woody species like lodgepole pine is challenging, with removal often failing to restore ecosystems to their previous state (Panetta 2012; Sapsford et al. 2020). Conifer invasions also alter soil chemistry, hydrology, and fungal communities, potentially increasing soil CO₂ emissions and reducing albedo, which exacerbates warming in temperate and cold regions (Popkin 2019; Nuñez et al. 2021).

Most criteria for assessment of vulnerability to IAS apply to Iceland: 1) it is an isolated oceanic island (Pyšek et al. 2020; Dueñas et al. 2021), 2) has a natural disturbance regime that periodically creates open ground, and 3) native ecosystems that suffered severe destruction and degradation in the wake of human settlement (Richardson et al. 1994; Barrio and Arnalds 2022). Finally, the indigenous flora mostly comprises low-growing shrubs and herbs, i.e. growth forms very different from that of the invader (here lodgepole pine, Richardson and Bond 1991). Among the many invasive strengths of lodgepole pine are the high dispersibility of its seeds and frequent copious seed output (Richardson et al. 1994). The future spread of lodgepole pine in Iceland is likely to be significantly accelerated by the widespread practice of establishing numerous small plantation projects across the country’s lowland regions.

Lodgepole pine in Iceland fulfils the scientific definitions of an invasive species. Its spread rates, exceeding 100 m in less than 50 years in Steinadalur, and significant negative impacts on biodiversity, comply with IAS criteria (Richardson et al. 2000; Pyšek et al. 2004). Established outside its original plantation for approximately 35–40 years, it has reached distances of nearly 3 km and demonstrated exponential growth. Research is needed to understand its effects on trophic networks, belowground biota, bird and mammal populations, and landscape-scale homogenization. Regional-scale studies are critical, as lodgepole pine appears to be spreading across Iceland, even in colder northern regions (Brynjólfsson 2022).

Perceptions of the reality and magnitude of the threat invasive species may pose are known to vary greatly among social groups (García-Llorente et al. 2008), not least when benefitting stakeholders differ from those concerned with negative impacts (Novoa et al. 2024). The failure of many countries to implement successful management and control practices has been attributed to little public and political awareness (Bertolino and Genovesi 2003), and lack of cohesion between scientific researchers, the commercial sector, and policy makers (Stokes et al. 2006). The emerging consensus is that effective management of invasive non-native species largely depends on the active support and collaboration of all relevant stakeholders (e.g. Brundu et al. 2020).

In Iceland, attitudes toward potentially invasive conifers are deeply polarised, with a sharp divide between those anticipating direct benefits and those expressing concerns about environmental impacts. Academics, biologists at state and regional institutes and environmental associations, have warned against indiscriminate use of introduced species and the widespread planting of lodgepole pine (Von Schmalensee 2010; Bjarnason et al. 2023; Jónsdóttir 2023). In contrast, the forestry sector, as well as national and regional afforestation associations have remained staunch advocates of continuing large-scale planting of lodgepole pine and dismiss it as a potential threat to Iceland’s biodiversity. The statements and arguments presented in Icelandic media and forestry publications reflect this divide. Claims include rejecting the term “invasive alien species” as invalid and arguing that concerns about invasive species are logically flawed (Sigurgeirsson 2014). Some maintain that lodgepole pine is not invasive (Eysteinsson 2021), that trees cannot become invasive, or that unwanted trees can be easily removed (Eysteinsson 2019). Invasion research has been portrayed as pseudoscience driven by nationalism (Sigurgeirsson 2005; Gardarsson 2022; Eysteinsson 2023). The most pressing task for Iceland is to enhance knowledge through rigorous scientific studies that provide a solid foundation for assessing the risks and long-term consequences of large-scale lodgepole pine cultivation. This evidence should facilitate informed dialogue among stakeholder groups.

The authors have declared that no competing interests exist.

No ethical statement was reported.

This research was supported by a grant from Kvískerjasjóður, awarded to Pawel Wasowicz in 2021.

Conceptualization: PW. Data curation: PW. Formal analysis: GÓ, PW. Funding acquisition: PW. Investigation: GÓ, PW. Methodology: GÓ, PW. Project administration: PW. Resources: PW. Supervision: PW. Validation: PW. Visualization: PW. Writing - original draft: PW, ÞEÞ, GÓ. Writing - review and editing: GÓ, PW, ÞEÞ.

Pawel Wasowicz https://orcid.org/0000-0002-6864-6786

Guðrún Óskarsdóttir https://orcid.org/0000-0001-8128-8306

Þóra Ellen Þórhallsdóttir https://orcid.org/0000-0003-2946-5963

All of the data that support the findings of this study are available in the main text or Supplementary Information.