Consequences of climate-induced range expansions on multiple ecosystem functions

Species range shifts, driven by climate change and other factors, are reshaping community composition in many ecosystems. These shifts introduce novel functional traits and alter food web interactions, potentially impacting ecosystem processes. While short-term experiments offer insights, predicting the long-term consequences in natural systems remains challenging. This is partly due to the difficulty in observing multiple range expansions over extended timeframes, allowing observation of the complex dynamics that unfold. Long-term community data, combined with species functional traits (characteristics linking organisms and ecosystem processes), offer a promising approach to understand how compositional shifts modulate species' contributions to ecosystem function. This study focuses on a 30-year census of larval caddisflies, dominant detritivores in subalpine ponds. This dataset includes three sequential upslope range expansions by different caddisfly species, alongside significant changes in the abundance of resident species. The high interspecific trait variation within this assemblage provides an excellent system to investigate the impact of range expansions on multiple ecosystem processes, specifically nutrient cycling (nitrogen and phosphorus supply) and detritus breakdown. The study aims to examine how the relative contributions of dominant resident, subdominant resident, and range-expanding species to these processes change over time during successive range expansions, and how changes in caddisfly assemblage evenness influence functional redundancy. The central hypotheses were that successive range expansions would reduce the dominant resident's abundance and relative contribution to ecosystem processes, while increasing the abundance and relative contribution of range-expanding species, and that increased assemblage evenness would increase functional redundancy.

The literature highlights the widespread occurrence of climate-driven species range shifts and their potential to alter community composition and ecosystem function. Studies emphasize the need for long-term data and the integration of functional traits to predict the impact of compositional changes. Previous research has established the importance of functional traits in linking organisms and ecosystem processes, such as plant stoichiometry and animal nutrient excretion rates. However, these studies often lack the temporal scale to assess the long-term dynamics of range expansions and their cumulative impact on ecosystem processes. Other relevant literature demonstrates the importance of animal-driven nutrient supply and detritus breakdown in aquatic ecosystems, as well as the increasing recognition of intraguild interactions and resource limitation as influential factors in community dynamics. Finally, studies have explored the relationship between species evenness and functional redundancy, indicating that more even assemblages may offer greater resilience to environmental change.

This study utilized a 30-year (1989-2019) dataset of larval caddisfly assemblages from subalpine permanent ponds in the Mexican Cut Nature Preserve, Colorado. Annual censuses documented three sequential upslope range expansions by caddisfly species (*Limnephilus picturatus*, *Grammotaulius lorretae*, and *Nemotaulius hostilis*) and changes in the abundance of resident species (*Limnephilus externus*, *Asynarchus nigriculus*, and *Agrypnia deflata*). Species-specific functional traits, including nitrogen and phosphorus excretion rates and detritus processing rates, were incorporated to predict species-specific contributions to nutrient supply and detritus processing. A random sampling framework was implemented to account for variation in pond size and habitat availability, ensuring robust estimates of species contributions in an average permanent pond. Predicted species-specific ecosystem process contributions were calculated as the products of average density, final instar mass, larval development rate, and nutrient excretion or detritus processing rate, with these individual rates randomly sampled from previously determined probability distributions. The relative contributions were then calculated to standardize for interannual variation in assemblage totals. Mixed effects models were used to analyze trends in species abundance and relative contributions to ecosystem processes during each range expansion period, accounting for pond-level variation and temporal autocorrelation. Linear contrasts were used to test hypotheses regarding trends in species’ abundances or groups’ relative contributions during the range expansions. Pielou's evenness index was used to quantify assemblage evenness, and simple linear models examined the relationship between evenness and the aggregate variability of species' contributions to ecosystem processes. Aggregate variability was calculated as the sum of species' among-pond variance in process contributions and twice the sum of contribution covariances among all species pairs.

The study revealed that species' relative abundances and contributions to ecosystem processes changed significantly throughout the three range expansions. The dominant resident species, *L. externus*, initially maintained its large relative contribution to phosphorus supply and detritus processing even after the first and second range expansions. However, the third range expansion (*N. hostilis*) led to significant declines in *L. externus*'s abundance and relative contributions to phosphorus supply and detritus processing. Conversely, *N. hostilis*'s relative contribution to phosphorus supply increased substantially. Interestingly, the subdominant resident species, *Ag. deflata*, consistently made significant relative contributions to nitrogen supply throughout all range expansions. Total ecosystem process rates, however, did not change significantly throughout the range expansions. The analysis of assemblage evenness revealed a decline over time, contrary to expectations, while aggregate variability in species contributions to ecosystem processes decreased with increasing evenness. This implies that changes in the relative abundance of the dominant resident, and consequent evenness changes, can greatly impact the assemblage-level contributions to ecosystem processes. Further investigation found that *N. hostilis* exceeded *L. externus*'s relative abundance and ecosystem process contributions far more frequently than the previous range-expanding species. This is possibly due to its unique fall hatching and development phenology, leading to reduced intraguild competition and predation pressure.

The findings highlight the complex interplay between climate-induced range expansions and ecosystem function. While total ecosystem process rates remained relatively constant, significant shifts occurred in the relative functional roles of resident and range-expanding species. The observation that changes in relative contributions can precede changes in total rates emphasizes the importance of considering species-specific roles. The dominant resident species' ability to regulate ecosystem processes was initially unaffected by the first two range expansions, emphasizing the resilience of established species and the potential for functional redundancy among subdominant species. However, the third range expansion illustrated the capacity of a range-expanding species with a unique life history to displace the functional role of a resident species, underscoring the importance of considering species traits beyond abundance and functional traits. The unexpected decline in evenness raises questions about the long-term impacts of range expansions on community structure and stability. This contrasts with the expectation of increased evenness and redundancy following multiple range expansions. The relatively unchanging total ecosystem process rates suggest constraints imposed by factors such as energetic equivalence or resource availability. Future research should address the possible role of other taxonomic groups in buffering the effects of caddisfly range expansions, considering the contributions of non-caddisfly taxa to nutrient cycling.

This study demonstrates that climate-induced range expansions can significantly alter the relative functional roles of species in an ecosystem, even without affecting overall process rates. The success of a range-expanding species in functionally displacing a resident species depends on a complex interplay of factors such as abundance, functional traits, and life history characteristics. The unexpected decline in evenness highlights the complex nature of community dynamics following range expansions. This study emphasizes the importance of long-term data and integration of species' functional traits and life history strategies in predicting the consequences of range shifts on ecosystem function. Future research could explore the generality of these findings in other systems and investigate the influence of other environmental factors on the dynamics observed.

The study is limited to caddisfly assemblages, potentially overlooking the contributions of other taxonomic groups to ecosystem processes. The focus on permanent ponds might not fully reflect the dynamics in other pond types. The statistical methods used assume linearity, which might not fully capture complex interactions between species and environmental factors. Furthermore, the observed changes in relative contributions might be transient, requiring longer-term monitoring to confirm sustained shifts in functional roles.