Abstract :
[en] The Anthropocene is characterized by unprecedent environmental transformations driven by human activities, including industrialization, agricultural intensification, and deforestation. These changes have deeply altered biogeochemical cycles, climate, and air quality, resulting in a dramatic global decline in biodiversity and disastrous impacts on human health. Among the major contemporary threats, climate change and air pollution are two of the main drivers of species loss. However, their respective effects are difficult to assess, as they both originate from similar anthropogenic sources and interact across spatial and temporal scales.
In this thesis, we aim to document and disentangle the respective impacts of air quality changes and climate change on epiphytic bryophyte communities across space and time. We first assess how improvements in air quality and concomitant climate change between 1980 and 2020 have shaped the spatio-temporal turnover of epiphytic bryophyte communities, evaluating the relative roles of major air pollutants (e.g., SO₂, NO₂, O₃, particulate matter) and climatic conditions. We further investigate whether air pollution continues to affect current epiphytic bryophyte distribution, identifying the specific pollutants involved—including particulate matter, black carbon, ammonium, heavy metals, and pesticides—and disentangling their effects from background environmental factors such as climate, topography, and forest structure.
The spectacular expansion of epiphytic bryophytes in southern Belgium since 1980 is characterized by increasing species frequency, species richness and the emergence of newcomers. Temporal beta diversity was twice higher as extant spatial beta diversity, indicating that community composition differed more at different time periods within the same site than among sites. The spatio-temporal variation in species composition reflected a clear time-series and not a geographic clustering. Variance partitioning analyses demonstrated that this temporal turnover was almost exclusively driven by the improvement in air quality and not by climate change. The sharp decline in SO₂ and NO₂ concentrations since the pollution peaks of the 1970s–1980s allowed acid-sensitive species to recolonize formerly polluted areas, while acidophilous species declined.
Although major pollutants such as SO₂ and NO₂ no longer appear to affect epiphytic bryophyte communities, other pollutants may have emerged. O₃ was identified as the main driver of extant species distributions. Its concentrations are, however, strongly correlated with climatic and land-use conditions, suggesting that it may not necessarily have an actual ecotoxicological impact but instead reflects an urban to rural gradient. The second most important pollutant identified was NH₃. Although the concentrations of the latter are slightly decreasing, NH₃ acts as a base, potentially counter-acting the acidifying effect of historical pollutants and contributing to the decrease of acidophilous species and the increase of nitrophilous species. The potential effects of other pollutants, such as pesticides or heavy metals, remain insufficiently documented, emphasizing the need for improved environmental monitoring.
The lack of any signature of climate change in the recovery of epiphytic bryophyte floras is surprising given the reliance of bryophytes on precipitation for water uptake and the sensitivity of temperate species to moderately warm temperatures. While climatic conditions are commonly assessed at a regional scale from standardized weather stations, organisms, however, actually experience local conditions shaped by topography, vegetation structure, solar radiation, and wind exposure. This is particularly true for small-size organisms like bryophytes, calling for a shift from a macro to a microclimatic perspective. The difference between macroclimate (open-field) and (in-situ) microclimate is called the microclimatic effect. In forest ecosystems, canopy cover typically generates a buffering effect that stabilizes temperature and humidity. Temperature and relative humidity were recorded in-situ for one year at 42 stands of Quercus-Fagus forests in southern Belgium. The microclimatic effect was characterized at the level of each sensor through the “slope and equilibrium” approach and its variation across sensors was modelled using spatially explicit variables of topography and vegetation structure obtained by satellite imagery. The slope of the linear relationship between macro and microclimate accurately characterized the microclimatic effect, evidencing both buffering and amplification for temperature and only buffering for relative humidity. The models to map the microclimatic effect using topography and vegetation structure as predictors performed, however, poorly. Increasing the number of dataloggers measuring microclimate in-situ to calibrate the models and adding new predictors, such as vegetation structure variables derived from LiDAR, appear as the next steps to increase model performance. Species response curves to the microclimatic effect allowed for a quantitative characterization of species indicator values for forest microclimates and their drivers. Overall, this preliminary study sets the premises towards a fine-scale mapping of forest microclimates and the characterization of species responses to microclimatic effects, opening the door to the identification of microrefugia and the formulation of recommendations for forest management under climate change.
