Spatial distribution of soil microbial biomass and respiration (H1)
Soil microbial biomass decreased with increasing distance from the trees in monospecific tree pairs. This could be explained on the one hand by higher water availability due to stemflow near the tree base, which can leach and transport nutrients and microorganisms from the canopy layer to the soils (Bittar et al. 2018). Soil moisture was shown to be important in many studies before (Cesarz et al. 2022; Schimel 2018; Serna-Chavez, Fierer, and van Bodegom 2013). High levels of soil moisture can increase soil enzyme activities, fluxes of soil nutrients, and oxygen availability (Brockett, Prescott, and Grayston 2012; Stark and Firestone 1995), and higher soil humidity can furthermore buffer possible negative changes in soil pH, suggesting it to be a key driver of soil microbial biomass (Cesarz et al. 2022). On the other hand it could also be explained by a higher rhizodeposition closer to the trees (Parker et al. 2017). Our findings would suggest the importance of forest density in modulating soil functioning.
As expected, we also found a negative effect of soil depth on both microbial properties. This is in line with previous findings where a lower amount of carbon and nutrients was found in deeper soil layers, as the main decomposition happens in the leaf litter cover and top soil layers (Goebes et al. 2019; Jobbágy and Jackson, n.d.; Prescott and Grayston 2013). Additionally, deeper soil layers often have a decreased amount of oxygen, soil water content, and contain less plant root biomass (Engelhardt et al. 2018; Fall et al. 2012; Serna-Chavez, Fierer, and van Bodegom 2013). Thus, the present results at the small scale are in line with previous findings at the larger scales, where soil organic carbon decreased with increasing soil depth and distance to trees (Rabearison et al. 2023). These similar results suggest that understanding interaction effects at small scales have the potential to be upscaled.
Spatial distribution of belowground overyielding (H2 - H3)
Our study showed no overyielding across all soil core positions and depth layers. However, we found significant differences between the soil layers for both soil microbial functions. Microbial biomass showed, on average, higher overyielding in the shallower soil (0-5 cm), whereas microbial respiration showed higher overyielding in the deeper soil (5-10 cm). The BEF-China experimental Site A was established in 2009 after a clear-cut of the previous plantation (Yang et al. 2013). Plant diversity effects on soil organic matter have been shown to become stronger in the topsoil layer over time (Lange et al. 2023). These findings from experimental grasslands could suggest that the soil microbial biomass in forests is strongly related to the organic matter content in the topsoil layer (Beugnon, Eisenhauer, et al. 2023) and that positive plant diversity effects could get stronger with stand age (Huang et al. 2018; Perles-Garcia et al. 2021). The increased microbial respiration in the deeper soils could suggest increased carbon sequestration by adding soil microbial necromass to the carbon pool (Buckeridge et al. 2020; Schmidt et al. 2011).
We expected heterospecific tree-tree interactions effects to be maximised in the middle between the planted trees. Contrary to our hypothesis (H3), we found an overyielding of soil microbial biomass and respiration close to L. formosana and underyielding close toS. saponaria , showing that microbes in mixed pairs perform better than expected close to L. formosana but less well than expected close to S. saponaria. The gradient from over- to underyielding of microbial respiration was less pronounced in the shallower soil than in the deeper soil. This could indicate that the presence of S. saponaria had a positive effect on microbes close to L. formosana and it is stronger in the deeper soil layer (5-10 cm), possibly through fine root exudates (Zheng, Wei, and Zhang 2017). Microbial respiration was less affected from the mixture than microbial biomass in the topsoil layer (0-5 cm). It was shown that the balsam of L. formosana contains acidic compounds, which were reported to be inhibitory for fungi (Chien et al. 2013). These could also be present in the leaf litter or root exudates (Öztürk et al. 2008) and inhibit microbial respiration more than microbial biomass. Together with a spatial distribution of litter in the heterospecific pairs (Beugnon, Eisenhauer, et al. 2023), this might lead to a small-scale change of soil pH. Soil pH was found to be a strong driver of microbial growth (Fierer and Jackson 2006), and additional pH measurements should be performed in future studies to better understand the opposing species identity effects of S. saponaria and L. formosana . It was shown that soil fungi and bacteria react differently to changes in soil pH: bacterial growth decreased with a more acidic pH, whereas fungal growth was shown to increase (Rousk, Brookes, and Bååth 2009). This might also explain the significant negative effect of L. formosana on microbial respiration in this experiment (Suppl. S2: Fig. S2). To better understand distribution patterns of microbial properties, belowground tree traits (e.g. specific root length, root diameter) should be taken into account. Recent studies could link them to carbon exudation and fine root density (Bergmann et al. 2020; Sun et al. 2021), as well as soil organic matter decomposition (Adamczyk et al. 2019).
The positive tree-tree interaction effect of the heterospecific tree pair on soil microbial biomass and respiration shows that neighbourhood effects are acting at small spatial scales, which could explain the inconsistencies of BEF relationships reported in previous forest studies (Beugnon et al. 2021; Cesarz et al. 2022; Li et al. 2019; Pei et al. 2016). Our results stress the need to standardize sampling methods by considering small-scale interactions to understand the mechanisms behind tree-soil interactions. In addition, measurements of soil microbial properties across a wider range of species transects are now needed to better understand tree-tree interactions in space and their biological drivers.