Aggregated species richness and habitat heterogeneity variables for testing the habitat-heterogeneity hypothesis, 2006-2018

Theory: Heterogeneity within a habitat comprises many different biotic and abiotic aspects. In a previous review (Stein & Kreft 2015, Biological Reviews), habitat heterogeneity was divided systematically in different subject areas: the biotic subject areas included land cover and vegetation, and the abiotic ones included climate, soil and topography. We adjusted this classification scheme system by restricting their subject areas to those relevant to our within-forest stand heterogeneity (thus excluding land cover), those for which heterogeneity data was available (thus excluding climate and soil, as measures on these properties were not replicated within plots for all of our study sites) and by adding dead wood as subject area. Both subject areas living vegetation and dead wood were then further divided into structural as well as species richness aspects. Finally, we considered in our analyses horizontal and vertical aspects of vegetation heterogeneity. Ultimately, six aspects of habitat heterogeneity were addressed: taxonomical diversity of dead wood, structural diversity of dead wood, plant diversity, heterogeneity in vertical structure and in horizontal structure as well as topography. References: Stein, A. & Kreft, H. (2015). Terminology and quantification of environmental heterogeneity in species-richness research. Biol Rev Camb Philos Soc, 90, 815–836. doi: 10.1111/brv.12135. Epub 2014 Aug 7.|Measurements Type: The data set comprises per plot both aggregated species richness and abundance measures as well as environmental data (diverse measurements based on Airborne Laser Scanning, phylogenetic and functional diversity of plants, richness of dead wood species and structures) Measures used in the main analysis: Species data used BExIS datasets: 20826; v1.4.5; Vegetation Records for Forest Eps, 2009-2016 4460; v1.11.15; Lichen diversity in forests, 2007-2008 4141; v1.6.10; Bryophyte diversity in relationship to forest management types in forests, 2007-2008 18547; v1.2.2; Deadwood inhabiting fungi presence absence, 2010 16868; v1.2.4; Window and ground traps on forest Eps in 2008 subset Araneae 16867; v1.2.4; Window and ground traps on forest Eps in 2008 subset Hemiptera 16866; v1.2.4; Window and ground traps on forest Eps in 2008 subset Coleoptera 21446; v3.1.4; Bird survey data 2008 21447; v3.2.1; Bird survey data 2009 21448; v3.1.3; Bird survey data 2010 19848; v1.1.4; Bat activity in all Exploratories, summer 2008, using acoustic monitoring 19849; v1.1.4; Bat activity in all Exploratories, summer 2009, using acoustic monitoring 19850; v1.1.5; Bat activity in all Exploratories, summer 2010, using acoustic monitoring for the other two projects, see Steigerwald-Project Doerfler, I., Gossner, M.M., Müller, J., Seibold, S. & Weisser, W.W. (2018). Deadwood enrichment combining integrative and segregative conservation elements enhances biodiversity of multiple taxa in managed forests. Biol. Conserv., 228, 70–78. and further, unpublished data which have been provided by Jörg Müller BIOKLIM-Project Bässler, C., Müller, J. & Dziock, F. (2010). Detection of Climate-Sensitive Zones and Identification of Climate Change Indicators: A Case Study from the Bavarian Forest National Park. Folia Geobot., 45, 163–182. Bässler, C., Müller, J., Dziock, F. & Brandl, R. (2010). Effects of resource availability and climate on the diversity of wood-decaying fungi. J. Ecol., 98, 822–832. Moning, C., Werth, S., Dziock, F., Bässler, C., Bradtka, J., Hothorn, T., et al. (2009). Lichen diversity in temperate montane forests is influenced by forest structure more than climate. For. Ecol. Manag., 258, 745–751. Müller, J. & Brandl, R. (2009). Assessing biodiversity by remote sensing in mountainous terrain: the potential of LiDAR to predict forest beetle assemblages. J. Appl. Ecol., 46, 897–905. Müller, J., Mehr, M., Bässler, C., Fenton, M.B., Hothorn, T., Pretzsch, H., et al. (2012). Aggregative response in bats: prey abundance versus habitat. Oecologia, 169, 673–684. Müller, J., Moning, C., Bässler, C., Heurich, M. & Brandl, R. (2009). Using airborne laser scanning to model potential abundance and assemblages of forest passerines. Basic Appl. Ecol., 10, 671–681. Raabe, S., Müller, J., Manthey, M., Dürhammer, O., Teuber, U., Göttlein, A., et al. (2010). Drivers of bryophyte diversity allow implications for forest management with a focus on climate change. For. Ecol. Manag., 260, 1956–1964. and further, unpublished data which have been provided by Jörg Müller To gain comparable species data across projects and regions, for each region only the data of year closest to the LiDAR-flights was chosen. The Biodiversity Exploratories used transect walks to record bats with a bat detector while in the other two areas fixed autonomous batcorders for single nights were used. Birds were monitored during breeding season acoustically and visually within a fixed time span. Arthropods were collected with pitfall, crossed flight-interception traps and low intensity light traps. Fungi, bryophytes and lichens were mapped on dead wood objects and plants on a fixed area. For each plot, species richness was summed over the whole collection period. Abundance (or activity) data were available for animals in all five regions. For bryophytes, fungi and plants we only used species richness, as the three projects used differing techniques for the estimation of abundance. Environmental data used BExIS datasets: 15386; v1.1.2; Dead Wood Inventory 2012 The taxonomical diversity of dead wood was calculated by counting the number of different tree species represented on a plot. The structural aspect was calculated by counting the number of different deadwood types classified by the diameter classes, the decomposition classes and deadwood types. We characterized plant diversity by calculating Faith’s PD considering all vascular plants in the tree, shrub and herb layers which have been determined in course of biodiversity assessments. Faith’s PD was calculated based on the phylogeny of Durka and Michalski (2012). We used high resolution Airborne Laser Scanning (ALS) to gain standardized measurements of the complex 3D structure of the forests. As a measure of vertical heterogeneity, we choose the standard deviation of vegetation height returns. For horizontal heterogeneity, we classified our plots in gap and non-gap area. We defined areas which had a minimum size of 50 m², a perimeter/area ratio under < 1.5 (thus excluding narrow linear structures as forest aisles), a height threshold of 2 m and a penetration ratio of more than 80% as gaps. Furthermore, we used the standard deviation of the slope with a grain size of 1 m x 1m across our one-hectare plots as a measure of topographic heterogeneity References: Durka, W. & Michalski, S.G. (2012). Daphne: a dated phylogeny of a large European flora for phylogenetically informed ecological analyses. Ecology, 93, 2297–2297. https://doi.org/10.1890/12-0743.1|instruments: Environmental measurements: Riegl Q560/VQ780i Biodiversity assessments: Pettersson D 1000x ultrasound detector (Pettersson Electronic AG, Uppsala, Sweden) Batcorder 2.0 bat detector (ecoObs GmbH, Nurnberg, Germany) Observations by hearing & sight tape measure, knife, pencil & eye 12V, 15W super-actinic light tubes, bucket and chloroform For more details see Appendix S1&S2 in Heidrich et al (2019)|procedures: The ALS datasets from the five regions were pre-processed and the metrics derived using the same methods for all datasets. The pre-processing was performed using LAStools (“LAStools” 2012). This included transformation of the raw data into LAZ file format (txt2las, las2las), coordinate transformations into a unified coordinate reference system, removal of isolated returns (lasnoise) and retiling of the point cloud into 500 x 500m tiles (lastile). Classification of returns into ground, vegetation and buildings was done using lasground and lasclassify. The elevation of points was normalized w.r.t. above ground level (AGL) using lasheight. From the ground returns, a DTM with 1m spatial resolution was generated using the blast2dem function. For the gap analysis, pit free canopy height models (CHM) with a spatial resolution of 1m were created following the concept of Khosravipour et al. (2014) which was implemented in the lidR R-package (Roussel et al. 2019). For the vertical heterogeneity, we calculated standard deviation (SD) of all the heights from the vegetation returns. The horizontal heterogeneity of the plots was analysed based on canopy gap masks. Gap masks were created from the normalized point cloud by calculating the penetration rate from the top of the canopy down to 2m for 1x1m raster cells. Here, penetration rate is the ratio of all returns with Z>2m to all returns <=2m. Cells with a penetration ratio > 80% were classified as gap cells. Using connected component labelling we grouped connected gap cells to gap objects and calculated area perimeter ratio for each gap using the landscapemetrics R-package (Hesselbarth et al. 2019). To remove very small or narrow gaps (e.g. skidding trails) we deleted all gaps with an area < 50m² or a perimeter-area-ratio <1.5 from the gap maps before we calculated the aggregated statistics on the plot level. Topographic variation was described based on the DTM by calculating the SD of slope values. References: LAStools. (2012). rapidlasso GmbH. Roussel, J.-R., Auty, D., De Boissieu, F., Meador, A.S: (2019). lidR: Airborne LiDAR Data Manipulation and Visualization for Forestry Applications. Available at: https://cran.r-project.org/web/packages/lidR/index.html. Khosravipour, A., Skidmore, A.K., Isenburg, M., Wang, T. & Hussin, Y.A. (2014). Generating Pit-free Canopy Height Models from Airborne Lidar. Available at: https://www.ingentaconnect.com/content/asprs/pers/2014/00000080/00000009/art00003. For details see Appendix S1&S2 in Heidrich et al (2019)|