Which factors affect landscape diversity

Precipitation was not significantly related to these variables. There was little correlation among variables of different groups delimited with bold black lines in Fig 2. Models containing 1—3 of the eight explanatory variables relating to diversity, environmental context and study design explained on average In the case of leaf area index, tree species richness log-transformed was among the most important predictors and explained, on average, Depending on sequential order, i.

The only variable exceeding the predictive power of tree species richness was the study design variable biogeographic region average R 2 : The coefficients of determination R 2 for models that include all combinations of 1—3 of all 8 predictors listed on y-axis are on average The R 2 contributions of every predictor when averaged over all possible model-orderings are indicated for each model with light-grey dots.

Results that were averaged across all models with 1—3 predictors are shown in black. Black dots represent average R 2 contributions over all possible model-orderings and black lines indicate the range of R 2 contributions when included first or last in the model, respectively.

A near-identical picture was present for growing-season length, with a similar amount of variance explained by the logarithm of tree species richness average R 2 : The main difference in the analysis of growing-season length compared with the analysis of leaf area index was that altitude and biogeographic region explained a very large fraction of the overall variance average R 2 : This is to be expected because growing-season length strongly depends on temperature.

The logarithm of tree species richness explained more variance in leaf area index and growing-season length than the number of trees average R 2 LAI: 0. Land-cover richness, landscape structure represented by edge density and stand structure represented by the number of trees only explained a marginal amount of variance and were not important in any model Fig 3.

We calculated effect sizes r partial of tree species richness log-transformed in separate models that each first adjusted for i. In the case of growing-season length, log-transformed tree species richness effect sizes were relatively similar in magnitude in most models range r partial : 0. Effect sizes of log-transformed tree species richness in these models always exceeded 0. Tree species richness effects on growing-season length in these models were even larger than the effects on leaf area index range r partial : 0.

Effect sizes for land-cover richness generally were low and insignificant for leaf area index and growing-season length Table 1 , although they increased in models of growing-season length when additionally adjusting for biogeographic region.

In these models the r partial of land-cover richness ranged from 0. We investigated the effects of local and landscape-level diversity and environmental context variables on forest functioning in managed landscapes. We found that local tree diversity, measured as the logarithm of species richness, was positively related to forest leaf area index and growing-season length.

Tree diversity was among the most important predictors for these functions compared with other variables related to landscape structure land-cover edge density , climate annual precipitation , topography mean altitude and stand structure number of trees. Local tree diversity effects were relatively robust, with little confounding with the environmental context variables.

Overall, our findings thus support all three hypotheses stated in the Introduction in the case of local tree diversity, and suggest that it is an important driver of forest productivity and phenology, with effects that remain relatively constant across the range of environmental conditions encountered in managed landscapes. In contrast, landscape-level land-cover richness did not show a strong direct relationship with forest productivity and phenology and hence, we found no support for the hypotheses stated in the Introduction in the case of landscape diversity.

However, land-cover richness strongly positively correlated with local tree species richness. It may thus be that landscape diversity contributes to local ecosystem functioning indirectly through effects mediated by local tree species diversity.

These effects of canopy packing can be predicted if co-existing species differ in crown architecture either genetically or via neighborhood driven plastic responses in crown growth and vertical leaf distribution [ 49 , 53 , 55 ].

Whether local diversity effects on basal area were truly absent or whether unaccounted drivers masked these, remains unclear.

Swiss forests are often managed by removing individual trees from stands, avoiding clear-cutting large areas. It may be that differences in management history among plots had a long lasting effect on stand basal area, whereas leaf area recovered faster from such interventions.

Independent of these considerations, our results suggest that productivity measured by leaf area index does not necessarily reflect productivity measured by woody biomass and that these two attributes are likely governed by different mechanisms. Hence, species richness effects on woody biomass production [ 59 — 61 ] might differ from species richness effects on the leaf area in forest stands.

In our study, tree species richness was positively correlated with both leaf area index and growing-season length. Correlation coefficients were similar for both variables but the relationship was only marginally significant for growing-season length.

The reason for the lower statistical power in the case of growing-season length is that only 22 of the 36 forest sites showed a seasonal pattern in light attenuation S1 Fig. This raised the possibility that effects of tree species richness on growing-season length might actually have been altitude effects in disguise.

However, we found that the variation in growing-season length across biogeographic regions masked tree diversity effects. These increased substantially in magnitude and significance after we accounted for biogeographic region.

Our findings of positive plot-level species richness effects on growing-season length parallel our earlier findings for mixed ecosystems including large amounts of non-forest ecosystems at the landscape scale [ 62 ]. Phenology plays an important role for species interactions [ 63 ] and the capability of communities to adapt to environmental change [ 28 , 64 ]. Hence, biodiversity might be important for the resilience of communities faced with global change under real-world conditions in complex landscapes.

There is conflicting evidence whether landscape context is important for local ecosystem functioning. In agricultural grasslands and crop fields, landscape heterogeneity can increase local biological control [ 13 , 16 ].

In tropical dry forest, landscape structure was less important for local aboveground biomass [ 65 ]. In the present study, land-cover richness did not directly affect our two measured local ecosystem functions.

However, the close correlation between land-cover richness and local tree species richness indicated that land-cover richness may have had a indirect effect on local ecosystem functioning via an effect on tree species richness.

This would be in accordance with the idea that environmental heterogeneity at larger scales drives local species richness [ 66 ]. Indirect effects of landscape diversity on local ecosystem functioning via positive effects on local biodiversity have indeed been proposed in theoretical studies [ 15 , 67 ] and are empirically supported in grassland and agricultural areas [ 13 , 16 — 19 ] as well as in forests [ 65 , 68 ].

Possible mechanisms include that landscape diversity increases the regional taxonomic or functional diversity via an increased range of spatially dissimilar environmental conditions [ 66 ]. Increased regional diversity could promote local biodiversity and ecosystem functions via the spatial insurance effect [ 67 ], by which local ecosystems may be colonized by species with well-suited traits, or by species that would otherwise go locally extinct [ 15 ].

For example, it has been found that adverse effects of local land-use intensification on biodiversity and ecosystem functions were masked by the spillover of species from the surrounding landscape [ 13 , 16 ].

In our study, the effects of log-transformed tree species richness stayed relatively constant when adjusting for different environmental context variables at the local and the landscape scale.

This suggests that environmental context does not alter the local relationship between biodiversity and ecosystem functioning. This reasoning is supported by the fact that we did not find significant interactions between log-transformed tree species richness and environmental context variables on leaf area index or growing-season length when we included such interaction terms in further exploratory analysis, using the general linear models described in the Materials and methods section.

The only exceptions were two significant interactions between log-transformed species richness and forest type or the number of trees on growing-season length. However, these interactions were essentially due to coniferous forest plots that had low numbers of trees and for which season length was ill-defined.

Generally, the potential dependence of biodiversity effects on environmental context remains poorly tested to date [ 3 , 6 ]. Experiments suggest that there is no such dependency [ 69 , 70 ]; but see [ 71 ] , whereas observational studies often suggest the contrary [ 59 — 61 , 72 — 75 ]. However, cause and effect of diversity were often not separated in those latter studies.

To conclude, we show that local tree species diversity is a powerful predictor of local forest functions in managed real-world landscapes, similarly to the effects of species richness that have been found in BEF experiments. Landscape diversity had only a low explanatory power, but was positively correlated with local tree species diversity and could thus indirectly affect local forest functions.

A general challenge in observational studies is to disentangle causes and effects of biodiversity [ 6 ]. Whereas knowing the causes of biodiversity enables the identification of conditions that facilitate biodiversity, it is only knowledge about the effects of biodiversity that enables an assessment of the consequences of biodiversity loss for ecosystem functioning and ultimately, human well-being.

A : Temporal development of average daily illuminance kilolux; i. B: Light attenuation by the forest canopy derived from the smoothed ratio of average daily i. We defined yearly start of the season SOS as the day of the year when light attenuation first exceeded the mean of its annual minimum and maximum value horizontal dashed line. Similarly, we derived the yearly end of the season EOS as the last day of the year before light attenuation fell below this threshold. We derived these metrics only for forests with a clear seasonal pattern in light attenuation, which we determined by applying the following restrictions: A minimum GSL of 60 days, a minimum growing season amplitude of 0.

In total, we obtained GSL values for a subset of 22 out of the 36 forest sites. Principal component analyses PCA of variables related to landscape structure A , climate and topography B and local stand structure C. A: Edge density ED; total length of borders between different land-cover types divided by the total landscape area , patch density PD; the number of different land-cover patches divided by the total area of the landscape and land cover richness LR; the number of different land-cover types closely clustered together and strongly differed in their loadings compared to the fraction of forest cover Ffrac and the connectivity of forest patches Fconn for the first principal component PC1.

B: Temperature temp and altitude alt showed the strongest negative covariance in loadings for PC1, indicating a strong temperature-altitude gradient in our dataset. We identified a second important gradient defined by precipitation precip that positively co-varied with slope and negatively co-varied with the northerly aspect N-aspect in their loadings for the second principal component PC2.

C: Our proxy for stand age and demographic structure DBHmax average diameter at breast height of the three largest trees clustered closely with cumulative basal area BA and both of these measures co-varied negatively with the number of trees ntrees in their loadings for PC1. We used information on the typical species composition of the 30 main Swiss forest types [ 34 ] to associate sites of the Swiss biodiversity monitoring program BDM; biodiversitymonitoring.

List of all the tree species we found in the tree inventory, the number of individuals across all study sites and number of study sites with a respective species present. Andy Hueni helped with the calibration of the light sensors. We acknowledge Daniel Trujillo and Richard Baxter for their help with collecting field data. Browse Subject Areas? Click through the PLOS taxonomy to find articles in your field. Abstract Research of the past decades has shown that biodiversity is a fundamental driver of ecosystem functioning.

Introduction Biodiversity-ecosystem functioning BEF studies of the past decades have revealed that biodiversity is an important driver of ecosystem functioning [ 1 ]. Materials and methods Study design Using 36 forest stands, we analyzed the relationship of local ecosystem functioning measures to local and landscape-level variables characterizing diversity and environmental conditions.

Download: PPT. Local tree diversity and stand structure We marked all trees in each inventory plot with a diameter at breast height DBH of at least 5 cm and determined their species identity.

Local ecosystem functioning Productivity. Data analysis We first assessed the overall correlation coefficients r among variables characterizing local and landscape-level diversity, landscape structure, topography, climate, stand structure and forest ecosystem functioning.

Fig 2. Correlations among local and landscape-level predictors and local ecosystem functions. Variance explained by local and landscape-level predictors Models containing 1—3 of the eight explanatory variables relating to diversity, environmental context and study design explained on average Fig 3.

Variance in local forest functions explained by diversity, environmental context and study design variables. Species and landscape diversity effects, adjusted for environmental context We calculated effect sizes r partial of tree species richness log-transformed in separate models that each first adjusted for i.

Table 1. Effects of local and landscape-level diversity on forest functions when fitted after environmental context variables. Discussion We investigated the effects of local and landscape-level diversity and environmental context variables on forest functioning in managed landscapes.

Supporting information. S1 Fig. Processing of light data. S2 Fig. S1 Table. Forest communities found in Switzerland. S2 Table. Results Total Species Richness For total species richness, the 3-km buffer was the best landscape scale i.

Figure 2. Relationship between species richness and proportion of forest cover at a landscape scale. Table 1. Wide-Ranging vs. Figure 3. Response of each species to proportion of forest cover at a landscape scale.

Interaction Terms Adding the interaction term between annual mean temperature and the proportion of forest cover improved model performance i. Figure 4. Effect of interaction between annual mean temperature and proportion of forest cover on the richness of narrow-ranging species.

Discussion The results largely supported our original hypotheses. Figure 5. Relationship between number of habitat types used by a species and range size.

Supporting Information. File S1. Supporting Appendixes. References 1. J Biogeogr 79— View Article Google Scholar 2. Ecol Lett — View Article Google Scholar 3. Fahrig L Effects of habitat fragmentation on biodiversity. Annu Rev Ecol Syst — View Article Google Scholar 4. Oikos 7: — View Article Google Scholar 5.

Trends Ecol Evol — View Article Google Scholar 6. Philos Trans R Soc B — View Article Google Scholar 7. Ecol Indic 41— View Article Google Scholar 8. Ecol Indic — View Article Google Scholar 9. Acta Zool Hung — View Article Google Scholar Biodivers Conserv — Ann Rev Ecol Syst — Science — Clavel J, Julliard R, Devictor V Worldwide decline of specialist species: toward a global functional homogenization?

Front Ecol Environ 9: — Balmford A Extinction filters and current resilience: the significance of past selection pressures for conservation biology. Biological Reviews — Jetz W, Rahbek C Geographic range size and determinants of avian species richness.

Ecol Lett 7: 81— Ecography 21— Chiron F, Filippi-Codaccioni O, Jiguet F, Devictor V Effects of non-cropped landscape diversity on spatial dynamics of farmland birds in intensive farming systems. Biol Conserv — Ibis — Biol Cons — Agric Ecosyst Environ 20— Clim Past 9: — Tokyo: Tsukijishokan. In Japanese. Ministry of the Environment Biodiversity Center of Japan. Oikos — Accessed May 9. Accessed July Miyashita T, Chishiki Y, Takagi SR Landscape heterogeneity at multiple spatial scales enhances spider species richness in an agricultural landscape.

Popul Ecol — Kato N, Yoshio M, Kobayashi R, Miyashita T Differential responses of two anuran species breeding in rice fields to landscape composition and spatial scale.

Wetlands — Cambridge: Cambridge University Press. Dormann CF Effects of incorporating spatial autocorrelation into the analysis of species distribution data. Global Ecol Biogeogr — Accessed August Accessed December Kissling WD, Carl G Spatial autocorrelation and the selection of simultaneous autoregressive models.

Global Ecol Biogeogr 59— Bivand R Spdep R package. Second edition. New York: Springer-Verlag. Gregory RD, Gaston KJ Explanations of commonness and rarity in British breeding birds: separating resource use and resource availability. J Anim Ecol — Holling CS Cross-scale morphology, geometry, and dynamics of ecosystems. Ecol Monogr — Szabo P, Meszena G Spatial ecological hierarchies: coexistence on heterogeneous landscapes via scale niche diversification.

Ecosystems 9: — Amano T Conserving bird species in Japanese farmland: past achievements and future challenges. Nature — Ecol Appl — Amano T, Yamaura Y Ecological and life-history traits related to range contractions among breeding birds in Japan.

The forests of birds. Japan: Tokai University Press in Japanese. Hirabuki, Y. Farmstead groves and traditional Lifestyle. Development process of two environmental educational programs based on local area field trail results. Proceedings Miyagi University Education, 9 , 59—66 in Japanese. Iida, S. Forest fragmentation and its effect on species diversity in sub-urban coppice forests in Japan. Forest Ecology and Management, 73 1 , — Imai, H. Environmental factors affecting the composition and diversity of avian community in mid-to late breeding season in urban parks and green spaces.

Landscape Urban Plann, 96 , — Ishida, K. The avian community in Lake Yamanaka around the Fuji forest of Tokyo university. Strix, 6 , 66—79 in Japanese. Japanese Society for the Preservation of Birds. Bird guide enlarged and revisedth ed. Karasawa, K. Relationship between fruit-eating birds and seed dispersal in urban areas.

Japan Journal of Ornithology, 27 , 1— Importance of sampling design and analysis in animal population studies: a comment on Sergio et al. Journal of Applied Ecology, 45 , — Proc Miyagi Univ Educ. Matsuda, M. Picture book of wild birds in Japan. Japan: Natsumesha in Japanese. Accessed 30 Nov Accessed 25 Dec in Japanese. Miura, S.

Homestead woodland developing with climate. Japan: Kahoku-Shinpo-sha in Japanese. Morioka, T. The birds of prey in Japan. Japan: Bun-ichi Sogo Press in Japanese. Osawa, S. The distribution of frogs on paddies locating different landforms in alluvial fan area. Journal of Rural Planning Associatoin, 19 4 , — in Japanese. Ecological diversity of birds in relation to the structure of urban green space. Landscape Urban Plann, 77 , 39— Savard, J.

Biodiversity concepts and urban ecosystems. Landscape Urban Plann, 48 , — Sergio, F. Ecologically justified charisma: preservation of top predators delivers biodiversity conservation. Journal of Applied Ecology, 43 , — Sheffield, L. Response of American kestrels and gray-tailed voles to vegetation height and supplemental perches.

Canadian Journal of Zoology, 79 , — Sieving, K. Heterospecific facilitation of forest-boundary crossing by mobbing understory birds in north-central Florida. Auk, , — Simberloff, D. Biological Conservation, 83 , — Suzuki, C.

Application of canonical correspondence analysis to soil microbial ecology. Soil Microorganisms, 63 , 32—38 in Japanese. Takagawa, S. Bird Res, 7 , R9—R12 in Japanese. Takano, S. Field guide wild birds of Japan. Takeuchi, K. Satoyama: the traditional rural landscape of Japan.

Japan: Springer. Book Google Scholar. Ter Braak, C. Canonical correspondence analysis: a new eigenvector technique for multivariate direct gradient analysis. Ecology, 67 , — Uezu, A. Can agroforest woodlots work as stepping stones for birds in the Atlantic forest region? Biodiversity and Conservation, 17 , — Watson, J. Habitat structure and proximity to forest edge affect the abundance and distribution of forest-dependent birds in tropical coastal forests of southeastern Madagascar.

Biological Conservation, , — Download references. We would like to thank Drs. Satoki and K. Hikosaka Tohoku University for their valuable advice. Kurokawa Forestry and Forest Products Research Institute gave me kind advice concerning the statistical analysis and support. All the members of the Laboratory of Functional Ecology gave helpful comments on our researches and encouragement through the discussion in all process of our study.

We sincerely thank all of them. The datasets used during the current study are available from the corresponding author on reasonable request. HI and TN designed the study. HI collected the data and analyzed with MO.

HI wrote the initial draft of the manuscript. All authors read and approved the final manuscript. You can also search for this author in PubMed Google Scholar. Correspondence to Haruka Imai. Reprints and Permissions. Environmental factors affecting the composition and diversity of the avian community in igune, a traditional agricultural landscape in northern Japan. Download citation. Received : 07 December Accepted : 16 February Published : 09 March Anyone you share the following link with will be able to read this content:.

Sorry, a shareable link is not currently available for this article. Provided by the Springer Nature SharedIt content-sharing initiative. Skip to main content. Search all BMC articles Search. Download PDF. Results The study was conducted in the suburban areas of Oshu and Hanamaki cities, Iwate Prefecture, Japan, at eight sites that varied in the density and area of igune woodlots within the landscape.

Conclusions Our results suggest that maintaining igune landscapes may enhance avian diversity within this landscape, although the effects of shrubs within igune varied; developed bush communities increased the evenness of the avian community, whereas some raptor species preferred an open forest understory.

Full size image. Table 1 Environmental factors selected in this study Full size table. Table 2 Classification of avian species into guild group and habitat type based on their ecological trait Full size table. Results Environmental factors Several environmental factors were significantly correlated with one another Table 3 , so we adopted only four variables density, distance, total area, and broadleaf for CCA.

Table 3 Single correlation coefficients r between environmental factors Full size table. Table 4 The correlation coefficients r between indices Full size table. Discussion The results of CCA indicated that the total area of woodlots most strongly affected the avian community in this agricultural landscape. References Clergeau, P.

Article Google Scholar Dunn, R.