Giant root-rat engineering and livestock grazing activities regulate plant functional trait diversity of an Afroalpine vegetation community in the Bale Mountains, Ethiopia

doi:10.21203/rs.3.rs-3168412/v1

Download PDF

Research Article

Giant root-rat engineering and livestock grazing activities regulate plant functional trait diversity of an Afroalpine vegetation community in the Bale Mountains, Ethiopia

https://doi.org/10.21203/rs.3.rs-3168412/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 01 Jun, 2024

Read the published version in Oecologia →

You are reading this latest preprint version

Disturbances from rodent engineering and human activities profoundly impact ecosystem structure and functioning. While we know that disturbances modulate plant communities, it is important to comprehend the mechanisms through which rodent and human disturbances modulate the functional trait composition of vegetation communities. Here, we evaluated the changes in functional trait diversity and composition of Afroalpine vegetation communities in the Bale Mountains of Ethiopia along gradients of engineering disturbances of a subterranean endemic rodent, the giant root-rat (Tachyoryctes macrocephalus RÜPPELL 1842) and human activities (settlement establishment and livestock grazing). We conducted RLQ (co-inertia analysis) and fourth-corner analyses to test for trait-disturbance (rodent engineering/human activities) covariation. Overall, our results show an increase in plant functional trait diversity with increasing root-rat engineering and increasing human activities. Functional traits, such as large seed mass, stoloniferous vegetative propagation and a prostrate stem were associated with increasing root-rat engineering, while leaf size and leaf nitrogen content were associated with increasing human activities. The results suggest that disturbances by rodents filter plant traits related to survival and reproduction strategies, whereas human activities such as livestock grazing act as filters for traits related to leaf economics spectrum along acquisitive resource use strategy. Overall, our results show that both rodent engineering and human activities act as habitat filters, but each favouring different combinations of traits reflecting specific adaptation strategies.

Functional trait dispersion

habitat filter

habitat heterogeneity

disturbance

subterranean rodent

Plants are a crucial component of ecosystems as they are the primary biomass producers and thus provide the energy that drives most ecosystem processes (Díaz et al. 2016). Thus, changes in plant communities often result in changes in important ecological processes, which in turn influence the extent, distribution, and diversity of organisms within ecosystems (Díaz et al. 2016; Eldridge and Soliveres 2023). Understanding the relationship of plant communities to environmental change caused by disturbances, e.g. bioturbation, and human activities, will allow projecting changes across trophic levels (Smith et al. 2022; Eldridge and Soliveres 2023).

One important factor impacting plant communities is the activity of bioturbating animals. Especially subterranean rodents are known as ecosystem engineers due to their extensive underground tunnel digging and mound building that greatly alter soil properties (Reichman and Seabloom 2002; Haussmann 2017) and impact plant communities by burying short vegetation and selectively feeding on preferred food plants (Jones et al. 1997; Escobedo et al. 2017; Asefa et al. 2022). On a landscape scale, they create a dynamic mosaic of burrow mounds varying in age, such as fresh active burrow mounds, old abandoned burrow mounds and mima-like mounds, characterized by different soil properties and plant communities (Jones et al. 1997; Cramer and Barger 2014; Šklíba et al. 2017; Asefa et al. 2022).

Consequently, engineering activities of subterranean rodents lead to increased microhabitat heterogeneity, which provides new resources such as open space and nutrients for species colonization and promote species coexistence within a community (Jones et al. 1997).

Similar to the engineering activities of rodents, human activities such as settlement establishment and livestock production also impact vegetation structure and plant community composition, mainly through grazing, trampling, defecation and urination (Dunne et al. 2011; Eldridge et al. 2016; Narantsetseg et al. 2018). For example, grazing and trampling by livestock reduces plant biomass, eliminates grazing intolerant species (Tessema et al. 2011) and damages seedlings and vegetative organs (Eldridge et al. 2016). At the same time, livestock activity creates open spaces for gap-colonizing plant species, promoting the dominance of unpalatable and grazing tolerant species (Tessema et al. 2011; Eldridge et al. 2016; Niu et al. 2019; Pavlů et al. 2019). Finally, livestock dung deposition and urination affect nutrient cycling and can reduce plant species diversity by facilitating encroachment of exploitive native and/or non-native plant species (Pavlů et al. 2019).

The degree to which rodent engineering and livestock grazing impact plant communities is largely modulated by plant functional traits. Functional traits, one or a combination of traits related to a specific adaptation strategy, impact plant growth, reproduction and survival, and thus, modulate the tolerance to disturbances (Voille et al. 2007). As such, quantifying plant functional traits along disturbance gradients can reveal changes in functional strategies of a community (Grime 2001; Westoby et al. 2002), e.g. favouring species with fast growth and low competitive ability in highly disturbed sites (Smith et al. 2022). These changes in the functional composition of plant communities can occur even if taxonomic measures like species richness remain unchanged due to disturbance (Smith et al. 2022). Thus, studying the functional composition is of importance to understand community drivers and mechanisms underpinning the changes.

A growing body of recent literature shows varying effects of different disturbance types on plant functional diversity and composition. For example, Choler (2005) has shown that the burrowing activity of the alpine marmot (Marmota marmota) benefits light-demanding rosette forbs, such as Alchemilla spp. In the alpine meadows of Qinghai–Tibet Plateau, Wang et al. (2020) have found negative associations between both plateau pika (Ochotona curzoniae) and plateau zokor (Myospalax baileyi) burrow densities and plant height. Similarly, an increase in burrowing activity of coruro (Spalacopus cyanus) leads to a decrease in plant functional diversity in an arid shrubland in Chile due to favouring a few invasive species (Escobedo et al. 2017). Narantsetseg et al. (2019) and Wang et al. (2020) have also reported that over-grazing by livestock leads to a decrease in functional diversity by sorting for species with acquisitive traits, e.g. larger leaf area and higher nitrogen content, thereby leading to functional trait convergence. Overall, findings of the above previous studies show that rodent engineering and human activities can act as habitat filters, and can potentially result in functional homogenization in species colonizing disturbed communities (Choler 2005; Escobedo et al. 2017). However, increased habitat heterogeneity associated with these disturbances increases niche opportunities which may also lead to trait diversification within species of a community (Mayfield and Levine 2010; de Bello et al. 2021). Thus, understanding the interplay between disturbances from rodent engineering and human activities in impacting plant functional diversity is important to disentangle their relative role in shaping biodiversity patterns. It is also important to predict future changes and to enable informed decisions in response to changing intensity of the disturbances.

In this study we evaluated the changes in functional trait diversity and composition of Afroalpine vegetation communities along gradients of engineering disturbances of a subterranean endemic rodent, the giant root-rat (Tachyoryctes macrocephalus RÜPPELL 1842), and human activities in the Bale Mountains of Ethiopia. To do so, we selected six traits to identify the main dimensions of variation in leaf and growth traits, which are known to have ecological functions related to resource use and acquisition, growth, survival and reproduction, and determined their association to root-rat engineering (fresh burrow density, old burrow density and presence of mima mound) and human disturbances (distance from settlement and livestock grazing intensity). Specifically, first we related multivariate functional trait diversity (functional divergence) to disturbance variables, and in a second step identified the association between each trait and disturbance. We expected increased functional trait diversity with increasing root-rat engineering and human activities since both disturbances are expected to create habitat heterogeneity and increased resources (space and nutrients). In the contrary, if these disturbances can act as habitat filters, we expected functional convergence along increasing disturbance gradients. Assuming that specific patterns of feeding and microhabitat changes are caused by rodents and livestock, we also expected disturbance specific associations with traits; specifically we expected strong association between giant root-rat and rosset (acaulscent, stoloniereous, rhizomatous) and larger seed mass traits, while human disturbance with acquisitive traits, e.g. larger leaf area and higher nitrogen content.

Study area

This study was conducted in the Afroalpine ecosystem of the Bale Mountains National Park in South-Eastern Ethiopia (BMNP; 6.508–7.178N, 39.508–39.928E; Fig. 1) between December 2020 and February 2021. The Bale Mountains represent the largest area of Afroalpine vegetation over 3,000 m above sea level in Africa (Yalden 1983). Elevation in the Bale Mountains ranges between 1,500 and 4,377 m asl. The area experiences two rainy seasons, with lighter rains from March to June and heavy rainy season from July to October, and a dry season between November and February; mean annual rainfall is approximately 1,000 mm (Miehe and Miehe 1994). The lowest and maximum recorded temperature in the Bale Mountains is -15°C and 26 ºC, respectively (Miehe and Miehe 1994; BMNP 2017). The soils in the Bale Mountains are entirely volcanic in origin and mainly derived from the basaltic and trachytic parent rock, are fairly fertile silty loams of reddish-brown to black colour (Hillman 1986; Miehe and Miehe 1994). Vegetation types of the Bale Mountains’ Afroalpine ecosystem include open grassland, grassland dotted with Artemisia afra shrub, Helichrysum dwarf-scrub, Alchemilla meadow, Lobelia rhychopetalum, and wetlands, such as alpine lakes, rivers, swamps and seasonal wetland grasslands; Tallents 2007).

Similar to many other alpine ecosystems in Africa, more rapid ecosystem changes have been detected in the Bale Mountains over the past 40 years (Tallents 2007; Johansson and Granström 2014; BMNP 2017). Subsistence livestock farming is one main income for the people and number of livestock is estimated at 200–300 heads km^− 2 (Vial et al. 2010). Reber et al. (2018) have recorded a total of 870 settlements in the Afroalpine zone of the Bale Mountains. A socio-economic survey conducted in 2013 showed 863 households, each having an average of four people, in the study area (BMNP, unpubl. data). For this study, we considered distance from settlement and livestock dung abundance as proxies for overall human activities. The Bale mountains also represent the world’s oldest known high-altitudinal residential site (Ossendorf et al. 2019) and the region is included in Conservation International’s Eastern Afro-Montane Biodiversity Hotspot, with the BMNP being recognized as the single most important conservation area in Ethiopia (BMNP 2017).

Giant root-rat and its disturbances

The giant root-rat (Tachyoryctes macrocephalus, Rüppell 1842), a subterranean rodent, is one of the several small mammal species restricted to the Afroalpine belt of the Bale Mountains (Yalden 1975). The species is restricted to < 1,000 km² area at altitudes from 3,000 to 4,150 m above sea level (Šumbera et al. 2018), where it is the main prey of the endangered Ethiopian wolf (Canis simensis) and numerous raptor species, such as golden eagle (Aquila chrysaetos), lesser-spotted eagle (A. pomarina), tawny eagle (A. rapax), Verreaux's eagle (A. verreauxi) and augur buzzard (Buteo rufofuscus) (Sillero-Zubiri et al. 1995; Asefa 2007). The giant root-rat is a diurnal species and occurs with a density of 22–90 animals ha^− 1 (Yalden 1985; Sillero-Zubiri et al. 1995; Tallents 2007).

Giant root-rats construct extensive large underground burrow systems. An individual root-rat burrow system extends up to 34 m, and branches into short tunnels that comprise nesting and food caching and defecation chambers (Beyene 1986; Sillero-Zubiri et al. 1995; Yaba et al. 2011). Giant root-rats produce three types of burrow marks: fresh burrows, old burrows, and mima mounds. Fresh burrows are easily distinguished from old burrows in that the former are freshly open or plugged holes that are currently active. However, root-rat old burrows are abandoned burrows, with holes open or plugged with weathered soil, partially or wholly covered by vegetation regrowth, and sometimes occupied by other small rodents. Mima mounds are rounded dome-shaped structures formed by continually burrowing activities of giant root-rats that measure up to 27 m in diameter and 1.5 m in height (Beyene 1986; Sillero-Zubiri et al. 1995; Cramer and Barger 2014; Šklíba et al. 2017; Wraase et al. 2023). Areas around root-rat mima mounds, which are giant root-rats’ favoured habitats, are characterized by the predominance of bare soil, as they eject soil from their burrow systems when excavating, and when plugging their burrow holes at night for thermoregulation (Yalden 1975). They also graze and gather vegetation for bedding around burrows, which further denudes the landscape (Beyene 1986; Yaba et al. 2011). As a result, giant root-rats have been known to cause changes in plant species diversity and composition (Tallents 2007; Šklíba et al. 2017; Asefa et al. 2022).

Data collection

Human activities, root-rat burrow density and plant species abundance

To examine the relationships between human settlement, livestock grazing, giant root-rat burrow density and plant functional traits, we systematically selected six study sites, between 5 to 20 km apart, spanning across the entire distribution range of the giant root-rat. At each site, we selected two adjacent settlements (3–5 km apart) that were known to be established 30 years ago (Hillman 1986; BMNP 2017). Starting at the centre of each settlement, we established three 1.5 km long transects at an angle of 80–120° (see the inset map on Fig. 1). Along each transect, we established six 25 m × 25 m plots at a distance of 250 m from each other. In total, there were 216 plots covering an area of 13.5 ha.

We undertook data collection in February and March of 2021. At each plot, we recorded (1) rodent engineering activities measured as fresh burrow density, old burrow density and presence-absence of mima mounds; (2) human activities measured as distance from settlement and abundance of livestock dung; and (3) plant species and species-specific cover.

We recorded plant species identity and species-specific cover data within two 10 m × 10 m subplots established at opposite corners of each plot. In each of these subplots, we identified all plants to species level, except grasses which were collectively recorded as a single morpho-species, and estimated, in 5% intervals, the cover of each species. For analyses on the plot level, we averaged cover values of each species obtained from the two subplots. Species occurring only in one plot, and grasses which were recorded as a single morphospecies were excluded from analysis. Thus, our data finally contained 61 species recorded across 216 plots.

Plant trait selection and data

Detecting significant associations between disturbance and functional traits depends on the list of examined traits (Westoby et al. 2002). Generally, a few traits, such as seed mass, leaf size and plant height, have been known to serve as surrogates for setting plant strategies related to the fundamental processes of plant life, i.e., dispersal, establishment, and persistence (Grime 2001; Westoby et al. 2002). As such, we selected six plant traits which are known to have ecological functions related to resource use and acquisition, growth, survival and reproduction and thus influence species responses to environmental changes caused by biotic disturbances (Díaz et al. 2016). These traits were: 1) adult plant maximum height (cm), 2) leaf area (mm²), 3) stem shoot growth form, three categories: acaulescent (without aboveground stem), prostrate or erect, 4) dispersal mode, in three categories: seed alone, seed and rhizome and seed and stolones, 5) leaf nitrogen content (mg g ^− 1), and 6) seed mass (mg). Species-specific trait data are provided in Online Resource 1.

Adult plant height is a measure of whole plant size and indicates ability to pre-empt resources, and therefore outcompete them. It also relates to plant resistance to damages from herbivory and burrowing activities of subterranean rodents (Díaz et al. 2016). Stem shoot growth form can also be linked with plant mechanical strength and resistance to biotic filters (Chave et al. 2009). We extracted information on species-specific maximum plant height and stem shoot growth form from botanical descriptions provided in the flora of Ethiopia and Eritrea (full references are provided in the Online Resource 2).

Leaf area (LA), one-sided surface area of an individual lamina, is a measure of leaf size and is relevant for light interception and has important consequences for leaf energy and water balance (Schrader et al. 2021). For leaf area estimation, we extracted information on minimum and maximum sizes of leaf length (L) and leaf width (W) the leaf blade (i.e. excluding petioles), as well as information on leaf shape type, from the flora of Ethiopia and Eritrea (for full list of references see Online Resource 2). In the case of compound leaves, single leaflets were treated as analogous to simple leaves with the exception of highly dissected pinnae for which we used the entire pinnae. Then, we estimated LA using Montemgory formula: Leaf area = cLW, where c is a correction factor to account for differences in leaf shape type among species (Schrader et al. 2021). We used c values, ranging between 0.55—0.79, reported by Schrader et al. (2021), as their analysis is based on global level datasets that encompass larger number of species and all of the ten leaf shape types we identified within our community samples (see Online Resource 2). In some cases, a species is characterized by having an intermediate shape between two shape types, e.g., elliptic to orbicular, in which case we used average values.

Leaf nitrogen mass is directly related to photosynthesis and respiration and reflects a trade-off between two different costs that increase with higher nitrogen (to acquire N, and potentially suffer more herbivory), on the one hand, and the greater photosynthetic potential that higher nitrogen allows, on the other hand (Díaz et al. 2016). For most species, we compiled data on leaf nitrogen content from the TRY Plant Trait Database30 (Kattge et al. 2022; https://www.try-db.org, accessed 20 February 2023), and for the remaining species for which traits are not included in the TRY database we extracted from published literature.

Seed mass (mass of an individual seed plus any additional structures that assist dispersal and do not easily detach) indexes species along a dimension describing the trade-off between seedling competitiveness and survival on the one hand, and dispersal and colonization ability on the other (Thompson et al. 1993). We compiled data on seed mass for all species from Seed Information Database (SER, INSR and RBGK 2023; https://ser-sid.org). Similarly, vegetative dispersal is an adaptation to resist to/recovery from herbivory damages and an adaptation strategy to compensate the erratic seed production and seedling establishment in alpine habitats (Choler 2005). We compiled information on vegetative dispersal from flora of Ethiopia and Eritrea (list of references are provided in the Online Resource 2).

Species names were standardized according to The World Flora Online database (www.worldfloraonline.org/; accessed April 2023), and each species and its synonyms, subspecies or local variety is represented by a single value for each trait. The number of observations per trait and species range from a single one to hundreds. Thus, we calculated the geometric mean of all the records of a trait compiled. All data were unit-standardized and subjected to error detection and quality control. Accordingly, we excluded trait records measured on juvenile plants and on plants grown under non-natural environmental conditions, duplicate trait records for the same species on a particular trait, and potential outliers – trait records with a distance of > 3 standard deviations from the mean of the species (Diaz et al. 2016). The remaining dataset was used to calculate species mean trait values.

Statistical analysis

To assess whether increased giant root-rat disturbance and human activities act as agents creating habitat heterogeneity and thus increasing functional divergence or as a habitat filters by decreasing community functional divergence, we calculated functional dispersion (FDis) based on principal coordinates analysis (PCoA) of a Hill-Smith dissimilarity matrix (FDis; Dray and Dufour 2007). FDis is the mean distance of each species to the centroid of all species in the community, weighted by its cover-abundance. Thus, a decrease in FDis means that community composition has shifted towards species that are more similar to each other, i.e. functional convergence, in response to increased disturbance. We used FDis because it is independent of species richness and takes into account species abundance; moreover, it can be used for multiple traits, as well as for continuous and categorical trait values (Dray and Dufour 2007). We calculated FDis using ade4 package (Dray and Dufour 2007).

To evaluate whether the observed FDis of traits in each plot was higher or lower than expected from a null model, we estimated the standardized effect size (SES_FDis) using 9999 randomly generated null communities with an ‘independent swap’ algorithm (Gotelli 2000). SES_FDis was calculated using the R package picante (Kembel et al. 2010) as the differences between the observed FDis and null FDis divided by the standard deviation (s.d.) of the null data; positive and negative SES_FDis values indicate higher or lower functional divergence than the null expectation values, respectively (Gotelli, 2000; Kembel et al. 2010). In terms of assembly processes, positive SES_FDis values are considered to be caused by competitive exclusion that limits functional similarity or by habitat heterogeneity that increases resources and thus reduces competition, negative SES_FDis values by habitat filtering of functionally similar species, and values near zero indicates random distribution of traits or concurrent occurrence of competition or habitat heterogeneity and habitat filtering, off-setting their effects (Edwards et al. 2021). Thus, we tested whether mean SES_FDis was significantly different from zero using t-test. To determine the relationship between SES_FDis values with the giant root rat and human disturbances, we also performed regression analysis based on GLMMs using glmmTMB package (Brooks et al. 2017), by including transect nested within site as random component. We checked for multicollinearity among predictors using the ‘performance’ R package (Lüdecke 2021); this confirmed lack of collinearity problem (VIF ranged between 1.11–1.73). We also used diagnostic plots in the DHARMa R package (Hartig 2021) and confirmed that model assumptions were met.

To directly measure the link between species traits and environmental data, we used Dray and Legendre’s (2008) novel version of the fourth-corner analysis provided in the ade4 package (Dray and Dufour 2007). Accordingly, we first conducted separate ordinations of three tables: ordination of table L (species abundance), which was done by a correspondence analysis (CA); table R (disturbance variables), done by a hillsmith function with row weights of table L; and table Q (species traits), done by a hillsmith function with the column weights of table L (Dray and Dufour 2007; Dray and Legendre 2008). To evaluate the global, or their joint multivariate relationship, significance of the trait-environment relationships, we conducted RLQ analysis using outputs of the ordinations described above. The RLQ analysis is a three-tables co-inertia analysis that tends to maximize the covariance between the sample site scores constrained by the environmental variables of table R and the species scores constrained by the traits of table Q (Dray and Legendre 2008). Finally, we undertook the fourth-corner analysis. In the fourth-corner procedure, a matrix L with species abundances is related to a matrix R with variables describing the extent of giant root-rat and human disturbances at the sample plots and a matrix Q describing species traits (Dray and Dufour 2007). The environmental matrix (R) contained the three root-rat disturbance variables: presence/absence of mima mound, root-rat old burrow density and root-rat fresh burrow density, and the two human activities: distance from settlement and livestock grazing intensity. The trait matrix (Q) was composed of six species traits: three continuous variables – plant adult height, leaf area and leaf nitrogen content, and three categorical variables: stem growth form (acaulescent, erect, and prostrate), and mode of propagation (seed only, seed and rhizome, and seed and stolones). We used the permutation model 6, with 999 permutations (Dray and Legendre 2008), which permutes all species within an entire column and row of the L matrix to test the null hypotheses of the observed pattern would be different from random. To evaluate the global, or their joint multivariate relationship, significance of the traits-environment relationship we applied a multivariate test using fourthcorner2 function of the ade4 package. The significance of observed statistic was tested based on 999 Monte-Carlo permutations (Dray and Dufour 2007). To measure the strength and significance of the links between individual trait and environmental variable, we used a Pearson correlation coefficient for two quantitative variables, a Pearson Chi² for two qualitative variables and a Pseudo-F for one quantitative variable and one qualitative variable (Dray and Dufour 2007). We conducted all analyses in the R environment (R Core Team 2020), and all R scripts used for analysis are provided in Online Resource 3.

Leaf area of plant species ranged between 5.6–490095 mm² (mean: 2442.45 ± 835.30 mm²), while leaf nitrogen content (Nmass) ranged between 7.53–54.82 mg g ^− 1 (mean: 24.26 ± 1.25 mg g ^− 1). Adult plant height ranged between 5–300 cm (mean: 68.27 ± 8.84 cm), and seed mass between 0.02–1477 mg (mean: 1.15 ± 0.27 mg) (Table 1). Over half (38 species; 62.3% of the total species) of the studied plant species were found to reproduce only by seed, while 14 (23.0%) and 9 (14.8%) reproduce, in addition to by seed, by rhizome and seed and stolones propagule organs, respectively. In terms of species shoot growth form, species with erect stem growth form contributed over a third to the total species (42 species; 68.9%), while those with prostate and acaulescent growth forms, respectively, contributed 14 species (23.0%) and 5 species (8.2%) (for detail see Online Resource 2). Mean FDis was 0.154 ± 0.003, while that of SES_FDis was − 0.067 ± 0.068 (Table 1).

Table 1

Mean ± standard error (SE) and range (minimum-maximum) values of quantitative giant root-rat (*Tachyoryctes macrocephalus*) disturbance variables (giant root-rat old burrow and giant root-rat fresh burrow) and human activities (livestock dung abundance and distance from human settlement), individual traits, functional trait diversity (FDis) and standardized effects size (SES_FDis.
Variable	Mean ± SE	Range
Disturbance variables
Root-rat old burrow (No. per plot)	44.86 ± 3.67	0–316
Root-rat fresh burrow (No. per plot)	30.76 ± 4.12	0.00–333
Livestock dung abundance (No. per plot)	52.24 ± 3.70	0–250
Distance from settlement (m)	625.00 ± 29.12	0–1250
Trait values
Leaf area (LA; mm2)	2442.45 ± 835.3	5.6–40095
Leaf nitrogen content (Nmass; mg g ^− 1)	24.26 ± 1.25	7.53–54.82
Adult plant height (Ht, cm)	68.27 ± 8.84	5–300
Seed mass (Smass, mg)	1.15 ± 0.27	0.02–1477
Functional dispersion (FDis)	0.15 ± 0.003	0–0.25
Standardized effect size (SES_FDis)	-0.07 ± 0.07	-1.83–2.76

We counted a mean (± SE) number of 44.86 ± 3.67 and 30.76 ± 4.12 burrows per plot of root-rat old burrow and fresh burrow densities, respectively. Root-rat mima mounds were encountered at 97 plots, or 45% of the total plots. Livestock dung abundance varied between 0–250 (mean: 52.24 ± 3.70) per plot (Table 1). Distance from human settlement ranged between 0 − 1250 m (mean: 500 ± 176.8).

Our test of if the observed SES_FDis of traits in each plot was higher or lower than expected from a null model revealed that mean SES_FDis was not significantly different from zero (t = -0.9764, df = 215, P = 0.33). However, analysing of the effects of giant root rat and human disturbances on the FDis, we found significantly increased SES_FDis with increasing root-rat fresh burrow density (Z = 2.566, P < 0.05) and livestock dung abundance (Z = 3.121, P < 0.01), but lower at mima mound sites than at sites where mima mounds were absent (Z = 2.212, P < 0.05) and decreased with increasing distance from settlement areas (Z = 3.756, P < 0.001) (Table 2; Fig. 2a–d).

Table 2

Summary statistics for linear mixed-effects model testing the effects of human disturbances (distance from human settlement and livestock dung abundance) and giant root-rat (*Tachyoryctes macrocephalus*) disturbances (giant root-rat fresh burrow density, presence of mima mound, giant root-rat old burrow density) on standardized effect size (SES_FDis) values.
Explanatory variable	Estimate ± S.E.	Z-value
Intercept	0.1484 ± 0.1957	0.754
Distance from settlement	-0.0006 ± 0.0002	3.756***
Livestock dung abundance	0.0045 ± 0.0014	3.121**
GRR fresh burrow density	0.0027 ± 0.0010	2.566*
Mima mound (present)	-0.2905 ± 0.1306	2.212*
GRR old burrow density	-0.0005 ± 0.0012	0.440
Asterisks indicate significance: P < 0.05, P < 0.01, **P < 0.001

Analyzing the multivariate link between species traits and environmental data, we found that the general association between plant traits and disturbance measures (root-rat and human) was larger than expected under random community assembly (Monte-Carlo test, observed covariance: 0.578, expected covariance under random community assembly: 0.231, P < 0.05). The first axis of the RLQ summarised most of the projected inertia between traits (Table R) and disturbance metrics (Table Q; eigenvalues: 0.55; covariance: 0.74; % total inertia: 87.7%), while the value was 9.1% (eigenvalues: 0.06; covariance: 0.74) for the second axis (Table 3; Fig. 3). Correlation between the two new sets of factorial scores projected onto the first two RLQ axes was 0.41 and 0.18, respectively. Projected inertia, i.e. the variance of each set of factorial scores computed in the first two axes of the RLQ analysis, was 71.8% and 88.7%, and 62.4% for the disturbances and 71.3% for traits (Table 3). Distance from human settlement and livestock dung abundance were associated with the first axis of the RLQ. Root-rat variables were associated with the second axis (Fig. 3).

Table 3

Summary results of RLQ analysis for the first two axes showing eigenvalues, percent of total coinertia, covariance, correlation, projected inertia of Table R (% of total inertia) and Table Q (% of total inertia).
	Axis 1	Axis 2
Eigenvalues	0.55	0.06
Percent of total coinertia	87.7	9.1
Covariance	0.74	0.24
Correlation	0.41	0.18
Projected inertia of Table R (% of total inertia)	1.30 (71.8)	1.18 (88.7)
Projected inertia of Table Q (% of total inertia)	1.41 (62.4)	1.14 (71.3)

Examining disturbance specific associations with traits based on the fourth-corner analyses, we found significant associations of individual plant traits and disturbance variables. While increasing root-rat engineering disturbances were associated with larger seed mass, stoloniferous vegetative propagation and prostrate stem shoot, increasing human disturbances were associated with larger leave size and increased leaf nitrogen content (Figs. 3 and 4). Certain species were found to be strongly associated with specific traits that in turn were associated with a specific disturbance type (Online Resource 4). For example, characteristic stolonifereous, prostrate rosette species, such as Alchemilla abyssinica, Helichrysum gofense and Euryops prostrates, were associated with root-rat mima mounds, and species such as Crepis rueppelii, Haplocarpha rueppelii, Kniphofia foliosa, Potentilla dentate and Umbilicus bostryoides with larger leaf size and higher nitrogen content were associated with human activities (Online Resource 4).

We identified associations between functional trait diversity and composition of Afroalpine vegetation communities and disturbances by giant root-rats and humans. While we did not find an overall signal of functional dispersion within our community, we found an increase in plant functional trait diversity with increasing root-rat (fresh burrows) engineering and increasing human activities. However, the diversity was lower at sites with root-rat mima mounds compared with sites where mima mounds were absent. Yet, each disturbance had specific associations with traits.

In line with our first prediction, both root-rat fresh burrows engineering disturbances and human activities (livestock grazing and settlement) appeared to enhance FDis, suggesting that these disturbances lead to increased habitat heterogeneity that, in turn, results in a functional trait divergence. These disturbances are known to cause increased soil nutrient, create open spaces and reduce of abundances of dominant species, which facilitate colonization and establishment of species with diverse traits (Choler 2005; Haussmann 2017). In our study area, Šklíba et al. (2017) have reported that soils of giant root-rat mounds contain higher nitrogen and other nutrients, which could enhance coexistence of species with different trait values. This enhanced resource availability explains our finding of a significantly increased SES_FDis with increasing root-rat fresh burrow density (see Choler 2005; Escobedo et al. 2017; Boet et al. 2020). Similar studies (e.g. Chambers et al. 1990; Choler 2005) have also reported the importance of disturbances from rodent engineering activities in maintaining or enhancing the structural and functional diversities of Arctic and Alpine meadow vegetation communities. Moderate level of disturbances [sensu Connel’s (1978) ‘Intermediate Disturbance Hypothesis’] caused by human activities, such as livestock grazing, often cause habitat heterogeneity and plant functional diversity, and thus have been used as management tool to improve rangeland conditions (Bokdam 2001; Eldridge et al. 2016; Pavlů et al. 2019).

In the contrary to the above habitat heterogeneity process, our results showed three evidences that these disturbances also act as habitat filters. Firstly, despite the positive relationship found between root-rat fresh burrow and human activities, the overall SES_FDis value was similar to that obtained from null model. Secondly, mean SES_FDis was lower at root-rat mima mounds, compared with at sites where mima mounds were absent. Finally, as predicted, we found disturbance specific associations with traits (see below). In the first case, such mean value of SES_FDis near zero can be the outcome of two assembly processes, which may off-set each other’s effects: random species distribution (no effect of disturbances) vs. concurrent occurrence of competition, or habitat heterogeneity, and habitat filtering processes (Edwards et al. 2021; de Bello et al. 2021). Different disturbance types influence specific assembly process and/or trait differentially (Choler 2005; Wang et al. 2020; Smith et al. 2022); for example, giant root-rat fresh burrow cause habitat heterogeneity while its mima mound acts as habitat filter. Thus, the lack of significant departure of mean SES_FDis from zero in our community can likely be attributed to the simultaneous operation of habitat heterogeneity and habitat filtering assembly processes, which leads their resultant effect at plot level to be stable functional trait diversity or mean SES_FDis value approaching zero (de Bello et al. 2021; Smith et al. 2022).

In line with our second prediction, we found disturbance specific associations with traits. Considering root-rat disturbance, for example, we found strong association of larger seed mass with increasing root-rat fresh burrow density, rhizomatous vegetative propagation with increasing giant root-rat old burrow, and stolonifereous vegetative propagation with presence of giant root-rat mima mound (Fig. 3). These results indicate that different engineering disturbance types of subterranean rodents to having varying filtering effects on biodiversity and ecosystem functioning and that considering different types of disturbances would help reveal the mechanisms how engineering activities of rodent impact plant functional trait diversity. More specifically, the filtering effect of the giant root-rat mima mound, for example, can be explained in relation to previous studies who reported strong associations with mima mounds of characteristic stolonifereous, prostrate rosette species, such as Alchemilla abyssinica, Helichrysum gofense and Euryops prostrates (Beyene 1986; Miehe and Miehe 1994; Tallents 2007; see also Fig. 3 and Online Resource 4). The latter two species are endemic to the Bale Mountains and have been recorded only at giant root-rat mima mound sites (Miehe and Miehe 1994; Tallents 2011). Overall, the stolonifereous trait is among commonly known traits of alpine plants that serve as a reproduction strategy to reduce resource allocation to seed production (Choler 2005). Thus, as mima mounds are relatively elevated compared with the surrounding areas (Wraase et al. 2022), being stoloniferous enables such characteristic species to persist on mima mounds by overcoming alpine wind pressure. Our finding of positive association of seed mass with root-rat fresh burrow density can be explained by the fact that this disturbance favours species with large seed size as large seeds facilitate survival through the early stages of recruitment, and higher establishment in the face of rodent herbivory and burrowing disturbances (Westoby et al. 2002; Muller-Landau 2010). Similar to our finding, Šklíba et al. (2017) have reported strong association of plant species with relatively larger seed mass, such as Urtica simensis, Carduus nyassanus and Salvia merjamie, with giant root-rat fresh mounds. The former two species are also characterized by mechanical defence against herbivores (Šklíba et al. 2017), similar to the high abundance of unpalatable plants observed in the Tibetan plateau as a consequence of the activity of the plateau zokors (Myospalax baileyi; Zhang and Liu 2003; Wang et al. 2020).

Similarly, larger leaf size and leaf nitrogen content are strongly associated with human activities (Fig. 3). This result is in line with our prediction and corroborates findings of previous studies that human activities related to livestock production are globally known to favour plants with resource acquisitive traits (Grime 2001; Wright et al. 2004; Dunne et al. 2011). Higher leaf area and leaf nitrogen content are widely known as indicative of a fast-growing, rapid nutrient acquisition strategy (Liu et al. 2017; Smith et al. 2022), as their combination increases energy exploitation through improved photosynthetic capacity (Wright et al. 2004). Overall, the increased levels of leaf area and leaf nitrogen content with increasing livestock grazing intensity permits the success of quick-growing plants that can quickly access resources (Bokdam 2001; Eldridge et al. 2016; Niu et al. 2019; Pavlů et al. 2019; Smith et al. 2022). Characteristic species with such acquisitive leaf economics spectrum and are associated with human activities include: Crepis rueppelii, Haplocarpha rueppelii, Kniphofia foliosa, Potentilla dentate and Umbilicus bostryoides. Most of these species have relatively larger leaf area (see Online Resource 1) and are known to be weedy growing in degraded areas (see Online Resource 2).

In conclusion, we found overall stable functional trait diversity, which likely is due to opposing effects (creation of habitat heterogeneity vs habitat filtering) of the different disturbance types and associations of each disturbance type with specific trait types. Generally, root-rat disturbances filtered plants with higher seed mass, and with stolonifereous and rhizomatous vegetative organs, while human disturbances filtered species with larger leaf area and higher leaf nitrogen content. From these findings, it is possible to conclude that two plant strategies can be revealed: traits related to survival and reproduction strategies are associated with root-rat disturbances, while leaf traits related to resource-acquisitive economics spectrum are associated with human disturbances (Díaz et al. 2016). In sum, identification of such associations between plant traits and disturbance can help predicting changes under future environmental change and on which trait-disturbance associations to focus in effective ecosystem mismanagement.

Acknowledgements

This research was funded by the German Research Council (DFG) within the framework of the joint Ethio-European DFG Research Unit 2358 “The Mountain Exile Hypothesis (NA 783/12-2, FA-925/14-1 und SCHA-2085/3-1, MI271/33-2): How humans benefited from and re-shaped African high-altitude ecosystems during Quaternary climate changes.” We thank the Ethiopian Wildlife Conservation Authority, the Frankfurt Zoological Society, the Ethiopian Wolf Project, and the Bale Mountains National Park for their cooperation and kind permission to conduct fieldwork. We are grateful to Katinka Thielsen for helping to prepare the fieldwork, Sena Gashe, Daniel Tilaye, Issa Hassan, Usman Abdella, Abu Abduku and Dinsho horsemen for helping with the field work.

Funding: This work was supported by the German Research Council (DFG) within the Research Unit 2358 (“The Mountain Exile Hypothesis”) [NA 783/12-2, FA-925/14-1 und SCHA-2085/3-1, MI271/33-2].

Conflicts of interest/Competing interests: The authors have no conflict of interest to declare.

Ethics approval: Not applicable.

Consent to participate: Not applicable.

Consent for publication: Not applicable.

Data availability: All relevant data are within the manuscript and its Supplementary Material.

Code availability: R code used for analysis is provided in Electronic Supplementary Material.

Authors’ contribution statement: All authors developed and designed the study. AA collected data. AA, VR, DGS and NF analyzed data and wrote the manuscript. Other authors reviewed and edited.

Asefa A (2007) Birds of Bale Mountains National Park (BMNP), Southeast Ethiopia. Walia 25:22–33
Asefa A, Reuber V, Miehe G, Wondafrash M, Wraase L, Wube T, Farwig N, Schabo DG (2002) The activity of a subterranean small mammal alters Afroalpine vegetation patterns and is positively affected by livestock grazing. Basic Appl Ecol 65:67–76. https://doi.org/10.1016/j.baae.2022.09.001
Beyene S (1986) A study on some ecological aspects of the giant molerat Tachyoryctes macrocephalus in Bale Mountains, Ethiopia. Master thesis, Department of Biology, Addis Ababa University, Addis Ababa, Ethiopia
BMNP (Bale Mountains National Park) (2017) BMNP General Management Plan 2017–2027. BMNP, Dinsho, Ethiopia
Boet O, Arnan X, Retana J (2020) The role of environmental vs. biotic filtering in the structure of European ant communities: A matter of trait type and spatial scale. PLoS One (2020 Feb 1915(2):e0228625. https://doi.org/10.1371/journal.pone.0228625
Bokdam J (2001) Effects of browsing and grazing on cyclic succession in nutrient-limited ecosystems. J Veg Sci 12:875–886
Brooks ME, Kristensen K, van Benthem KJ, Magnusson A, Berg CW, Nielsen A, Skaug HJ, Maechler M, Bolker BM (2017) glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. R J 9(2):378–400
Chambers JC, Mcmahon JA, Brown RW (1990) Alpine seedling establishment: The influence of disturbance type. Ecology 71:553–563
Chave J et al (2009) Towards a worldwide wood economics spectrum. Ecol Lett 12:351–366
Choler P (2005) Consistent shifts in alpine plant traits along a mesotopographical gradient. Arct Antarct Alp Res 37(4):444–453
Cramer MD, Barger NN (2014) Are mima-like mounds the consequence of long-term stability of vegetation spatial patterning? Palaeogeogr Palaeoclimatol Palaeoecol 409:72–83. http://dx.doi.org/10.1016/j.palaeo.2014.04.026
de Bello F, Carmona CP, Dias ATC, Götzenberger L, Moretti M, Berg MP (2021) Handbook of trait-based ecology: From theory to R tools. Cambridge University Press, Cambridge
Díaz S et al (2016) The global spectrum of plant form and function. Nature 529:167–171
Díaz S et al (2019) Pervasive human-driven decline of life on Earth points to the need for transformative change. Science 366(6471):eaax3100. https://doi.org/10.1126/science.aax3100
Dray S, Dufour A (2007) The ade4 Package: Implementing the duality diagram for ecologists. J Stat Softw 22(4):1–20. https://doi.org/10.18637/jss.v022.i04
Dray S, Legendre P (2008) Testing the species traits-environment relationships: the fourth-corner problem revisited. Ecology 89:3400–3412
Dunne T, Western D, Dietrich WE (2011) Effects of cattle trampling on vegetation, infiltration, and erosion in a tropical rangeland. J Arid Environ 75(1):58–69. https://doi.org/10.1016/j.jaridenv.2010.09.001
Edwards FA, Edwards DP, Hamer KC, Fayle TM (2021) Tropical land–use change alters trait–based community assembly rules for dung beetles and birds. Oecologia 195:705–717. https://doi.org/10.1007/s00442-020-04829-z
Eldridge DJ, Soliveres S (2023) Rewilding soil-disturbing vertebrates to rehabilitate degraded landscapes: benefits and risks. Biol Lett 19:20220544. https://doi.org/10.1098/rsbl.2022.0544
Eldridge DJ, Poore AGB, Ruiz-Colmenero M, Letnic M, Soliveres S (2016) Ecosystem structure, function, and composition in rangelands are negatively affected by livestock grazing. Ecol Applic 26(4):1273–1283
Escobedo VM, Rodrigo S, Rios RS, Salgado-Luarte C, Stotz GC, Gianoli E (2017) Disturbance by an endemic rodent in an arid shrubland is a habitat filter: effects on plant invasion and taxonomical, functional and phylogenetic community structure. Ann Bot 119:659–670. https://doi.org/10.1093/aob/mcw258
Gotelli NJ (2000) Null model analysis of species co-occurrence patterns. Ecology 81:2606–2621
Grime JP (2001) Plant Strategies, Vegetation Processes, and Ecosystem Properties, 2nd edn. John Wiley & Sons, Chichester, UK
Hartig F (2022) DHARMa: Residual Diagnostics for Hierarchical (Multi-Level / Mixed) Regression Models. R package version 0.4.5. http://florianhartig.github.io/DHARMa/
Haussmann NS (2017) Soil movement by burrowing mammals: A review comparing excavation size and rate to body mass of excavators. Progr Phys Geogr 41(1):29–45. https://doi.org/10.1177/0309133316662569
Hillman JC (1986) Management Plant of Bale Mountains National Park. BMNP, Dinsho, Ethiopia
Johansson MU, Granström A (2014) Fuel, fire and cattle in African highlands: Traditional management maintains a mosaic heathland landscape. J Appl Ecol 51:1396–1405
Jones CG, Lawton JH, Shachak M (1997) Positive and negative effects of organisms as physical ecosystem engineers. Ecology 78(7):1946–1957
Kattage J et al (2022) Plant trait data in the TRY File Archive (https://www.try-db.org/TryWeb/Data.php)
Kembel S, Cowan P, Helmus M, Cornwell W, Morlon H, Ackerly D, Blomberg S, Webb C (2010) Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26:1463–1464
Lüdecke D (2021) Package ‘performance’: Assessment of Regression Models Performance, R package Version 0.8.0. https://easystats.github.io/performance/
Mayfield MM, Levine JM (2010) Opposing effects of competitive exclusion on the phylogenetic structure of communities. Ecol Lett 13(9):1085–1093. https://doi.org/10.1111/j.1461-0248.2010.01509.x
Miehe G, Miehe S (1994) Ericaceous Forests and Heathlands in the Bale Mountains of South Ethiopia: Ecology and Man’s Impact. Stiftung Walderhaltung in Afrika & Bundesforschungsanstalt fur Forst- und Holzwirtschaft, Hamburg
Muller-Landau HC (2010) The tolerance-fecundity trade-off and the maintenance of diversity in seed size. Proc Natl Acad Sci USA 107:4242–4247
Narantsetseg A, Kang S, Ko D (2018) Livestock grazing and trampling effects on plant functional composition at three wells in the desert steppe of Mongolia. J Ecol Environ 42: 13 (2018). https://doi.org/10.1186/s41610-018-0075-2
Niu Y, Zhou J, Yang S, Chu B, Zhu H, Zhang B, Fang Q, Tang Z, Hua L (2019) Plant diversity is closely related to the density of zokor mounds in three alpine rangelands on the Tibetan Plateau. PeerJ 7:e6921. https:doi.org/10.7717/peerj.6921
Ossendorf G et al (2019) Middle Stone Age foragers resided in high elevations of the glaciated Bale Mountains, Ethiopia. Science 365(6453):583–587. https://doi.org//10.1126/science.aaw8942
Pavlů K, Kassahun T, Nwaogu C, Pavlů L, Gaisler J, Homolka P, Pavlů V (2019) Effect of grazing intensity and dung on herbage and soil nutrients. Plant Soil Environ 65:343–348. https://doi.org/10.17221/177/2019-PSE
R Core Team (2020) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org
Reber D, Fekadu M, Detsch F, Vogelsang R, Bekele T, Nauss T, Miehe G (2018) High-Altitude Rock Shelters and Settlements in an African Alpine Ecosystem: The Bale Mountains National Park, Ethiopia. Hum Ecol 46:587–600. https://doi.org/10.1007/s10745-018-9999-5
Reichman OJ, Seabloom EW (2002) The Role of Pocket Gophers as subterranean ecosystem engineers. Tren Ecol Evol 17(1):44–49
SER INSR, RBGK [Society for Ecological Restoration, International Network for Seed Based Restoration, Royal Botanic Gardens Kew] (2023) Seed Information Database (SID). Available from: https://ser-sid.org/ (updated February 2023; accessed 28 April 2023)
Schrader J, Shi P, Royer DL, Peppe DJ, Gallagher RV, Li Y, Wang R, Wright IJ (2021) Leaf size estimation based on leaf length, width and shape. Ann Bot 128(4):395–406. https://doi.org/10.1093/aob/mcab078
Sillero-Zubiri C, Tattersall FH, Macdonald DW (1995) Habitat selection and daily activity of Giant Molerats Tachyoryctes macrocephalus: Significance to the Ethiopian Wolf Canis Simensis in the Afroalpine ecosystem. Biol Conserva 72(1):77–84
Šklíba J, Vlasatá T, Lövy M, Hrouzková E, Meheretu Y, Sillero-Zubiri C, Šumbera R (2017) Ecological role of the Giant Root-Rat (Tachyoryctes macrocephalus) in the Afroalpine ecosystem. Integr Zool 12(4):333–344
Smith EA, Holden EM, Brown C, Cahill JF Jr (2022) Disturbance has lasting effects on functional traits and diversity of grassland plant communities. PeerJ 10:e13179. https://doi.org/10.7717/peerj.13179
Šumbera R, Krásová J, Lavrenchenko LA, Mengistu S, Bekele A, Mikula O, Bryja J (2018) Ethiopian highlands as a cradle of the African subterranean root-rats (genus Tachyoryctes), the genetic evidence. Mol Phylogen Evol 126:105–115. https://doi.org/10.1016/j.ympev.2018.04.003
Tallents LA (2007) Determinants of reproductive success in Ethiopian wolves. PhD Thesis, Zoology Department, University of Oxford, Oxford
Tessema ZK, de Boer WF, Baars RMT, Prins HHT (2011) Changes in soil nutrients, vegetation structure and herbaceous biomass in response to grazing in a semi-arid savanna of Ethiopia. J Arid Environ 75:662–670. https://doi.org/10.1016/j.jaridenv.2011.02.004
Thompson K, Band SR, Hodgson JG (1993) Seed size and shape predict persistence in soil. Funct Ecol 7:236–241
Vial F, Sillero-Zubiri C, Marino J, Haydon DT, Macdonald DW (2010) An analysis of long-term trends in the abundance of domestic livestock and free-roaming dogs in the Bale Mountains National Park, Ethiopia. Afr J Ecol 49:91–102
Voille C, Navas M-L, Vile D, Kazakou E, Fortunel C, Hummel I, Garnier E (2007) Let the concept of trait be functional! Oikos 116(5):882–892. https://doi.org/10.1111/j.0030-1299.2007.15559.x
Wang Y, Zhang X, Sun Y, Chang S, Wang Z, Li G, Hou F (2020) Pika burrow and zokor mound density and their relationship with grazing management and sheep production in alpine meadow. Ecosphere 11(5):e03088. https://doi.org/10.1002/ecs2.3088
Westoby M, Falster DS, Moles AT, Vesk PA, Wright IJ (2002) Plant ecological strategies: some leading dimensions of variation between species. Annu Rev Ecol Syst 33:125–159
Wraase L, Reuber VM, Kurth P, Fekadu M, Demissew S, Miehe G, Opgenoorth L, Selig U, Woldu Z, Zeuss D, Schabo DG, Farwig N, Nauss T (2023) Remote sensing-supported mapping of the activity of a subterranean landscape engineer across an Afro-alpine ecosystem. Remote Sens Ecol Conserv 9(2):195–209. https://doi.org/10.1002/rse2.303
Wright IJ et al (2004) The worldwide leaf economics spectrum. Nature 428:821–827
Yaba M, Mekonnen T, Bekele A, Malcolm J (2011) Food selection and feeding behavior of giant mole rat (Tachyoryctes Macrocephalus, Ruppell, 1842) from the Sanetti plateau of Bale Mountains National Park, Ethiopia. Asian J Appl Sci 4(7):735–740
Yalden DW (1975) Some observations on the giant mole-rat Tachyoryctes macrocephaluus (Rüppell, 1842) (Mammalia Rhizomydae) of Ethiopia. Monit Zool Ital Suppl 6(1):275–303. https://doi.org/10.1080/03749444.1975.10736820
Yalden DW (1985) Tachyoryctes macrocephalus. Mamm Species 237:1–3
Yalden DW (1983) The extent of high ground in Ethiopia compared to the rest of Africa. SINET: Ethiop J Sci 6:35–38
Zhang Y, Liu J (2003) Effects of plateau zokors (Myospalax fontanierii) on plant community and soil in an alpine meadow. J Mamm 84(2):644–651

AsefaetalMS3FunctionalTraitESMOecologia.docx

Download PDF

Journal Publication

published 01 Jun, 2024

Read the published version in Oecologia →

Reviewers agreed at journal
04 Sep, 2023
Reviewers invited by journal
01 Sep, 2023
Editor assigned by journal
14 Jul, 2023
First submitted to journal
13 Jul, 2023

You are reading this latest preprint version

Giant root-rat engineering and livestock grazing activities regulate plant functional trait diversity of an Afroalpine vegetation community in the Bale Mountains, Ethiopia

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and methods

Study area

Giant root-rat and its disturbances

Data collection

Human activities, root-rat burrow density and plant species abundance

Plant trait selection and data

Statistical analysis

Results

Discussion

Declarations

References

Supplementary Files

Status:

Journal Publication

Version 1