Exp Appl Acarol (2015) 67:357–372 DOI 10.1007/s10493-015-9955-7
Effects of topography on soil and litter mites (Acari: Oribatida, Mesostigmata) in a tropical monsoon forest in Southern Vietnam Maria A. Minor1 • Sergey G. Ermilov2
Received: 14 June 2015 / Accepted: 19 July 2015 / Published online: 25 July 2015 Ó Springer International Publishing Switzerland 2015
Abstract The effects of topographic variables (elevation above sea level, slope position, topographic (wetness) index, and global solar radiation) on mite abundances and on quantitative composition of Oribatida communities in soil and litter have been studied in six sites along a hill slope in a tropical lowland forest in the Bu Gia Map National Park, Southern Vietnam. A positive relationship existed between abundance and species richness of Oribatida in soil cores, and global solar radiation (W h m-2) which quantifies the total sun energy available to the local ecosystem. There was no significant relationship between abundance of Mesostigmata and topographic variables. The Oribatida community composition in soil and in litter was significantly different, with a large number of species unique to either litter or soil. The canonical correspondence analysis (CCA) showed that the topographic variables together explained 75 % (in litter) and 83 % (in soil) of the variation in Oribatida community structure. The species-topography relationship was globally significant in the litter, weaker in the soil; the eigenvalue of the CCA axis 1 (related to elevation and global solar radiation) was significant in both substrates. CCA ordinations identified groups of species associated with high landscape positions (hill crest, high elevation, high global solar radiation) versus species associated with low-lying landscape positions, where moisture tends to accumulate (hill footslope, low elevation, low solar radiation, high topographic index values). The importance of relief and geographical position for soil Oribatida is discussed. Keywords Microarthropods Abundance Species richness Community structure Substrate effects Topography Elevation Topographic index Solar radiation
& Maria A. Minor
[email protected] Sergey G. Ermilov
[email protected] 1
Institute of Agriculture and Environment, Massey University, Private Bag 11222, Palmerston North, New Zealand
2
Tyumen State University, Tyumen, Russia
123
358
Exp Appl Acarol (2015) 67:357–372
Introduction Free-living mites (Acari) comprise a major part of the soil microarthropod community. Mites, especially Oribatida, are well suited as indicator organisms for monitoring the impact of land use practices on soil biodiversity, because their population densities are high, they are sensitive to soil conditions, and their sampling methods are well-developed. However, in the tropics their use as indicators of soil conditions and ecosystem disturbance is still limited, due to poor knowledge of fauna and frequently due to the lack of baseline ecological data (Noti et al. 2003). The fauna of oribatid mites in Southern Vietnam is relatively well-studied: the species lists and taxonomic descriptions of Oribatida from Bu Gia Map National Park and other National Parks in Southern Vietnam have been published (Ermilov and Anichkin 2011, 2012; Ermilov 2011; Ermilov et al. 2012, 2013, 2014a b, c; Niedbała and Ermilov 2014). However, quantitative ecological information on soil communities in Vietnam is scarce (Vu and Nguyen 2000; Vu 2011; Nguyen and Vu 2012). The lack of detailed data on ecological structure and functioning of soil mite communities in tropical forests of Vietnam, and in South-East Asia in general, is a serious deficiency, especially as soils in this region are under increasing pressure from land development to accommodate growing population. Species-level identification and information on environmental preferences, distribution, and life histories of species in the community are of upmost importance for using community data in biodiversity conservation (Lawton et al. 1998; Franklin et al. 2007). Borcard and Legendre (1994) suggested gradient analysis as a successful approach to investigating environmental preferences of species, and for analysing relationships between spatial structure of communities and their environment. This approach allows to integrate information from a range of environmental variables and to quantify the spatial structure of soil communities at different scales. Factors such as the quality and quantity of organic matter (Hansen 1999; Hasegawa et al. 2006), soil moisture (Lindberg et al. 2002), tree species composition (Hasegawa et al. 2013), microhabitat heterogeneity (Hansen and Coleman 1998), soil nutrient content (Kaspari and Yanoviak 2009) and biotic factors (Bardgett et al. 1998; Rooney et al. 2000) have been linked to abundance, diversity and community patterns of soil microarthropods. Some of these factors (e.g., biotic factors such as predation and competition) can be difficult to observe and to measure. Local physical and chemical factors (vegetation, soil properties, soil water content, etc.) can be measured directly, but such field data are sometimes difficult or expensive to collect, especially in remote or poorly studied areas. In vegetation and soil studies, local measurements are frequently and successfully substituted with topographic variables, which are derived from digital elevation models (DEMs) and describe geomorphological features of the land surface. The DEMs are available as worldwide coverage (Jarvis et al. 2008), topographic variables are relatively easily derived, and have been used in many studies as proxies for predicting local physical and ecological conditions, such as soil properties, microclimate, hydrological regime, plant species distributions, etc. (e.g., Moore et al. 1993; Guisan and Zimmermann 2000; Matsuura and Suzuki 2013). The use of DEM-derived parameters is less frequent for soil fauna (but see Dress and Boerner 2004), yet it could enhance our understanding of distribution of soil communities in less-studied areas. In this paper we aim to integrate data on soil biodiversity and landscape-related factors, and investigate patterns of Acari (Mesostigmata and Oribatida) abundances and quantitative composition of Oribatida communities in soil and litter along a hill slope in response to
123
Exp Appl Acarol (2015) 67:357–372
359
topographic variables (elevation, slope position, topographic wetness index, and global solar radiation) in a tropical lowland forest in Bu Gia Map National Park, Southern Vietnam. We also aim to add to the general knowledge of soil Acari in tropical lowland forests.
Methods Study sites The Bu Gia Map National Park (12o050 N, 107o030 E to 12o180 N, 107o140 E) is located in the lowlands of Southern Vietnam in the north-east of the Binh Phuoc province, along the Vietnamese-Cambodian border. The total area of the park is 26,032 ha, including
Fig. 1 Location and elevational gradient of study sites in the Bu Gia Map National Park, Binh Phuoc, Southern Vietnam, 13–14 September, 2013
123
360
Exp Appl Acarol (2015) 67:357–372
18,100 ha of a strict protection area. The park is surrounded by 15,200 ha of increasingly populated buffer zone in Binh Phuoc and Dac Nong provinces. The terrain is moderately rugged (elevation 300–700 m above sea level) with numerous stream and river valleys, including the Dak Huyet and Dak Sam rivers (Sourcebook 2004). The highest point in the area is 738 m asl. The region has distinct dry (December–April) and rainy (May– November) seasons. The rainy season is characterized by high humidity and high temperatures (mean high temperature exceeding 25 °C) (World Weather OnlineÒ). The park is well-forested; natural forest makes up 21,376 ha. The main vegetation type is lowland tropical monsoon forest, represented by semi-deciduous mixed closed forest and evergreen closed forest. Open woodland is found on hill slopes, which is dominated by bamboo in many areas. There are approximately 808 species of vascular plants present in the park (Anon 1997). Samples were collected from six localities (sites) on 13–14 September 2013. The sites formed a topographic gradient from shoulder of the hill (site 1) to the low-lying footslope near the river (site 6) (Fig. 1). The brief description of collection sites is given in Table 1. Coordinates and elevation were recorded using GPS.
Sampling In each site, ten soil samples were collected using 7.8 cm dia., 10 cm deep corer. Leaf litter was brushed off prior to collection of soil cores. Intact cores in metal cylinders were transported to the laboratory, where soil mites were extracted for 10 days using Berlese funnels heated with 40 W lamps. Additionally, 16 litter samples were collected in each site using a 50 9 50 cm stainless steel frame. The litter samples were mixed together and passed through a sifter (mesh size 2 9 2 cm). The resulting fine fraction was extracted at room temperature for 20 days using Winkler extractor. Mites were extracted and stored in 75 % ethanol. Adult Oribatida were identified to a species level (the species check-list is given in Ermilov and Bayartogtokh 2015) except for ptyctimous mites, which were not identified further. Juvenile oribatid mites were included in the abundance counts, but excluded from species richness estimates and community composition analysis. Table 1 Study sites in the Bu Gia Map National Park, Binh Phuoc, Southern Vietnam, 13–14 September, 2013 Site
Coordinates, elevation (a.s.l.)
Description
1
12°110 31.2800 N, 107°120 13.0600 E, 570 m
Dipterocarp forest (Dipterocarpus costatus), crest of the hill, dark loamy soil
2
12°110 30.2400 N, 107°120 16.6300 E, 539 m
Palm forest (Arecaceae) in the gorge, dark loamy soil
3
12°120 02.3300 N, 107°120 17.4900 E, 408 m
Mixed forest (Anacardiaceae, Dipterocarpaceae, Verbenaceae, Irvingiaceae and other), middle of the slope, dark loamy soil
4
12°120 06.7200 N, 107°120 16.0000 E, 398 m
Mixed broadleaf-bamboo forest, middle of the slope, dark loamy soil
5
12°120 12.4900 N, 107°120 13.0200 E, 390 m
Mixed forest (Anacardiaceae, Dipterocarpacea, Verbenaceae, Irvingiaceae and other), lower slope, dark loamy soil
6
12°120 19.6100 N, 107°120 12.4200 E, 341 m
Mixed forest (Dilleniaceae, Lecythidaceae, Anacardiaceae, Verbenaceae, Irvingiaceae), footslope, dark loamy soil
123
Exp Appl Acarol (2015) 67:357–372
361
Topographic variables The topography-related parameters were derived from the DEM of the area. We used the SRTM v.4.1 (srtm_58_10.tif, Geotiff format, GCS WGS84, cell size 90 m) derived from the USGS/NASA SRTM data for the world (Jarvis et al. 2008; http://srtm.csi.cgiar.org). The DEM was filled, projected to UTM zone 48 N, clipped to the area of interest (Bu Gia Map National Park), manually registered with the topomap of the area, and used for geomorphometry. The ArcGIS Geomorphometry and Gradient Metrics Toolbox 1.0 (Evans et al. 2014) were used to derive slope position and topographic index. Additionally, the global solar radiation was derived from the DEM in Spatial Analyst: 1. 2.
Slope position—a scaled unitless variable representing position of the slope from top/ ridge (positive values) to footslope/depression (negative values). Topographic Index (TI) is a steady state soil wetness index (Gessler et al. 1995). The TI is a function of landscape position, and incorporates both the slope and the upslope contributing area: TI ¼ lnðAs= tan bÞ;
3.
where As is the upslope contributing area, calculated as (flow accumulation ? 1) 9 (cell size in m2), and b is the slope in radians. Global solar radiation (W h m-2) is the cumulative potential insolation based on incident solar radiation for a specified number of days of the year, calculated using solar radiation physics according to slope, aspect, and latitude (Rich et al. 1994; Fu and Rich 2002). We used the time period from 1 January 2013 to 13 September 2013 (the year of sampling) for this calculation.
Nearest neighbour sampling was used to extract geomorphometric information for the six sites. All data manipulation and analysis were done in ArcMap 10 (ESRI) with Spatial Analyst extension.
Statistical analysis Densities of mites in soil and litter were standardised to individuals m-2. The site differences were quantified using PROC GLM in SAS 9.3 (SAS Institute). Diversity of Oribatida was described as observed species richness, the Shannon’s diversity index, and expected species richness (Chao2 diversity estimate) using EstimateS 9.10 for Windows (Colwell 2013). The blocked multi-response permutation procedures method (MRBP) in PC-ORD (MjM software, version 5) was applied to species presence-absence matrix to tests the null hypothesis of no difference in community composition of Oribatida between soil and litter substrates (blocked by site). Presence-absence rather than abundances was used because soil and litter abundances are not directly comparable; rare species (\5 individuals across all sites) were excluded. Associations between species of Oribatida (presence-absence data) and types of substrate were analysed using the Indicator Species Analysis (Dufreˆne and Legendre 1997) in PC-ORD. The presence of directional trends for mite abundance (all three orders) and species richness (Oribatida only) in soil cores in response to topography was checked using OLS regression model in PROC REG in SAS 9.3. The topographic variables included elevation above sea level, slope position, topographic index, and global solar radiation for Jan 1–Sept 13 2013. Data values for this analysis were square-root transformed to reduce variance heterogeneity inherent to count data. The collinearity analysis indicated that some of
123
123
15.75
102.75
± 28.94
122.0
21.5
1192.88 ± 11.69
c
1778.85ab ± 29.14
Site 3
86.25
14.75
4625.02
ab
± 67.59
2762.45a ± 44.10
Site 4
8.75
3.5
2971.73
2
1
Pooled data (table values represent means, but variance estimates are not available); Winkler extraction of sifted litter
n = 10; Berlese extraction of soil cores
bc
± 40.77
1569.58ab ± 32.01
Site 5
Values are mean ± standard error. Within a row, means with the same letter are not significantly different (P [ 0.05, Tukey’s HSD)
7.75
71.75
Oribatida
1820.71
bc
6571.29 ± 145.53
1276.59b ± 16.04
a
Site 2
2720.60a ± 30.88
Mesostigmata
Litter
2
Oribatida
Mesostigmata
Soil1
Site 1
14.5
4.75
2385.75bc ± 45.55
1841.64ab ± 35.13
Site 6
Table 2 Density (ind. m-2) of Acari in the soil and litter along a tropical forest hillslope in the Bu Gia Map National Park, Southern Vietnam (September 2013)
362 Exp Appl Acarol (2015) 67:357–372
12
3.20
53.43 ± 5.77
H’
Chao2 ± SD
6
–
–
H’
Chao2 ± SD
–
–
H’
Chao2 ± SD
2
–
–
9
51
–
–
13
39
46.45 ± 11.08
2.69
4
30
Site 2
–
–
4
44
–
–
6
30
40.01 ± 14.85
2.78
4
21
Site 3
Pooled data (diversity estimates not available); Winkler extraction of sifted litter
n = 10; Berlese extraction of soil cores
11
Unique spp
1
56
Sobs
Total for soil and litter
23
Sobs
Unique spp
Litter2
45
Sobs
Unique spp
Soil1
Site 1
–
–
6
48
–
–
5
28
56.04 ± 16.18
2.77
6
33
Site 4
–
–
0
36
–
–
0
7
40.07 ± 7.26
3.0
0
31
Site 5
–
–
1
32
–
–
0
8
36.25 ± 7.07
2.99
3
27
Site 6
117.33 ± 10.04
3.47
–
96
94.25 ± 15.58
2.95
23
63
93.46 ± 9.63
3.49
32
74
All sites
Table 3 Diversity of oribatid mites (Acari: Oribatida) in the Bu Gia Map National Park, Southern Vietnam (September 2013)—observed species richness (Sobs), Shannon’s diversity index (H’), estimated species richness (classic Chao2)
Exp Appl Acarol (2015) 67:357–372 363
123
364
Exp Appl Acarol (2015) 67:357–372
topographic variables were collinear (due in part to the way they were derived); therefore, only two uncorrelated variables (global solar radiation and topographic index) representing two first Principal Component Analysis (PCA) axes were selected for the regression model. These two variables are also the most biologically meaningful, with global solar radiation representing total energy available to the ecosystem, and topographic index representing soil wetness pattern. The relationship between Oribatida communities and topography was investigated using canonical correspondence analysis (CCA) in PC-ORD. Soil and litter data were considered separately, and species with\10 individuals across all sites were excluded. To check if any of the variables would be redundant for CCA, correlation structure of topographic variables matrix was examined in PCA the global solar radiation and elevation were strongly associated with the first principal axis (correlation coefficients -0.957 and -0.952, respectively), while topographic index and slope position were correlated with the second principal axis (correlation coefficients -0.954 and 0.661); therefore, all four variables were included in CCA. In CCA settings, the data were biplot scaled, the rows and columns were compromised, and the null hypothesis of no relationships between community structure and topography was tested using a Monte Carlo test with 999 permutations. Significance level a = 0.05 was used for all statistical tests.
Results A total of 3442 Acari were collected in the samples; Oribatida (74.3 %) were most abundant, followed by Mesostigmata (24.5 %). Densities were higher in the soil samples (Table 2). Densities of Mesostigmata and Oribatida varied significantly among sites; the highest density for both groups was found in site 1 (Table 2). There was no significant relationship between abundance of Mesostigmata and topographic variables (reg. model F = 1.81, P = 0.17). For Oribatida, a positive relationship was detected between abundance (model F = 4.77, P = 0.012) and species richness (model F = 5.45, P = 0.007) in soil cores, and global solar radiation (abundance: t = 2.88, P = 0.006; species richness: t = 2.95, P = 0.005), but not topographic index (abundance: t = -0.40, P = 0.69; richness: t = -0.75, P = 0.46). In soil and litter combined, Oribatida belonging to 96 species were recorded (the species check-list is given in Ermilov and Bayartogtokh 2015), with variable diversity in soil and litter samples (Table 3). The Oribatida communities composition (based on presenceTable 4 Eigenvalues (inertias) of the CCA axes, and % variance explained in the ‘‘species’’ data set
Axis
Eigenvalue
% of variance explained This axis
Cumulative
Soil 1
0.306
43.0
43.0
2
0.166
23.4
66.4
3
0.114
16.1
82.5
1
0.379
35.9
35.9
2
0.250
23.7
59.6
3
0.158
15.0
74.6
Litter Total variance (‘‘inertia’’) in the species data: litter 1.0555, soil 0.7108
123
Exp Appl Acarol (2015) 67:357–372
365
absence) in soil and in litter was significantly different (MRPP for substrate, blocked by site: A = 0.1538, P = 0.007), with a large number of species unique to either litter or the soil (Table 3, unique spp all sites; see also ‘‘Appendix 1‘‘ for species-substrate associations). The CCA showed that the topographic variables together explain a very large proportion of the Oribatida community variation (Table 4). The species-topography relationship was globally significant in the litter (P = 0.024), but weaker in the soil (P = 0.050). However, the eigenvalue of the first axis in the soil was much higher than the range expected by chance (Monte-Carlo test P = 0.033, 998 runs); this axis is strongly related to elevation and global solar radiation. Correlations between the first three canonical axes and topographic variables are shown in Table 5. Figure 2 shows scatterplots of points representing topographic variables and species on axes 1–3. In the soil, species which contributed most to the inertia of axis 1 were: Oppiella nova and Pergalumna hauseri, associated with high elevation points and high global solar radiation; Scheloribates kraepelini, Ctenobelba bugiamapensis and Ramuselloppia vietnamica, associated with low elevation/low solar radiation (Fig. 2b, c). In litter, species with highest contribution to inertia of axis 1 were Peloribates kaszabi and Tegeozetes tunicatus (high elevation/high solar radiation); Ramusella insculpta and R. vietnamica (low elevation/low solar radiation) (Fig. 2a). Axis 2 identified a group of species associated with low-lying landscape position, where moisture tends to accumulate (high topographic index values). A second group of species is associated with high landscape positions which would be more free-draining (high slope position values). The results for the same species in litter and in the soil are not consistent—for example, O. nova in litter was associated with higher soil moisture in low-relief sites (Fig. 2a), whereas in the soil the same species was correlated more strongly with high elevation and high solar radiation (Fig. 2b, c). Some species had strong correlation with a particular axis without large contribution to its inertia; examples of these are Peloribates rangiroaensis in litter (axis 1, Fig. 2a) and Meristacarus sundensis in the soil (axis 2, Fig. 2b).
Discussion The observed densities of soil Mesostigmata and Oribatida are within the range reported for tropical forests in Vietnam (Vu and Nguyen 2000; Vu 2011) and elsewhere (PalaciosVargas et al. 2007; Banerjee et al. 2009; Kaspari and Yanoviak 2009). The results also Table 5 Correlations between the first three CCA axes and topographic variables
Axis 1
Axis 2
Axis 3
Soil Elevation Slope position Solar radiation Topographic index
0.902
-0.324
0.162
-0.407
0.724
-0.006
0.987
-0.146
-0.062
-0.247
-0.111
0.511 -0.003
Litter 0.845
-0.330
Slope position
Elevation
-0.347
0.843
0.342
Solar radiation
0.907
0.048
-0.187
Topographic index
0.225
-0.540
-0.310
123
366
Exp Appl Acarol (2015) 67:357–372
(a)
(b)
1.5
D.mon
s3
O.ger
P.ran
s5
P.par L.mol
Axis 2
N.jac T.ovu
S.kra T.tun
R.vie Ptyct
P.gla
s1
S.pra
TOPO_INDEX
s6 O.nova P.pse
-1.5
s2
D.mon
M.dor
s4
L.kue P.hir
M.sun
P.hau
P.pse
s1 A.ser
SOLAR S.fim
s3
ELEVATION J.kue
O.nova
P.ver
-1.0
G.don
POSITION
Ptyct E.bre
s6
0.0
SOLAR
U.sph ELEVATION
-0.5
s5
C.bug
Axis 2
0.5
1.0
P.kas
A.inc M.sun N.jac A.pus U.asi T.ovu s4 POSITION R.ins R.vie
V.sin
T.nip B.orn
S.pra P.par
G.pra
s2
-2.5 -1.5
-0.5
0.5
1.5
Axis 1
2.5
-2.0 -1.5
-0.5
0.5
1.5
Axis 1
(c) 2.0 C.bug
R.vie
1.0
s6
S.kra
Axis 3
A.ser
s5
S.fim
TOPO_INDEX N.jac
Ptyct
V.sin E.bre
0.0
P.par P.hir M.dor
P.hau
s1
O.nova
ELEVATION
s2
SOLAR L.kue
M.sun P.ver T.ovu
s3
P.pse D.mon
J.kue
s4
-1.0
S.pra
-1.5
-0.5
0.5
1.5
Axis 1
Fig. 2 CCA ordinations (LC scores) of points representing species and topographic variables on pairs of canonical axes. The vectors linking each point to the coordinate origin are shown for topographic variables but not shown for species. Points which are further away from the coordinate origin contribute more to the inertia of respective CCA axes. a Litter, axes 1 and 2; b soil, axes 1 and 2; c soil, axes 1 and 3. Legend: S1– S6—sites 1-6; POSITION—slope position; SOLAR – global solar radiation; TOPO_INDEX—topographic (wetness) index. Species: see ‘‘Appendix 2’’
highlight the importance of using complementary sampling techniques to achieve best community census, as indicated by the significant difference in Oribatida assemblages between soil and litter samples (based on presence-absence since abundances are not directly comparable due to different sampling techniques). Other authors observed distinct species composition of Oribatida communities in soil and litter layers (Luxton 1981; Berg et al. 1998; Osler et al. 2006) and reported groups of species associated with either litter or soil. We can only speculate what are the processes behind these species-substrate associations, but it is clear that the two substrates provide different living environments. We observed increase in abundance of Oribatida with higher global solar radiation; this variable reflects the solar energy input available to the ecosystem, and correlates with higher elevation of these sites. A number of other studies have observed the link between
123
Exp Appl Acarol (2015) 67:357–372
367
species richness and abundance of Oribatida and topographic factors, most frequently landscape position (Melamud et al. 2007; Illig et al. 2010). Dress and Boerner (2004) found that abundance of Oribatida in hardwood forests in Ohio was significantly greater in xeric (high and dry) landscape positions comparing to intermediate and mesic (low and wet) positions; landscape position accounted for *18 % of the variation in oribatid mite abundance in their study. For Mesostigmata, Dress and Boerner (2004) found no effect of landscape position on abundance, similar to what we observed. O’Lear and Blair (1999) found that while microarthropod abundance decreased with increasing soil moisture in a tallgrass prairie ecosystem, the effect was not the same for different taxa. Lopes de Gerenyu et al. (2011) suggested that the effects of topography on soil biota in a monsoon tropical forest can be linked to soil drainage; during the rainy period the soils in the depressed positions become oversaturated with moisture, which inhibits respiration by soil communities. Hasegawa et al. (2006) found that elevation influenced Oribatida in a tropical montane forest, but the effect of elevation was clearer in morphospecies richness and community composition, and less in total density. The elevation gradient in our study is likely to be too short to detect the effects of elevation per se, but our results suggest that the effects related to relative elevation and curvature of the land surface operate on all scales. The CCA ordinations show stratification of Oribatida assemblages in response to two groups of factors—first, solar energy input/elevation (CCA axis 1; these two variables were highly correlated), second, soil moisture regime, as slope position/topographic wetness index (CCA axis 2), with high slope position sites more free-draining, while high wetness index sites subject to moisture accumulation. The specific local associations (topoecogroups) of Oribatida could be identified, adding to the small pool of information on ecological preferences of tropical species. For example, Oppioidae such as R. vietnamica, R. insculpta and O. nova (in litter) were associated with low landscape positions/moisture accumulation areas. Woas (2002) suggested that in tropical rainforests higher population density of Oppioidea may reflect greater volume of decaying plant material in the soil. However, these preferences can be inconsistent between substrates—for example, in our results O. nova in the soil was associated with high elevation/high solar radiation sites, possibly reflecting different hydrological regimes in soil and litter. The very high proportion of species variation explained by topography in Bu Gia Map suggests that topography contributes strongly to the structuring of Oribatida communities. This is not surprising, as topography has a defining influence on soil environment. Soil scientists (e.g., Moore et al. 1993; Gessler et al. 1995) have demonstrated that catenary soil development is closely linked to the water flow paths through the terrain, and that on a meso-scale DEM-derived topographic variables such as slope and wetness index are significantly correlated with directly measured surface soil attributes, accounting for *50 % of the variability in A-horizon thickness, organic matter content, pH, extractable P, and soil texture. All of these soil attributes are important for soil microflora and fauna, and contribute to the structuring of soil mite communities (e.g., Bardgett et al. 1998; Hansen 1999; Lindberg et al. 2002; Hasegawa et al. 2006, 2013). Lopes de Gerenyu et al. (2011) highlighted the importance of topography for soil biota, and showed that in a monsoon tropical forests (Nam Kat Tien National Park, Southern Vietnam) the topographic position of the plots with similar type of vegetation and soil was the main factor influencing the integrated soil respiration rate (CO2 flux), which combines respiration by soil microorganisms, roots, and fauna. Topography also strongly influences physical and ecological conditions (e.g., solar radiation, soil moisture, slope position) that affect the distribution of vegetation, and is one of the main drivers behind vegetation cover, plant community composition, litter type, and
123
368
Exp Appl Acarol (2015) 67:357–372
other related factors (Franklin et al. 2000; Dress and Boerner 2004; Ediriweera et al. 2008; Matsuura and Suzuki 2013), all of which affect soil fauna. Ermilov and Bayartogtokh (2015) suggested existence of specific faunistic complexes of Oribatida related to certain types of plant associations in Bu Gia Map Park, although they only used presence-absence data, and did not separate litter and soil communities. Franklin et al. (2007) also found that the differences in vegetation cover and the litter type influence species composition of Oribatida in tropical settings. It should be said that, although the DEM-derived variables ‘‘explain’’ oribatid community structure very well in our study, they provide no insight into the mechanism of this influence, which we speculate could be related to soil/water/plant community response to topography. Soil biodiversity is the result of complex interactions of continental, regional, and local physical, chemical and biological processes. In tropical forests, as in other environments, factors such as climate, soil type, relief, vegetation cover and disturbance events interact to create complex patterns seen in soil communities. At a local level, patterns of microarthropod abundance and diversity are even more complex to predict, as they are determined by elements of spatial heterogeneity, such as local plant species and litter composition, microrelief, etc. (Striganova 1996; Dress and Boerner 2004). Future studies should aim to separate the effects of different factors by having replicated combinations of topographic and environmental variables. To conclude, our results show significant correlation between terrain attributes and assemblages of Oribatida in litter and soil of a tropical monsoon forest. The exact patterns of oribatid mite abundance and diversity probably reflect a response to a complex of terrain-dependant factors, such as soil moisture regime, vegetation, and soil attributes, rather than simply to elevation above sea level and slope position. Water and energy balance on the terrain have been suggested as good predictors for local distribution patterns of soil organisms (Melamud et al. 2007), and these are directly linked to topography. Because topographic variables can be remotely obtained (DEM, satellite data, etc.), they are useful to enhance our understanding of distribution of soil communities in less-studied areas. Acknowledgments We thank Dr. Andrey A. Yurtaev (Tyumen State University, Tyumen, Russia) for help in creating the maps of study sites in Fig. 1, Dr. Alexander E. Anichkin (Severtsov Institute of Ecology and Evolution, Russian Academy of Sciences, Moscow, Russia) for assistance in sampling and for description of vegetation in study sites, and the staff of the Bi Gia Map National Park for support during the field work. We also thank two anonymous reviewers for comments which helped to improve the manuscript. The study was supported by the Russian Foundation for Basic Research (Project 14-04-31183 mol_a).
Appendix 1 Indicator analysis (Dufreˆne and Legendre 1997) of species-substrate associations (based on presence-absence), Bu Gia Map National Park, Binh Phuoc, Southern Vietnam, 13–14 September, 2013. Species
Substrate
Indicator value
P value
Papillacarus hirsutus (Aoki, 1961)
Soil
100.0
0.002
Perxylobates vermiseta (Balogh and Mahunka, 1968)
‘‘
85.7
0.015
Javacarus kuehnelti Balogh, 1961
‘‘
83.3
0.015
Eremobelba breviseta Balogh, 1968
‘‘
83.3
0.018
123
Exp Appl Acarol (2015) 67:357–372
369
Species
Substrate
Indicator value
P value
Lasiobelba kuehnelti (Csisza´r, 1961)
‘‘
83.3
0.016
Tecteremaeus incompletus Mahunka, 1988
‘‘
66.7
0.057
Oribatella gerdweigmanni Ermilov and Anichkin, 2012
Litter
83.3
0.016
Peloribates rangiroaensis Hammer, 1972
‘‘
79.7
0.032
Afronothrus incisivus Wallwork, 1961
‘‘
66.7
0.061
Appendix 2 Species included in the CCA analysis, their ordination codes, and relative abundance in the soil and litter across all sites, Bu Gia Map National Park, Binh Phuoc, Southern Vietnam, 13–14 September, 2013. Rare species (\10 individuals across all sites) were excluded.
CCA ordination code
Species
Relative abundance (%) In litter
In soil 0.99
Ptyct
Ptyctimous mites
3.63
A.pus
Allozetes pusillus (Berlese, 1913)
0.40
2.81
A.ser
Arcoppia serrutala (Balogh & Mahunka, 1980)
3.10
0.08
A.inc
Afronothrus incisivus Wallwork, 1961
–
3.87
B.orn
Berlesezetes ornatissimus (Berlese, 1913)
1.21
0.91
C.bug
Ctenobelba bugiamapensis Ermilov, Shtanchaeva, Subı´as & Anichkin, 2014
1.75
0.15
D.mon
Dendrohermannia monstruosa (Aoki, 1977)
6.86
0.83
E.bre
Eremobelba breviseta Balogh, 1968
1.62
–
G.don
Galumna dongnaiensis Ermilov & Anichkin, 2013 Galumna praeoccupata Subı´as, 2004
0.40
0.91
–
0.99
G.pra J.kue
Javacarus kuehnelti Balogh, 1961
1.48
–
L.mol
0.94
16.22
L.kue
Lamellobates cf. molecula (Berlese, 1916) Lasiobelba kuehnelti (Csisza´r, 1961)
5.79
–
M.dor
Malaconothrus dorsofoveolatus Hammer, 1979
3.10
0.30
M.sun
Meristacarus sundensis Hammer, 1979
1.35
0.76
N.jac
Neoribates jacoti (Balogh & Mahunka, 1967)
1.62
2.65
O.nova
Oppiella nova (Oudemans, 1902)
2.42
1.52
O.ger
Oribatella gerdweigmanni Ermilov & Anichkin, 2012
–
4.40
P.hir
Papillacarus hirsutus (Aoki, 1961)
2.42
–
P.kas
Peloribates kaszabi Mahunka, 1988
0.27
0.99
P.ran
Peloribates rangiroaensis Hammer, 1972
0.94
11.75
P.hau
Pergalumna hauseri Mahunka, 1995
2.69
0.30
P.pseu
Pergalumna pseudosejugalis Ermilov & Anichkin, 2012
7.27
9.33
P.ver
Perxylobates vermiseta (Balogh & Mahunka, 1968)
9.56
0.08
P.gla
Phyllhermannia gladiata Aoki, 1965
1.08
2.81
P.par
Protoribates paracapucinus (Mahunka, 1988)
8.34
3.26
123
370
CCA ordination code
Exp Appl Acarol (2015) 67:357–372
Species
Relative abundance (%) In litter
In soil 0.91
R.ins
Ramusella insculpta (Paoli, 1908)
–
R.vie
Ramuselloppia vietnamica Ermilov & Anichkin, 2013
2.02
2.96
S.fim
Scheloribates fimbriatus Thor, 1930
2.69
0.15
S.kra
Scheloribates kraepelini (Berlese, 1908)
2.42
–
S.pra
Scheloribates praeincisus (Berlese, 1910)
3.10
18.95
T.tun
Tegeozetes tunicatus Berlese, 1913
0.13
2.50
T.ovu
Trachyoribates ovulum Berlese, 1908
11.71
2.50
T.nip
Trichogalumna cf. nipponica (Aoki, 1966)
–
0.91
U.asi
Unguizetes asiaticus Ermilov & Anichkin, 2012
0.27
0.91
U.sph
Unguizetes sphaerula (Berlese, 1905)
–
0.91
V.sin
Vilhenabates sinatus (Aoki, 1965)
1.35
0.30
References Anonymous (1997) Nature elements of Bu Gia Map Nature Reserve. Unpublished report to the Binh Phuoc Provincial Department of Science, Technology and the Environment [in Vietnamese]. Cited from: Tran DQ, Nguyen HTN, Tran HTM, Stephenson S (2014) Myxomycetes recorded from three lowland tropical forests in Vietnam. Mycosphere 5:662–672 Banerjee S, Sanyal AK, Moitra MN (2009) Abundance and group diversity of soil mite population in relation to four edaphic factors at Chintamani Abhayaranya, Narendrapur, South 24-Parganas, West Bengal. Proc Zool Soc (Calcutta) 62:57–65 Bardgett RD, Keiler S, Cook R, Gilburn AS (1998) Dynamic interactions between soil animals and microorganisms in upland grassland soils amended with sheep dung: a microcosm study. Soil Biol Biochem 30:531–539 Berg MP, Kniese JP, Bedaux JJM, Verhoef HA (1998) Dynamics and stratification of functional groups of micro- and mesoarthropods in the organic layer of a Scots pine forest. Biol Fert Soils 26:268–284 Borcard D, Legendre P (1994) Environmental control and spatial structure in ecological communities: an example using oribatid mites (Acari, Oribatei). Environ Ecol Stat 1:37–61 Colwell RK (2013) EstimateS: statistical estimation of species richness and shared species from samples. Version 9. User’s guide and application. http://purl.oclc.org/estimates. Accessed 21 April 2015 Dress WJ, Boerner REJ (2004) Patterns of microarthropod abundance in oak–hickory forest ecosystems in relation to prescribed fire and landscape position. Pedobiologia 48:1–8 Dufreˆne M, Legendre P (1997) Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol Monogr 67:345–366 Ediriweera S, Singhakumara BMP, Ashton MS (2008) Variation in canopy structure, light and soil nutrition across elevation of a Sri Lankan tropical rain forest. For Ecol Manag 256:1339–1349 Ermilov SG (2011) A new genus and species of Amerobelbidae (Acari, Oribatida) from Vietnam. Acarologia 51:275–282 Ermilov SG, Anichkin AE (2011) New oribatid mites of the genera Pulchroppia and Lineoppia (Acari: Oribatida: Oppiidae) from Cat Tien National Park in southern Vietnam. Acarologia 51:31–42 Ermilov SG, Anichkin AE (2012) Two new species of oribatid mites (Acari: Oribatida) from Bu Gia Map National Park (Vietnam). Zoosystematica Rossica 21:18–27 Ermilov SG, Bayartogtokh B (2015) The oribatid mite fauna (Acari, Oribatida) of the Bu Gia Map National Park (southern Vietnam), with description of two new species. Int J Acarol 41:220–231 Ermilov SG, Anichkin AE, Wu D (2012) Oribatid mites from Bu Gia Map National Park (Southern Vietnam), with description of a new species of Dolicheremaeus (Tetracondylidae) (Acari: Oribatida). Genus 23:591–601 Ermilov SG, Anichkin AE, Wu D (2013) A new species of oribatid mites of the genus Sphaerozetes (Oribatida, Ceratozetidae) from Vietnam. Entomol Rev 94:131–135
123
Exp Appl Acarol (2015) 67:357–372
371
Ermilov SG, Anichkin AE, Tolstikov AV (2014a) The oribatid mite genus Papillocepheus (Acari, Oribatida, Tetracondylidae), with description of a new species from southern Vietnam. ZooKeys 381:1–10 Ermilov SG, Anichkin AE, Tolstikov AV (2014b) Umashtanchaeviella plethotricha, a new genus and species of the family Tetracondylidae (Acari, Oribatida). ZooKeys 408:51–59 Ermilov SG, Shtanchaeva UY, Subı´as LS, Anichkin AE (2014c) The family Ctenobelbidae (Acari, Oribatida), with description of a new species and discussion on systematic placement and taxonomic status of the genus Berndamerus Mahunka, 1977. ZooKeys 395:1–10 Evans JS, Oakleaf J, Cushman SA, Theobald D (2014) An ArcGIS Toolbox for surface gradient and geomorphometric modeling, version 2.0-0. http://evansmurphy.wix.com/evansspatial. Accessed 30 April 2015 Franklin J, McCullough P, Gray C (2000) Terrain variables used for predictive mapping of vegetation communities in Southern California. In: Wilson JP, Gallant JC (eds) Terrain analysis—principles and applications. Wiley, New York, pp 331–353 Franklin E, Santos EMR, Albuquerque MIC (2007) Edaphic and arboricolous oribatid mites (Acari; Oribatida) in tropical environments: changes in the distribution of higher level taxonomic groups in the communities of species. Braz J Biol 67:447–458 Fu P, Rich PM (2002) A geometric solar radiation model with applications in agriculture and forestry. Comput Electron Agric 37:25–35 Gessler PE, Moore ID, McKenzie NJ, Ryan PJ (1995) Soil-landscape modelling and spatial prediction of soil attributes. Int J Geogr Inf Syst 9:421–432 Guisan A, Zimmermann NE (2000) Predictive habitat distribution models in ecology. Ecol Model 135:147–186 Hansen RA (1999) Red oak litter promotes a microarthropod functional group that accelerates its decomposition. Plant Soil 209:37–45 Hansen RA, Coleman DC (1998) Litter complexity and composition are determinants of the diversity and species composition of oribatid mites (Acari:Oribatida) in litterbags. Appl Soil Ecol 9:17–23 Hasegawa M, Ito MT, Kitayama K (2006) Community structure of oribatid mites in relation to elevation and geology on the slope of Mount Kinabalu, Sabah, Malaysia. Eur J Soil Biol 42:S191–S196 Hasegawa M, Okabe K, Fukuyama K, Makino S, Okochi I, Tanaka H, Goto H, Mizoguchi T, Sakata T (2013) Community structures of Mesostigmata, Prostigmata and Oribatida in broad-leaved regeneration forests and conifer plantations of various ages. Exp Appl Acarol 59:391–408 Illig J, Norton RA, Scheu S, Maraun M (2010) Density and community structure of soil- and bark-dwelling microarthropods along an altitudinal gradient in a tropical montane rainforest. Exp Appl Acarol 52:49–62 Jarvis A, Reuter HI, Nelson A, Guevara E (2008) Hole-filled seamless SRTM data V4, International Centre for Tropical Agriculture (CIAT). http://srtm.csi.cgiar.org. Accessed 30 April 2015 Kaspari M, Yanoviak SP (2009) Biogeochemistry and the structure of tropical brown food webs. Ecology 90:3342–3351 Lawton JH, Naeem S, Thompson LJ, Hector A, Crawley MJ (1998) Biodiversity and ecosystem function: getting the Ecotron experiment in its correct context. Funct Ecol 12:848–852 Lindberg N, Engtsson JB, Persson T (2002) Effects of experimental irrigation and drought on the composition and diversity of soil fauna in a coniferous stand. J Appl Ecol 39:924–936 Lopes de Gerenyu VO, Kurbatova YA, Kurganova IN, Tiunov AV, Anichkin AY, Myakshina TN, Kuznetsov AN (2011) Daily and seasonal dynamics of CO2 fluxes from soils under different stands of monsoon tropical forest. Eurasian Soil Sci 44:984–990 Luxton M (1981) Studies on the oribatid mites of a Danish beech wood soil. V. Vertical distribution. Pedobiologia 21:365–386 Matsuura T, Suzuki W (2013) Analysis of topography and vegetation distribution using a digital elevation model: case study of a snowy mountain basin in northeastern Japan. Landsc Ecol Eng 9:143–155 Melamud V, Beharav A, Pavlicek T, Nevo E (2007) Biodiversity interslope divergence of oribatid mites at ‘‘Evolution Canyon’’ Mount Carmel, Israel. Acta Zool Acad Sci Hung 53:381–396 Moore D, Gessler PE, Nielsen GA, Peterson GA (1993) Soil attribute prediction using terrain analysis. Soil Sci Soc Am J 57:443–452 Nguyen HT, Vu QM (2012) Oribatid mite (Acari: Oribatida) community structures—bioindicators of environment changes in Phong Nha–Ke Bang National Park, province of Quang Binh. J Plant Protect 41:41–44 (in Vietnamese with English summary) Niedbała W, Ermilov SG (2014) Ptyctimous mites (Acari, Oribatida) from the joint Russian-vietnamese biological expedition (October 2013–April 2014). Zootaxa 3884:156–168
123
372
Exp Appl Acarol (2015) 67:357–372
Noti MI, Andre´ HW, Ducarme X, Lebrun P (2003) Diversity of soil oribatid mites (Acari:Oribatida) from high katanga (Democratic Republic of Congo): a multiscale and multifactor approach. Biodivers Conserv 12:767–785 O’Lear HA, Blair JM (1999) Responses of soil microarthropods to changes in soil water availability in tallgrass prairie. Biol Fert Soils 29:207–217 Osler GHR, Korycinska A, Cole L (2006) Differences in litter mass change mite assemblage structure on a deciduous forest floor. Ecography 29:811–818 Palacios-Vargas JG, Castano-Meneses G, Gomez-Anaya JA, Martı´nez-Yrizar A, Mejı´a-Recamier BE, Martı´nez-Sa´nchez J (2007) Litter and soil arthropods diversity and density in a tropical dry forest ecosystem in Western Mexico. Biodivers Conserv 16:3703–3717 Rich PM, Dubayah R, Hetrick WA, Saving SC (1994) Using viewshed models to calculate intercepted solar radiation: applications in ecology. ASPRS technical papers, p 524–529 Rooney TP, Antolik C, Moran MD (2000) The impact of salamander predation on Collembola abundance. Proc Entomol Soc Wash 102:308–312 Sourcebook (2004). Sourcebook of existing and proposed protected areas in Vietnam, 2nd edn. BirdLife International in Indochina and Ministry of Agriculture and Rural Development. http://thiennhienviet. org.vn/sourcebook/source_book/index_EN.html. Accessed 15 April 2015 Striganova B (1996) Transect approach to the assessment of soil macrofauna diversity. Biol Int 33:17–23 Vu QM (2011) Oribatid soil mite (Acari: Oribatida) of northern Vietnam: species distribution and densities according to soil and habitat type. Pan-Pacific Entomol 87:209–222 Vu QM, Nguyen TT (2000) Microarthropod community structures (Oribatei and Collembola) in Tam Dao National Park, Vietnam. J Biosci 25:379–386 Woas S (2002) Acari. In: Adis J (ed) Amazonian Arachnida and Myriapoda. Pensoft, Sofia-Moscow, pp 21–291
123
