Fig 1.
Study area and experimental plots.
Light and dark colors show low and high elevation, respectively.
Table 1.
Geographical data and abbreviations for nine mountains surveyed on the east of the Loess Plateau with two, four, and two mountains in its northern, central, and southern parts, respectively.
Table 2.
Demarcations on geographical gradient belts of various mountains.
Latitudinal, longitudinal, and elevational gradients were divided into five, five, and six belts with intervals of 0.5°, 0.45°, and 100 m, respectively. The initial belts were all numbered 1 with different gradient ranges. Mountain names indicated by abbreviated letters are shown in Table 1.
Fig 2.
Spatial distribution of α-diversity in subalpine meadows.
α-Diversity indices included Simpson, Shannon, Pielou, and Patrick indices. The data were collected from six quadrats on each of nine mountain sites (54 quadrats in total); means for each mountain site were used in the analysis. Therefore, each α-diversity index had nine values, one for each site.
Table 3.
Correlation and significance analyses of α-diversity with elevation and among various mountains, respectively.
α-Diversity indices included the Patrick, Simpson, Shannon, and Pielou indices. Mountains were DD, BT, DT, ML, HY, YZ, YD, SU, and SE; mountain names indicated by abbreviated letters are shown in Table 1. Different or the same small letters present significant or insignificant differences (P<0.05 and P>0.05), respectively. The data for correlation coefficients in the analyses of α-diversity with elevation are listed with the corresponding P values in brackets. In significance analyses of α-diversity among various mountains, significance levels were expressed with different small letters. The data were collected from six quadrats on each of nine mountain sites (54 quadrats in total); means for each mountain site were used in the analysis. Therefore, each mountain site had one elevation, so nine α-diversity values were used in each correlation analysis. However, the six plots in each mountain were not averaged, so six value were used in each significance analysis for each site.
Fig 3.
Variations of α-diversity with latitude and longitude in subalpine meadows.
α-Diversity indices included the Patrick, Simpson, Shannon, and Pielou indices. The data were collected from six quadrats on each of nine mountain sites (54 quadrats in total) with different latitude and longitude; means for each mountain site were used in the analysis. Therefore, each index had nine values, one for each site.
Fig 4.
Changes on spatial distribution of β-diversity in subalpine meadows.
β-Diversity indices included the Cody, Sørenson, and Bray-Curtis indices. Numbers from 1 to 6 in x-coordinates represented corresponding numbers of latitudinal, longitudinal, and elevational belts in Table 2. Intervals of 1–2, 2–3, 3–4, 4–5, and 5–6 indicated comparisons between adjacent belts, whereas 1–2, 1–3, 1–4, 1–5, and 1–6 indicated comparisons of the initial belt (belt 1) with other belts. The study employed five latitudinal, five longitudinal, and six elevational belts. Each index had values for four latitudinal, four longitudinal, and five elevational belts.
Fig 5.
Spatial distribution of γ-diversity in subalpine meadows.
The study employed corresponding belts along five latitudinal, five longitudinal, and six elevational gradients, with five, five, and six values respectively. Each belt had one value.
Table 4.
Correlation analyses among α, β, and γ diversity.
βC is Cody index, βS is Sørenson index, and βB–C is Bray-Curtis index in β-diversity. There are 7, 7 and 9 gradient belts with latitude, longitude and altitude, respectively. Simulated values of β-diversity are calculated with common functions of β = γ/α, β = γ–α, and β = 1–α/γ. α and γ diversity are recalculated basing on species number in each gradient belt. So 23 data are shown for each diversity. The data are correlation coefficients and the levels of correlation analyses are all significant (P<0.01).
Table 5.
Spatial distribution of biomass in subalpine meadows.
Mountain names indicated by abbreviated letters are shown in Table 1. AB, BB, TB, and R/S represent aboveground, belowground, and total biomass, as well as the root:shoot ratio, respectively. Different or the same small letters indicate significant and insignificant differences (P<0.05 and P>0.05), respectively. The data are shown with mean ± S.E. with six values given for each mountain.
Fig 6.
Variations of biomass with latitude and longitude in subalpine meadows.
Biomass indices included aboveground, belowground, and total biomass as well as the root:shoot ratio. The data were collected from six quadrats on each of nine mountain sites (54 quadrats in total); means for each mountain site were used in the analysis. Therefore, each index had nine values.
Fig 7.
Changes of biomass with elevation in subalpine meadows.
The data were collected from six quadrats on each of nine mountain sites (54 quadrats in total); means for each mountain site were used in the analysis. Therefore, each biomass index had nine values.
Fig 8.
Regression analysis between α-diversity and biomass in subalpine meadows.
The data were collected from six quadrats on each of nine mountain sites (54 quadrats in total); means for each mountain site were used in the analysis. Therefore, each α-diversity and biomass index had 54 values. Significant relationships were selected between α-diversity and biomass indices.