The effects of grazing management on plant characteristics
Investigations into FG, RG, and NG sites across two rural communities of Artashavan and Avan show a significant effect of grazing on the study plant parameters. Significant differences are established between mean values of height, above-ground biomass, below-ground biomass, grasses, legumes, forbs, and root-to-shoot (Table 2, Fig. 2).
Table 2
The effects of grazing management strategies on plant biomass and community parameters (means + SE)
Sites/ management | Artashavan | Avan |
NG | RG | FG | NG | RG | FG |
H, cm | 33.93 ± 0.77a | 22.90 ± 0.66b | 11.76 ± 0.40c | 38.15 ± 0.71a | 24.38 ± 0.84b | 5.53 ± 0.35c |
AGB, g m− 2 | 266.0 ± 9.2a | 133.5 ± 4.3b | 86.7 ± 2.8c | 250.1 ± 12.1a | 105.8 ± 5.0b | 63.9 ± 3.1c |
G, g m− 2 | 106.5 ± 7.87a | 43.96 ± 3.70b | 19.19 ± 1.73c | 108.60 ± 10.70a | 41.41 ± 4.24b | 10.10 ± 1.11c |
L,g m− 2 | 85.17 ± 8.59a | 26.51 ± 2.49b | 4.14 ± 0.50c | 39.21 ± 5.70a | 6.73 ± 0.96b | 2.77 ± 0.62b |
F, g m− 2 | 74.26 ± 5.39a | 62.99 ± 4.67a | 63.40 ± 2.81a | 102.24 ± 7.82a | 57.61 ± 4.36b | 51.02 ± 2.97b |
BGB, g m− 2 | 1857.0 ± 67.5a | 2100.0 ± 62.87a | 1145.1 ± 35.4b | 1476.0 ± 49.3a | 1512.0 ± 27.5a | 1259.0 ± 49.4b |
R:S | 6.98 ± 0.36 | 15.73 ± 1.02 | 13.20 ± 0.68 | 5.90 ± 0.09 | 14.29 ± 0.021 | 19.70 ± 1.69 |
Different lowercase letters indicate significant differences in means among different grazing management (ANOVA, Tukey's test, significance level α < 0.05). |
For all parameters, the difference between the three sites with different grazing is maximally significant (p < 0.001), with the exception of grass biomass in the RG and FG Artashavan sites (p = 0.128; F = 0.09). The height of vegetation cover is maximal during the peak of the vegetative period on NG sites Avan (56.1cm) and Artashavan (49.5cm), which is 1.5 times higher vs. RG management sites in both communities and respectively 6.9 and 2.9 times higher as compared with FG sites in both rural communities. Almost the same regularity is established for AGB: the biomass on the NG site in Avan is in excess of 2.4 and 3.9 times as compared to RG and FG sites. For Artashavan sites these values make 2.0 and 3.1, respectively. Of the studied plant groups, the strongest impact of grazing is exposed grasses and legumes. On average, the biomass of legumes, grasses, and forbs on FG versus NG sites Artashavan is lower 18.0; 7.3; and 1.5 times, respectively. A root-to-shoot is the highest on FG and RG sites. At the same time, the size of AGB, BGB, and root-to-shoot on all sites in Artashavan are always reliably higher as compared to Avan (6.98 and 5.90, respectively).
Redundancy analysis (RA) is used to test the response of plant community parameters and soil nutrient properties to grazing management strategies. Figure 3 clearly shows the relationships between soil nutrient variables, plant biomass and species, and different grazing strategies. The magnitude of the vectors in the ordinate space indicates the strength of the contribution of each variable, angles in between – and the strength of relations.
The parameters: height, AGB, grasses, legumes, forbs, and BGB are linearly interrelated, positively correlated with the NG management strategies, and negatively - with a root system. These variables explain the variance F1 = 75.83%. Soil nitrogen, phosphorus, and potassium strongly correlate with each other and RG plots and less strongly - with biomass and plant species. Organic carbon content correlates with FG, while no relationship with plant variables has been established for hydrogen. Soil variables explain F2 = 22.33% of the variance. Overall, factors F1 and F2 explain 98.17% of the variance (F = 6.77, p = 0.086) and are not linearly related (α > 0.05).
Effect of grazing management on soil and plants nutrient
The results of a three-year long research, parameters of soil, AGB and BGB are given in Table 3.
Table 3
The effect of grazing management on soil (depth: 0–20 cm), AGB and BGB quality parameters (mean ± SE)
Sites/ management | Artashavan | Avan |
NG | RG | FG | NG | RG | FG |
Soil parameters |
N, % | 0.32 ± 0.03a | 0.34 ± 0.01ab | 0.40 ± 0.01b | 0.34 ± 0.01a | 0.34 ± 0.02a | 0.43 ± 0.01b |
P, % | 0.34 ± 0.01a | 0.45 ± 0.02b | 0.46 ± 0.03b | 0.30 ± 0.01a | 0.39 ± 0.02b | 0.43 ± 0.01b |
K, % | 0.35 ± 0.02a | 0.38 ± 0.04ab | 0.45 ± 0.01b | 0.25 ± 0.01a | 0.30 ± 0.03a | 0.32 ± 0.01a |
OC, % | 2.36 ± 0.23a | 2.48 ± 0.18a | 2.91 ± 0.24a | 1.64 ± 0.12a | 2.38 ± 0.05b | 2.52 ± 0.08b |
pH | 6.71 ± 0.07a | 6.73 ± 0.03a | 6.75 ± 0.04a | 7.43 ± 0.07a | 7.33 ± 0.09a | 7.38 ± 0.07a |
Above-ground biomass parameters |
N, % | 1.15 ± 0.04a | 1.31 ± 0.06a | 1.23 ± 0.04a | 0.89 ± 0.03a | 1.01 ± 0.05b | 1.17 ± 0.01c |
P, % | 0.82 ± 0.01a | 0.92 ± 0.01b | 0.90 ± 0.02b | 1.08 ± 0.02a | 1.02 ± 0.02a | 1.06 ± 0.02a |
K, % | 1.18 ± 0.03a | 1.16 ± 0.02a | 0.97 ± 0.03b | 1.12 ± 0.05a | 1.28 ± 0.01a | 1.28 ± 0.04a |
Below-ground biomass parameters |
N, % | 1.08 ± 0.03ab | 1.14 ± 0.03a | 1.03 ± 0.02b | 0.89 ± 0.01a | 1.01 ± 0.02b | 1.13 ± 0.03c |
P, % | 0.88 ± 0.02a | 0.88 ± 0.04a | 1.09 ± 0.02b | 1.01 ± 0.05a | 1.02 ± 0.04a | 1.13 ± 0.04a |
K, % | 0.40 ± 0.02a | 0.37 ± 0.02ab | 0.33 ± 0.01b | 0.25 ± 0.01a | 0.24 ± 0.01a | 0.30 ± 0.01b |
Different lowercase letters indicate significant differences in means among different grazing management (ANOVA, Tukey's test, significance level α < 0.05). |
The concentration of soil nitrogen on the FG site in Artashavan statistically credibly exceeds that on NG sites by 1.21 times, and in Avan – 1.29 times. For phosphorus, these values make 1.37 and 1.42, respectively. Potassium concentration in FG versus NG soils in Artashavan is excessive at 1.29 times, and organic carbon in Avan − 1.54 times. In other cases, no significant differences are observed between the average values for different grazing areas, with the exception of total nitrogen (p < 0.001, F = 25.5) and organic carbon (p < 0.001, F = 28.3) in Avan, as well as phosphorus in Artashavan (p < 0.01, F = 8.9) and Avan (p < 0.001, F = 18.0) areas. For the rest of the substances, grazing management has had no considerable effect on soil properties; in contrast to plant parameters, the differences between mean values for soils are less significant. For different management strategies, significant differences between mean concentrations of nutrients in AGB are observed for Avan sites by total nitrogen (р<0,001, F = 63.3) and phosphorus (р<0,001, F = 13.0), for Artashavan sites – by potassium (р<0,001, F = 11.5); in BGB – total nitrogen (р<0,001, 32.1) and potassium (р<0,001, F = 13.0) (Avan) as well as phosphorus (р<0,001, F = 21.8) (Artashavan). It is worth noting, however, that no clear regularity regarding changes in concentrations of the studied parameters under different grazing strategies has been established; nitrogen contents in AGB on FG sites exceed those on NG sites 1.10–1.54 times, phosphorus contents in BGB is 1.12–1.23 times excessive. In the rest of cases, the concentrations of nutrients are lower or equal for sites under different grazing loads.
Pearson's correlation coefficient was calculated to identify relationships between soil nutrients and AGB and BGB under different grazing management strategies. At a significance level of α = 0.05, the nutrient content in the soil does not indicate a significant relationship with the content of AGB and BGB. The exception is the positive correlation of potassium content in the soils of the NG and AGB (K = 0.73) and BGB (K = 0.69) plots and the negative - in the FG plot in AGB (K = -0.75). Thus, the nutrient content of AGB does not always depend on the nutrient content of the soil. At the same time, the nutrient contents of AGB and BGB show good correlation; the Pearson correlation coefficient for nitrogen at the NG, RG, and FG sites is 0.64, 0.58, and 0.71, respectively; for phosphorus 0.55, 0.58, and 0.41; for potassium 0.80, -0.45, 0.41.
The relationship between soils, AGB, and BGB properties under different grazing strategies according to redundancy analysis (RDA2) is presented in Fig. 4.
The FG management strategy positively correlates with all studied quality parameters of AGB and BGB with the exception of potassium contents in the case of RG; plant root potassium negatively correlates with its contents in AGB. The parameters of plants make 76.18% dispersion; their contents in ABG and BGB show good correlation and also correlate well with FG and RG management strategies. The relationship between soil nutrients and a management strategy is less distinct. The contents of potassium in soils negatively correlate with those of potassium and nitrogen in plants. Soil variables account for 17.38% dispersion. Wholly, F1, and F2 factors explain 91.67% dispersion (F = 3.482, p-value = 0.002) and are linearly interrelated (α < 0.05).