Fig. 1.
Flowchart of research
Source: own elaboration.
Fig. 2.
Research study area of Bornuur sub-province, Tuv province
Source: own elaboration.
Fig. 3.
Field work points location
Source: own elaboration.
Table 1.
Summary of hayfield yield and forage quality (Bornuur)
| Variable | Min | Max | Mean (approx.) |
|---|---|---|---|
| Area (ha) | 10.3 | 6 000 | – |
| Yield (g/m2) | 54.1 | 132.9 | 93 |
| Moisture (%) | 4.92 | 7.69 | 5.9 |
| Dry matter (%) | 92.31 | 95.08 | 94.1 |
| Crude protein (%) | 9.12 | 16.56 | 11.9 |
| Fibre (%) | 22.8 | 36.96 | 30.1 |
| Digestibility (%) | 55.66 | 68.59 | 61.7 |
| Energy (MJ) | 8.02 | 9.87 | 8.8 |
[i] Source: own elaboration. Across all sampling sites, the mean forage yield was 9.34 c/ha (SD = 2.52), indicating moderate spatial variability. Nutritional quality exhibited a mean value of 8.80 (SD = 0.55) with low dispersion and near-normal distribution, suggesting relatively stable forage quality conditions across the study area.
Fig. 4.
Land surface factors map of Bornuur sub-province. a – elevation, b – slope, c – aspect
Source: own elaboration.
Table 2.
Descriptive statistical results of Land surface factors
| Land surface factor | Mean | Std | Var | Median | Skewness | Kurtosis | Q25 | Q75 |
|---|---|---|---|---|---|---|---|---|
| Elevation | 1 114.7 | 84.1 | 7 060.9 | 1133.0 | –0.6 | 3.3 | 1 050.8 | 1 159.5 |
| Aspect | 166.6 | 118.6 | 14 076.4 | 160.2 | 0.1 | 1.8 | 58.3 | 250.1 |
| Slope | 7.1 | 2.4 | 5.5 | 6.9 | –0.2 | 2.9 | 5.7 | 8.7 |
Table 3.
Descriptive statistical results of vegetation indices
| Indices | Mean | Std | Var | Median | Skewness | Kurtosis | Q25 | Q75 |
|---|---|---|---|---|---|---|---|---|
| NDVI | 0.702038 | 0.083245 | 0.006930 | 0.723672 | –0.071365 | 2.045454 | 0.642460 | 0.756097 |
| NDWI | –0.538684 | 0.026957 | 0.000727 | –0.549065 | 1.081511 | 3.079622 | –0.554977 | –0.531618 |
| SAVI | 0.940731 | 0.120926 | 0.014623 | 0.986801 | –1.529100 | 4.264492 | 0.929660 | 1.006785 |
| EVI | 0.505570 | 0.088888 | 0.007901 | 0.491353 | –0.698589 | 4.182447 | 0.460148 | 0.579817 |
Fig. 5.
Vegetation indices map of Bornuur sub-province. a – EVI, b – NDWI, c – NDVI, d – SAVI
Source: own elaboration.
Fig. 6.
Correlation matrix of variables *p < 0.05, **p < 0.01, ***p < 0.001. Values without asterisks are not statistically significant
Source: own elaboration.
Fig. 7.
Linear regression plots of forage yield versus vegetation indices (NDVI, EVI, SAVI, NDWI)
Source: own elaboration.
Fig. 8.
Linear regression analysis: Yield and land surface factors
Source: own elaboration.
Table 5.
Variable importance estimates from multivariate analysis
| Variable | Estimate | Std. error | p-value | 95% CI |
|---|---|---|---|---|
| NDVI | 18.827 | 6.659 | 0.022* | (3.470, 34.184) |
| EVI | 7.855 | 8.227 | 0.368 | (–11.115, 26.826) |
| NDWI | 7.797 | 16.349 | 0.646 | (–29.905, 45.499) |
| Slope | –0.289 | 0.242 | 0.266 | (–0.846, 0.268) |