We used unmanned aerial vehicles (UAVs) to carry out a relatively complete population census of large wild herbivores in Maduo County on the Tibetan Plateau in the spring of 2017. The effective area covered by aerial surveys was 326.6 km², and 23,784 images were acquired. Interpretation tag libraries for UAV images were created for wild animals, including Kiang (Equus kiang), Tibetan gazelle (Procapra picticaudata), and blue sheep (Pseudois nayaur), as well as livestock, including yaks and Tibetan sheep. Large wild herbivores in the survey transect were identified through manual imagery interpretation. Densities ranged from 1.15/km² for Kiang, 0.61/km² for Tibetan gazelle, 0.62/km² for blue sheep, 4.12/km² for domestic yak, and 7.34/km² for domestic sheep. A method based on meadows in the cold and warm seasons was used for estimating the densities and numbers of large wild herbivores and livestock, and was verified against records of livestock numbers. Population estimates for Kiang, Tibetan gazelle, blue sheep, domestic yak, and domestic sheep were 17,109, 15,961, 9324, 70,846, and 102,194, respectively. Based on published consumption estimates, the results suggest that domestic stock consume 4.5 times the amount of vegetation of large wild herbivores. Compared with traditional ground survey methods, performance of UAV remote sensing surveys of large wild herbivore populations was fast, economical and reliable, providing an effective future method for surveying wild animals.

The Loess Plateau is one of the most fragile areas in the world, where the problem of soil erosion is particularly prominent. The spatial and temporal variation characteristics and mechanisms of soil erosion in this region have always been hot topics for researchers. In this study, Revised Universal Soil Loss Equation (RUSLE) is used to estimate the soil erosion modulus of the Loess Plateau from 2000 to 2015, the dynamic characteristics of its temporal and spatial variations and driving mechanisms are determined, and meteorological data are combined with remote sensing data to quantitatively calculate the contribution rate of human activities. The results show that from 2000 to 2015, the soil erosion modulus of the Loess Plateau had a downward trend as a whole, with a rate of −0.6408 t/ha/a, but the downward trend gradually slowed down. Precipitation mainly resulted in changes in the soil erosion modulus in the northwestern part of the Loess Plateau, where a significant positive correlation was seen. Meanwhile, the Vegetation Fractional Coverage (VFC) mainly affected the southeastern part, where a significant negative correlation was measured. The human-activity contribution rate was −1.0774 on the Loess Plateau, which means human activities effectively reduced the soil erosion modulus while climate change promoted soil erosion combined with the result of the analysis of variance (ANOVA). “Hilly and gully regions” and “Gully region of Loess Plateau” as the main implementation areas of ecological projects, human activities had contribution rate of 0.5513 and 0.7805 toward the declining of soil erosion, respectively. Interestingly, the spatial differentiation characteristic of the soil erosion driving mechanisms and human contribution rates on the Loess Plateau showed the same boundary line from northeast to southwest, which was well explained by the 400-mm isohyetal line and Hu’s Line. This boundary can guide the geographical layout of the ecological management projects and urban development spaces on the Loess Plateau.

Unmanned aerial vehicle surveys were conducted in the summer season of 2016 and the winter season of 2017 to investigate the large wild herbivore population, including kiangs, Tibetan gazelles and bharals, in Madoi County; the source region of the Yellow River. The study generated forage grass production data in 30 m spatial resolution in Madoi County in 2016 using a downscaling algorithm; estimated a forage-livestock balance including wild animals and domestic animals; and analyzed the effect of the large wild herbivore population on the balance between forage grass and herbivory in Madoi County. The large wild herbivore population was estimated based on the density of the animals in the survey sample strip and compared and verified with available statistical data and the two survey results from the summer season of 2016 and winter season of 2017. The results showed that: (1) in the winter season of 2017, the populations of kiang, Tibetan gazelle and bharal were 17,100, 16,000 and 9300, respectively, while the populations of domestic yak, Tibetan sheep and horse were 70,800, 102,200 and 1200, respectively. The total population of large wild herbivores and domestic animals was 475,000 (sheep units). The ratio (in sheep units) between large wild herbivores and domestic animals was 1:4.5; (2) When only considering domestic animals, the grazing pressure index was 1.13, indicating slight overloading of the grassland. When considering domestic animals and large wild herbivores (kiang, Tibetan gazelle and bharal), the grazing pressure index was 1.38, suggesting moderate overloading of the grassland; (3) If large wild herbivores are not taken into consideration when the forage-livestock balance is calculated, the grazing pressure will be under-estimated by 22%. Overgrazing is the major cause of grassland degradation in Madoi County. An additional 79,000 tons of hay or a 30% reduction in domestic animals is required to maintain a forage-livestock balance in Madoi County.