2.1 Study site and sampling
The study was conducted at Haibei Alpine Meadow Ecosystem Research Station, belonging to the Northwest Institute of Plateau Biology of the Chinese Academy of Sciences. The site is situated within a large valley surrounded by the Qilian Mountains, in the northeast of the Tibetan Plateau (37°29’-37°45’ N, 101°12’-101°23’ E, altitude of 3200-3500 m a.s.l.). The area has a typical plateau continental climate, with mean annual temperatures and precipitation of –1.7 °C and 582 mm, respectively. More than 80% of precipitation falls from May to September during the plant growing season. The site has Mat Cry-gelic Cambisols and silty clay loam soils, based on Inceptisols of the U.S. soil taxonomic system.
The dominant population of alpine grassland gradually changes from Gramineae-Kobresia humiliscommunity to Kobresia humilis and Kobresia pygmaeacommunities under long-term overgrazing. Furthermore, long-term overgrazing increases root accumulation and decreases surface soil bulk density and the available soil nutrients, resulting in soil nutrient imbalance between demand and supply. Moreover, the thickening of the mattic epipedon reduces the water infiltration capacity. It forms dark spots and local death in Kobresia pygmaea community, which eventually evolve into ”Black-soil-type” degraded land. Therefore, the experimental sites with different grazing intensities were defined as Gramineae-Kobresia humilis community (GK), Kobresia humilis community (K), thickening-in-mattic epipedon of Kobresia pygmaea community (KT), cracks-in-mattic epipedon of Kobresia pygmaea community (KC), and Forb-black soil type grassland (FB) (Table 1). The sites were named after the succession of the plant community after a long period of grazing lambs and sheep. The grazing intensities of the GK, K, KT, KC sites were 3.65, 7.50, 11.25, and 13.00 sheep ha-1, respectively. FB caused substantial degradation and could not be used for additional grazing. The sites were used as winter pastures with a grazing period from 20th September to 20th May. Each site was approximately 30-40 ha, and was fenced. The alpine meadow has experienced different grazing intensities since 1995. The aboveground plant tissues were not removed or consumed by grazing in the summer since the meadow was used as winter pasture.
The nutrient addition experiment was conducted in the Gramineae-Kobresia humilis community in Haibei station to determine the effect of nutrients on the grassland. The above experiment was a completely randomized block group experimental design, with four treatments: nitrogen addition (N), phosphorus addition (P), nitrogen and phosphorus mixed addition (NP), and control (CK). The treatments were randomly arranged in six plots (six repeats) (24 plots), measuring 6 m×6 m, with protection lines between all the plots. N fertilizer (100 kg/ (hm2·year)) and P fertilizer (50 kg/(hm2·year) were added in the form of urea, and heavy superphosphate, respectively. NP was added as a mixture of N and P fertilizers, and the control treatment had no nutrients. Fertilizing was conducted in the middle of the planting season, rainy or sunny evenings. Fertilizer was evenly distributed in the corresponding area by hand. Fertilizer was applied twice in each district to ensure the uniformity of fertilization. Nutrients were added from 2011 to 2015, on 1st June, 1st July, and 1st August of each year.
Five meadow sites under different grazing intensities and nutrient addition experimental sites were selected on 23rd July 2015 to determine GRSP and soil properties. Each site was divided into six plots. Soil was collected from five randomized points in each plot at depths of 0-10 cm and 10-20 cm. The soil samples of each soil depth in every plot were mixed into one sample, yielding 32 (five succession stages) and 48 (nutrient addition) samples, then kept on ice before being stored at 4 ºC in the laboratory as quickly as possible. Fresh soil was passed through a 2-mm mesh sieve after removing visible litter, roots, and stones. The soil sample was thoroughly mixed and stored at 4 °C for the determination of the biogeochemical properties.