| Alpine grassland on Tibetan plateau is one of important pastoral areas in China and is the general basis for the survival of Tibetan people. The stability of alpine grassland ecosystem has enormous impact on the climate of Eastern and Southwestern China, but also has a significant impact on the climate of northern hemisphere, even the world's climate. However, the increasingly serious problem of grassland degradation has posed a serious threat on the Tibetan Plateau even surrounding area's ecological environment and socio-economic sustainable development.On Tibet Plateau, yak dung was collected from the alpine grassland to fuel and was herders'reliable fuel fueled for cooking, making tea and warming. Collecting dung lead to nutrients and seeds decrease in alpine grassland, but burning yak dung reduced bush logging in the shrub grassland. In fact, it has protected the shrub grassland by fueling the yak dung. The yak dung has the role of maintenance for Qinghai-Tibet Plateau alpine grassland health, which has not been reported in literature. So a series experimental tests were studied systematically which included: 1) investigating the total volume of cow dung systematically fueled by herders in Tibetan plateau pasturing area, study how much areas shrub grassland is protected from damage and destruction with fueling yak dung by investigating yak dung amount fueled and testing calorific value of yank dung and shrubs, quantizing how much grassland ecosystem nutrients were lost by collecting yak dung; 2) the influence of yak dung accumulation on alpine grassland ecosystem by artificial accumulated yak dung on alpine grassland; 3) observing the defecation characteristics of yak, studying the digestibility characteristics of yak and sheep, quantizing how much cow dung needed one person per year as fuel; 4) the mechanism and process of endozoochorous seed dispersal by yak and sheep in Tianzhu alpine grassland on alpine grassland seeds, and whether the seeds defecated by yak and sheep are easily germinating; 5) the influence of grazing system which grazing yaks and sheep only and mixed on grassland soil, species biodiversity and forage product. The main results obtained are as followed.Moisture content of the yak dung applied for accumulation experiment was 75%. Total N, P and K was 1.33%, 0.159% and 0.626%, total organic matter was 51.87%, pH was 6.84. Dung was still present in the field after one year'accumulation. Yak dung (treatment) decreased soil pH, increased soil EC, improved soil physical properties of alpine meadow, reduced soil bulk density and increased soil moisture significantly. The accumulation of the yak dung increased the available N, total N, available P, total P, available K and organic matter content of soil under yak dung and around dung 10 cm.Yak dung accumulation caused above-ground plants death under dung for poor oxygen content and less light in the first year. After yak dung accumulating one year and five months, total forage product around dung 10 cm was significantly higher than the control. For yak dung patch overlapped plants and caused the death of forage, its accumulation caused forage product decline. It is a dilemma that moving down yak dung from alpine grassland made soil nutrient and soil seed losses while accumulation made forage production decreased.The calorific value of yak dung ranged from 15.38 to 17.89 KJ / g, and the average is 16.47 KJ / g. Among 21 species shrubs tested, the calorific value ranged from 17.80 to 24.26 KJ / g, and the average is 21.22 KJ / g. The average volume of yak dung fueled annual each herdsman in Zhuaxixiulong Township of Tianzhu County in Gansu Province, Huangcheng Township of Sunan County in Gansu Province, Dawu Township of Maqin County in Qinghai Province and GuluTownship of Naqu County in Tibet was 2182 kg, 590 kg, 9233 kg and 3595 kg, respectively. The average volume of shrub yields which are reduced by fueling yak dung in the four places were 1702, 460, 7202 and 2804 kg or area equivalent 1229, 332, 5200 and 2025 m2.The monthly average of total phosphorus, total nitrogen, total potassium and total organic carbon content in yak dung is 0.4974%, 2.031%, 0.665% and 39.41%, while which in sheep dung is 0.5026 %, 1.954 %, 0.707 % and 41.14 %, respectively. Approximately, 11, 3, 46 and 18 kg P, 44, 12, 188 and 73 kg N, 15, 4, 61 and 24 kg K was fueled each herdsman annually in four places mentioned frontly, respectively.Integrating the defecation volume of whole year, the annual defecation amount of yak in cluster was 2.0 kg / d ? head. So, 5.2, 1.5, 12.7 and 5.0 yaks can meet herder's yak dung needs for four places mentioned frontly, respectively.At the end of April, in Zhaxixiulong Township of Tianzhu County, grazing time on the grassland is 7 hours and 30 minutes. The proportion of dung weight defecated on the grassland was 35.65 percent. According to whole group of yaks defecation, total occupation area was per yak defecated was 0.194m2, and grassland area covered by each yak's defecation per day was about 0.078 m2.Yak apparent digestibility of mixed-grass in October and February was 63.22 % and 61.81%, of highland barley straw in January was 61.81%, and of oat hay in April was 59.42%. Sheep apparent digestibility of mixed grass in October and February was 53.68% and 57.64%, of oat hay in January and April were 57.25% and 55.18%.Recoverage proportion of alpine grassland plant seeds defecated by yak and sheep was different by animal and species. Oxytropis ochrocephala, Iris lacteal, Polygonum viviparnm, Rumex crispus, Thermopsis lanceolata, Stellera chamaejasme, Pocockia ruthenia, Achnatherum inebrians, Scirpus distigmaticus, Elsholtzia densa, Chenopodium glaucum, O. falcate, Blysmus sinocompressus, Scirpus validus, Carex przewalski, C. enervis, Pedicularis verticillata, P. multicaulis, Plantago asiatica seeds and bulbil of P. viviparnm recoverage propoion fefacated by yak was greater than 20 percent; Anemone rivularis and Hedysarum multifugum seeds recoverage propoion was smaller than one percent; Elymus nutans, Aconitum szechenyianum, Sphallerocarpus gracilis, Allium sikkimense, Roegneria hirsute, Kobresia capillifolia, C. angustifructus, Bromus inermis, Festuca ruba and Leymus secalinum seeds recoverage propoion was range from 1 to 10 percent; and C. moorcroftii and Sophora moocroftiana seeds recoverage proportion was range from 10 to 20 percent.O. ochrocephala, T. lanceolate and O. falcata seeds recoverage propoion fefacated by sheep was greater than 20 percent, E. nutans, A. szechenyianum, S. gracilis, R. hirsute, H. multifugum, A. rivularis, B. inermis and F. ruba seeds recoverage propoion was smaller than one percent, and C. moorcroftii, I. lacteal, R. crispus, S. chamaejasme, A. sikkimense, K. capillifolia, A. inebrians, S. moocroftiana and P. viviparnm seeds and Bulbil of P. viviparnm seeds recoverage propoion was range from 1 to 10 percent, C. enervis, P. ruthenia, P. verticillata, S. distigmaticus, E. densa, C. glaucum, B. sinocompressus, S. validus, C. przewalski and P. asiatica seeds recoverage proportion was range from 10 to 20 percent.Seed of C. enervis, C. przewalski, O. falcate, C. glaucum and S. distigmaticus defecated by yak and sheep germination was promoted. P. viviparnm, K. capillifolia and B. sinocompressus seeds germination was promoted passed yak's digestive tract but that was not passed sheep's'. Grasses, O. ochrocephala, R. crispus, A. szechenyianum, A. sikkimense, P. verticillata, T. lanceolata, E. densa and P. asiatica seeds germination was reduced, and seeds of C. moorcroftii, I. lacteal, S. chamaejasme, S. moocroftiana, P. ruthenia and S. Validus seeds, and P. viviparnm bulbil germination was unchanged.There are 43 species identified in yak dung from Nov. to April, and there are 45 grain seeds per piece of yak dung.In three grassland plots grazed with similar intensity including grazing yak moderately, and grazing sheep moderately, whether in 2008 or 2009, the forage yield of plot only grazing sheep was higher than the remaining two, and which of grazing yak and sheep was higher than the plot grazing yak he yak The average forage yield of 2008 and 2009 on grazing sheep only, yak and sheep, and only yak is 581, 476 and 411 g/m2, respectively. Similar results were appeared in plant height and total shoot density.The Shannon-Wiener diversity index of the grassland grazing sheep moderately is 4.5825, lower than that of grazing yak and sheep moderately and that of grazing yak moderately, of grazing yak. Similar result appeared in Pielou evenness index .The biggest Shannon-Wiener diversity index appeared on moderate grazing plot and smallest appeared on heavy grazing plot in mixed grazing with yak and sheep plots. Species numbers of the grassland grazing yak moderately is more than mixed grazing moderately plot with yak and sheep and grazing sheep moderately. Yak and sheep mixed rangeland, the species number increased firstly, and then decreased as grazing intensity increasing, that is, under moderate grazing the species number was highest.Soil bulk density in plot with grazing sheep moderately was smaller than grazing mixed with yak and sheep, the biggest was the grassland grazing yak moderately. Opposite result appeared in soil available potassium, available nitrogen, available phosphorus, total nitrogen and total phosphorus content. Soil organic matter content in grazing sheep moderately is significantly higher than that of grazing mixed with yak and sheep and grazing yak moderately, while which of grazing yak moderately is higher than that grazing mixed with yak and sheep.Soil neutral phosphatase, catalase and urease enzyme activity in plot with grazing sheep moderately was higher than grazing mixed with yak and sheep, the smallest was the grassland grazing yak moderately except soil urease enzyme activity form 0 to 10 cm.On Tibet Plateau alpine grassland, burning yak dung reduced bush logging, so it has protected the shrub grassland by fueling the yak dung. For accumulation of yak dung defecated in cold season on alpine grassland led to forage product decline at least two years while removed yak dung from alpine grassland made the harm caused by yak dung accumulation eliminated. As there are lots of N, P, K and other components and some germinable seeds in yak dung defecated in cold season, so collecting dung lead to nutrients and seeds losses in alpine grassland, and the losses increased with the growing grazing intensity. Which weakened the maintenance of yak dung for Qinghai-Tibet Plateau alpine grassland health. Break down yak dung on alpine grassland or return to grassland after composting, which can resolve the dilemma that dung accumulation and removing, and maintenance of yak dung on the alpine grassland health is maintained. Most of yak dung was removed from grassland while sheep dung was keeping, which was one of reasons made grazing sheep was more benefit alpine grassland than grazing yak on Tibet Plateau. So, for above reason,"more sheep and fewer yaks"grazing livestock management mode on alpine grassland was proposed in this paper. |