Бидний тухай
Багш ажилтан
There are a small number of vegetation maps of the greater Gobi region, but for Dzungarian Gobi only coarse maps are available, and these are not based on a state-of-the-art remote sensing data and sound ground-truthing. Detailed data on vegetation–environment relationships are lacking, since climate, soils and vegetation have never been comprehensively analysed. The present study is based on data collected over three different sampling periods in the Mongolian parts of the Dzungarian Basin: we sampled 208, 152 and 280 plots in 2003, 2010 and 2012-2014, using a modified Braun-Blanquet approach. All vascular plants species were recorded along with their cover, which was estimated in absolute percentages. Sample sites were deliberately chosen to represent relevant vegetation types, ranging from those of the high mountains to the oases in the lowland. A vegetation map was then prepared based on the supervised classification of remote sensing imagery. For the vegetation map, we used remote sensing imagery (Landsat mission 8, merged together from path-row WRS2 combinations 139-028, 029; 140-027, 028, 029; 141-027, 028, 029), with data being derived from remote sensing Landsat scenes using Tasseled Cap calculations for greenness, brightness, and wetness. Cloud cover for all scenes was below 10 percent. Data from the shuttle topographic radar mission was used for altitude, from which the slope of the terrain of the sample was also calculated using the spatial calculation tool “aspect” in ArcGIS. The original Landsat scenes (red, green, and blue; short-wave-infrared-1/swir1; short-wave-infrared-2/swir2; near-infrared/NIR; coastal-aerosol) were used to feed the Random Forest model. The R package raster was used to load the raster stack into the R software with the command brick. The R package RGDAL was used to import the relevé data shapefile in ArcMap GIS software. The R package geo was used to extract the information from the extract function’s locational overlaying with the raster stack for each relevé using the Landsat/Tasseled Cap/Altitude data and described derivates. We mapped four main vegetation types (forest-steppe, steppe, desert, and oasis) encompassing twenty-one plant communities, with each type showing a clear altitudinal distribution, except for the oases. The vegetation map of the Mongolian part of the Dzungarian Gobi clearly demonstrates the strong influence of relief in the area. Altitudinal gradients are reflected in a sequence of major plant community groups. The Altai ranges govern the northern part of the study region, where their steep slopes host various types of mountain steppes and related communities, including moist high-altitude pastures in the extensive upper valleys. Due to the steep altitudinal gradient (1050-2700 m), we could distinguish several types of mountain-, meadow- and desert steppes. Forest steppes are scattered between 2100-2500 m, where they cover relatively small areas on northern exposures. North-facing slopes of the upper mountain ranges are often covered by coniferous forest while more south-facing slopes, and all exposures on the lower mountain ranges, are covered by steppes locally intermingled with Juniper stands. Steppes and related units cover roughly 56% of the total study area, and the various desert communities cover about 40%. In conclusion, combining remote sensing data and ground-truthing to map the vegetation helps to accurately identify and classify different plant communities and their distribution patterns.
The Dzungarian Basin between the Altai and the Tien Shan represents the transition between Middle and Central Asia. The region hosts special plant communities that are characterized by a combination of Kazakh-Dzungarian (Dzungario-Turanian) flora elements, which extend from Middle Asia, and typical Central Asian (semi-) desert species. For the present study, we aimed to understand the plant species composition and richness in these regions by calculating the Shannon-Wiener diversity index. We sampled 644 plots using a modified Braun-Blanquet approach. Most vegetation types were rather homogenous, and we therefore selected reléve sample of 10 x 10 m2 in size. All vascular plants were recorded along with their cover, which was estimated in absolute percentage. Sample sites were deliberately chosen to represent all relevant vegetation types, ranging from the desert to the higher mountains, and including azonal vegetation like oases. Based on our results we found four different main vegetation types (forest-steppe, steppe, desert and oasis) encompassing twenty-one communities with a total of 255 vascular plant taxa belonging to 142 genera and 41 families recorded. The results indicated that Geranio pseudosibirici-Laricetum sibiricae community in the forest steppe belt at 2430 m a.sl and Halerpesto-Hordeetum brevisubulati typicum in azonal vegetation had higher indices of diversity and richness whereas, lowest in Stipo glareosae-Anabasietum brevifoliae typicum in desert steppe at 1100 m a.s.l. Changes in elevation is associated with changes in precipitation, temperature, soil type, and other factors that influence in floristic diversity in plant communities. In conclusion, plant species diversity strongly correlated with altitude in the Dzungarian Gobi and South-Western Mongolian Altai region. To further understand plant species diversity in this region, it is necessary to conduct more comprehensive studies that incorporate various environmental variables. These variables may include precipitation, temperature, soil characteristics, and grazing intensity.
Зүүнгарын говийн ургамалжлыг ойт хээр, цөлөрхөг хээр, цөл, бүсийн бус гэсэн 4 үндсэн хэвшинжид хамаарах 21 бүлэг эвшилд ангилав.
The Dzungarian Gobi (DzG), one of 16 phytogeographical regions in the country, is located in the southwestern part of Khovd province in western Mongolia. It comprises some of Mongolia’s largest reserves, namely the Great Gobi B Strictly Protected Area and the National Park Bulgan gol-Ikh Ongog. We conducted a comprehensive survey of the area’s floristic diversity between 2009 and 2019 by collecting vascular plants from different vegetation types in various seasons. In addition, we critically checked relevant published literature and material from the herbaria ALTB, GLM, GWF, HAL, KHU, LE, MW, NS, OSBU, UBA, and UBU to determine the occurrence of vascular plant species in the DzG region. Based on our collection data, a comprehensive checklist of DzG’s flora was compiled, representing 913 vascular plant taxa (including 34 subspecies and one variety) belonging to 329 genera and 70 families. Twenty-one taxa were newly found in the DzG region. We also investigated the conservation status of all species noted, and 19 endemic plants and 96 threatened species, including six critically endangered, 26 endangered, 57 vulnerable, and seven near threatened plants were recognized in this region. Eight rare species were newly assessed according to regional conservation status based on GeoCat and IUCN. The richest plant families found were Asteraceae (153 species), Fabaceae (77 species), Amaranthaceae (69 species), and Poaceae (68 species). Several uncertain endemic and non-endemic plants remain still discussion, such as Papaver baitagense and Rosa baitagensis; thus, further studies are needed on their taxonomic and conservation status. For each taxon, we provide its distribution in the region, elevation range, voucher number, and additional references. Finally, we analyzed species hotspots of DzG, based on three different plant species richness criteria: i. all recorded species, ii. endemic species, and iii. threatened species using our georeferenced records. The most diverse hotspot area in DzG is the Baitag Bogd Mountain area, which comprises the highest species number of all three richness criteria.
Central Asia hosts grassland and desert regions that are globally important to nature conservation and local livelihoods. Several major vegetation surveys have been published on the region, with the majority focussing on areas within central Mongolia and north-eastern China. Much less information is available on plant community composition on the Dzungarian Basin, which forms a transitional zone between Central and Middle Asia in terms of flora. The Altai Mountains extend along the northern and eastern boundaries of the basin and introduce further environmental heterogeneity. For the present study, we assess horizontal and vertical gradients in plant community composition in both the Chinese and Mongolian parts of the Dzungarian Basin and the adjacent Altai Mountains. We show that steep environmental gradients trigger notable differences in plant community composition over relatively short distances. The study is based on data collected over three different sampling periods in the Mongolian part: In 2003, we sampled 208 plots; in 2010, 152 plots; and during the vegetation periods of 2012–2014, 280 plots. On the Chinese side, 58 samples were taken in 2013. Sample sites were deliberately chosen to represent relevant vegetation types, ranging from those of the high mountains to the lower oases. We collected biomass and mixed topsoil samples for soil analysis and compiled additional environmental data. A vegetation map was then prepared based on the supervised classification of remote sensing imagery. Our results reveal that four different main vegetation types (forest-steppe, steppe, desert and oasis) encompass twenty-one plant communities, with each type showing a clear altitudinal distribution, except for that at oases. Detrended Correspondence Analysis revealed the expected close correlation between altitude and species composition and productivity, but it also highlighted significant differences between the vegetation of neighbouring mountain ranges in the Chinese and Mongolian Altai.
In this study, we investigate the performance of machine learning classification approaches and different remotely sensed data sources for identifying and mapping three types of grassland communities found in the Mongolian Steppe region (Artemisia, Caragana and grass dominated steppes). The Mongolian steppe is intensively used as pasture and provides the economic basis for approximately 1 million herders. The grassland types differ in their forage values, which is why a spatially-explicit estimation of their occurrence is of high importance. We compared different sensors, classifiers, and training-sample strategies to identify the most effective approaches for mapping these communities. Ten datasets were used: Landsat 8 OLI (30 m), pan-sharpened Landsat 8 (15 m), Landsat 8 Surface Reflectance (30 m), Sentinel 2 (10 m), Sentinel 2 (20 m), Worldview 3 (0.5 m and 1.2 m), integrated Landsat 8 and Sentinel 2 (30 m), temporal Landsat 8, and temporal Sentinel 2. The two foremost classifiers at producing high accuracy of land cover classification, SVM and RF, were applied with the same training datasets. The training samples were collected in a manner so that they could be used for different spatial resolutions (i.e., ranging from 0.5 to 30 m) with the least effect from mixed training samples and spatial autocorrelation. The results of this study indicate that remote sensing is a viable method for the identification of different grassland communities in the Mongolian Steppe region.
