Dataset Development of the Hotan Oasis, Water System, Watershed, and Elevation

Lin, J. W.^1,2,3  Gui, D. W.^1,2,3*  Zhang, S. Y.^1,3  Liu, C.⁴

1. Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi 830011, China;

2. University of Chinese Academy of Sciences, Beijing 100049, China;

3. Cele National Station of Observation and Research for Desert?CGrassland Ecosystems, Cele 848300, China;

4. Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China

Abstract: The oasis, as a nonzonal geographical unit in arid regions, sustains the production and livelihoods of residents, playing an irreplaceable role in regional socioeconomic development. In this study, remote sensing imagery from Google Earth Pro was utilized as the main data source. Through visual interpretation, we generated data for the Hotan Oasis and water systems in 2015. Simultaneously, we employed digital elevation models (DEMs) and ArcGIS hydrological analysis tools to obtain both extent and elevation classification data for the mountainous regions in the Hotan River watershed. The research findings revealed that the total area of the Hotan Oasis is 4,820.94 km², with the mountainous region of the watershed covering 40,812.80 km². With the use of the hexagonal grid method for verifying the data accuracy, an accuracy of 96.58% was achieved. The dataset comprises the following components: (1) boundary data of the oasis; (2) boundary data of the mountainous part of the catchment; (3) data of the river system in the catchment; and (4) elevation classification data of the mountainous part of the catchment. This dataset encompasses 49 files stored in .kmz, .shp, and .tif formats, with a total data size of 91.40 MB (compressed into 5 files, totaling 45.80 MB).

Keywords: oasis; Hotan; water systems; watershed; elevation

DOI: https://doi.org/10.3974/geodp.2023.03.10

CSTR: https://cstr.escience.org.cn/CSTR:20146.14.2023.03.10

Dataset Availability Statement:

The dataset supporting this paper was published and is accessible through the Digital Journal of Global Change Data Repository at: https://doi.org/10.3974/geodb.2020.09.13.V1 or https://cstr.escience.org.cn/CSTR:20146.11.2020.09.13.V1.
https://doi.org/10.3974/geodb.2022.07.02.V1 or https://cstr.escience.org.cn/CSTR:20146.11.2022.07.02.V1.

1 Introduction

Arid regions constitute a vital aspect of the earth??s geographical composition, accounting for approximately 45% of the global land area^[1]. These areas experience sparse precipitation and high evaporation rates, rendering the ecological environment highly fragile and excep­tionally sensitive to climate change and human activities^[2]. Oases represent a distinctive landscape where natural and human elements converge in arid regions. In these landscapes, deserts form the matrix, and oases serve as embedded jewels. China hosts the largest oasis area globally^[3], characterized by an extensive coverage and diverse types. Oases, which are characterized by abundant water sources, fertile soil, and lush vegetation, sharply differ from the arid and barren surroundings and play a crucial role in the arid regions of Northwest China. Oases not only serve as significant production bases for crops such as arid grains and cotton but also serve as processing centers for natural resources^[4]. Simultaneously, oases, which function as focal points for human settlement and activities, are preferred locations for urban development. They accommodate more than 95% of the population in arid regions, contributing more than 90% to the output for the creation of social wealth^[5].

Therefore, oases have become a focal point of research and attention in arid regions. However, due to the late development of the oasis discipline and relatively unique chara­cteristics, the oasis concept has not been fully elucidated, and different researchers have proposed different definitions of the oasis for different research purposes and needs^[6?C8]; as a result, the precise distribution of oases has been plagued by unclear basic records and large discrepancies in data, which has impeded the development of the oasis discipline. Through a comprehensive comparison of various viewpoints and combining characteristics observed in remote sensing images and on-site survey data, in this paper, oases are ultimately defined as exhibiting three essential features: 1) existing in arid and semiarid regions; 2) situated in a desert matrix; and 3) providing a stable water supply, forming a heterogeneous landscape unit with a certain vegetation cover or economic productivity.

The precise definition of an oasis is crucial for determining its spatial extent and disting­uishing it from the background desert area. This approach serves as the basis for extracting oasis boundaries. After clarifying the definition of an oasis, the Hotan Oasis is used as an example in this paper, and high-precision remote sensing imagery, terrain data, and other multisource data are employed to accurately locate and delineate the spatial extent of the Hotan Oasis and Hotan River watershed, leading to the development of the Hotan Oasis, river system, watershed, and elevation classification dataset. This dataset provides data support for the development and planning of the Hotan region and lays the foundation for and establishes a clear direction for the creation and publication of oasis datasets for other regions of China in the future.

2 Metadata of the Dataset

The metadata of the Hotan Oasis, river system, watershed, and elevation classification dataset^[9,10] are summarized in Table 1.

3 Methods

3.1 Study Area

The Hotan region is one of the five major regions in the Xinjiang Uygur autonomous region. This region boasts a rich history and is renowned for producing high-quality Hotan jade. It holds an important position in Xinjiang??s economy. The Hotan region is a multiethnic settlement area where the cultures of various ethnic groups, including the Uighur and Han

Table 1  Metadata summary of the Hotan Oasis, river system, watershed, and elevation classification dataset

people, are blended. Positioned as a crucial hub along the ancient Silk Road, Hotan plays a pivotal role in the historical and cultural heritage of Xinjiang. The region experiences extremely arid conditions, with hot and rain-scarce summers and dry, cold winters^[12]. Also referred to as the Ho-Mo-Luo Oasis, the Hotan Oasis is the largest oasis on the northern slope of the Kunlun Mountains. As shown in Figure 1, the Hotan Oasis is situated at the southern edge of the Tarim Basin and is surrounded by the Tianshan Mountains and Pamir Plateau, facing the vast Taklamakan Desert. Due to prolonged human activity and development, much of the area in the Hotan Oasis has transitioned from a natural oasis to an artificial oasis. The Hotan Oasis serves as the primary support for the social and economic activities of the residents in the Hotan region, serving as a foundational base for both agricultural production and industrial development.

3.2 Data Sources and Methods

The data for this study primarily included Hotan Oasis data, Hotan water system data, range data for the mountainous areas in the Hotan River watershed, and elevation classification data. The oasis and Hotan water system data are sourced from Google Earth Pro remote

Figure 1  Geo-location of the Hotan area and remote sensing image of the Hotan Oasis

sensing imagery. This dataset is compiled from various commercial image providers, such as DigitalGlobe, EarthSat, SPOT, and government agencies, to construct multiple sets of satellite map data. These datasets exhibit different zoom levels, extensive geographical coverage, and high spatial resolution from optical sensors (achieving submeter levels). This approach is advantageous for obtaining high-quality, detailed geographical information data for the Hotan region, facilitating further analysis and extraction.

While there are several automated^[13] or semiautomated^[14] methods available for feature extraction, the spatial heterogeneity of vegetation types within oases and the diversity of oasis boundary patterns often lead to a suboptimal extraction accuracy. Therefore, we emp­loyed a visual interpretation method to extract oasis and water system data, ensuring data high accuracy and high quality. First, we selected remote sensing images of the summer of 2015 for oasis boundary extraction. In summer, when the vegetation in oases is mostly lush and there is no snow cover, the contrast between the oasis and desert is highest, facilitating the use of image features to distinguish different land cover types. Second, we zoomed the images to the maximum level on Google Earth, with the minimum requirement a view height less than 1.5 km and a spatial resolution less than 1 m. Subsequently, based on the actual ground conditions, we identified different land cover types, used mapping functions to set control points, and saved the generated oasis boundary and water system data as a .kmz file. Next, we imported the .kmz file into ArcGIS and converted it into a shp-format vector file for subsequent geographical information analysis and processing. In ArcGIS, we conducted thorough spatial topology checks, including addressing issues such as connecting floating points, eliminating redundant line segments, and correcting polygon geometric deformations, ensuring data consistency and accuracy. Finally, we converted the oasis data polyline file into a polygon file, added information such as names in both Chinese and English, area, and length to the attribute table and removed patches with an area less than 0.01 km² to better represent the oasis spatial extent and geographical features.

The range data for the mountainous areas in the Hotan River watershed and elevation classification data were developed using the ASTER GDEM 30 m product sourced from the Geospatial Data Cloud website[1]. The watershed extraction process relies on hydrological analysis tools in ArcGIS software. The specific steps are as follows: firstly, the DEM surface of the Hotan region is filled to prevent unrealistic or erroneous flow directions during analysis caused by the presence of depressions. Then, the flow direction of each pixel in the raster is computed, generating the flow accumulation amount. With the use of the raster calculator, the expression was defined as Con(DEM_Flow>1000,1) to create river network raster data. Subsequently, the generated raster river network is converted into vector format, outlets are defined, and a preliminary Hotan River watershed range is obtained. Finally, this approximate watershed range is imported into Google Earth  as a baseline. Based on the topography of the mountainous areas in the images, high-precision boundary data for the mountainous regions in the Hotan River watershed are obtained. The production of elevation classification data involves using the obtained range for the mountainous areas in the Hotan River watershed as a mask to extract DEM data. Then, by overlaying river data and correc­ting abnormal pixel values based on the natural principle of water flowing downhill, elevation classification is conducted in ArcGIS using the raster reclassification tool with intervals of 500 m for the mountainous regions in the Hotan River watershed.

4 Data Results and Validation

4.1 Data Composition

The dataset comprises a total of 49 data files, archived in .kmz, .shp, and .tif data formats, with a total data size of 91.40 MB. These files have been compressed into 5 files, totaling 45.80 MB.

4.2 Data Results

As shown in Figure 2, the specific geog­raphic coordinates of the Hotan Oasis are 36??55ʹ45ʺN?C37??55ʹ11ʺN and 78??03ʹ11ʺE?C 80??32ʹ52ʺE. In 2015, the total area of the Hotan Oasis, calculated based on the Al­bers projection, was 4,820.94 km², and the total perimeter, based on the Lambert projection, was 2,358.08 km. Among the various counties in the Hotan region, the Hotan Oasis is predominantly located in Moyu county (34.32%), followed by Hotan county (28.46%) and Luopu county (20.71%). Hotan city (8.40%), Kunyu city (5.08%), and Pishan county (3.03%) exhibit smaller areas.

The existence of water defines an oasis, while the absence of water characterizes a desert. The survival and development of the Hotan Oasis primarily depend on surface runoff from melting snow and ice in the Kunlun Mountains and a small amount of atmospheric precipitation, which forms the Hotan River. The spatial variations in the Hotan River and its tributaries are critical drivers of the formation of the Hotan Oasis, significantly influencing its spatial distribution. The Hotan River system comprises two main branches, i.e., the eastern Yulong Kashgar River (referred to as the Yuhe River), which originates from the northern foothills of the Kunlun Mountains and extends for 554.84 km, and the western Karakash River (referred to as the Kahe), which originates from the northern foothills of the Karakoram Mountains and extends for 797.21 km. The Hotan Oasis was formed by irriga­ting the alluvial plains of the Kunlun Mountains with the source of the two rivers as the centerline and secondary rivers and irrigation canals as links. As shown in Figure 3, the geographic coordinates of the mountainous part of the Hotan River basin are 34??50ʹ40ʺN?C37??11ʹ37ʺN and 77??23ʹ55ʺE?C81??40ʹ52ʺE. The area of the mountainous region in the Hotan River watershed is 40,812.80 km², with a boundary length of 1,829.46 km. The elevation of the mountainous region in the Hotan River watershed is classified into 12 categories, ranging from <1,500 m at the lowest point to >6,500 m at the highest point.

Figure 3  Extent and elevation classification of the mountainous part of the Hotan River basin

4.3 Data Validation

We validated the accuracy of the Wada Oasis dataset using a method typically used to validate the accuracy of remote sensing classification. First, a 500 m ?? 500 m hexagonal grid was constructed to cover all the Wada oasis areas. Subsequently, different numbers of validation points were randomly generated based on the proportion of the oasis coverage area in each hexagonal grid cell: 5 points were generated for areas with a proportion of 0%?C 20% of the coverage area, 10 points were generated for areas with a proportion of 20%?C40% of the coverage area, 15 points were generated for areas with a proportion of 41%?C60% of the coverage area, 20 points were generated for areas with a proportion of 61%?C80% of the coverage area, and 25 points were generated for areas with a proportion of 80%?C100% of the coverage area. Next, after extracting the latitudinal and longitudinal coordinates of each validation point and determining whether they were located in the oasis region specified in the dataset, we imported these points into Google Earth Pro to visually determine whether they were indeed oasis points. As shown in Figure 5, this process resulted in the generation of 332 grids, the evaluation of 5,549 points, and a total of 5,359 points correctly categorized as oasis points, with an accuracy rate of 96.58%.

Figure 4  Distribution map of the hexagonal grids and validation points

5 Discussion and Conclusion

In this study, we constructed a Hotan Oasis, water system, mountainous part of the watershed and elevation classification dataset based on Google Earth Pro images and DEM data. This dataset is the first high-precision dataset of an oasis in China. It finely captures the spatial distribution of the Hotan Oasis in 2015 and promotes the development of oasis research from qualitative to quantitative. In this paper, the area of the Hotan River oasis in 2015 was 4,820.94 km², and the area of the mountainous part of the Hotan River basin was 40,812.80 km². It should be noted that this dataset is based on Google Earth Pro time-specific imagery, and it may be necessary to adjust the time of the imagery when using it in order to avoid updating the imagery so that it does not correspond to the dataset.

Author Contributions

Gui, D. W. and Liu C. developed the overall design for dataset development. Lin, J. W. collected and processed the data. Zhang, S. Y. validated the data. Lin, J. W. wrote the paper.

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]      Prăvălie, R. J. E-S R. Drylands extent and environmental issues [J]. A global approach, 2016, 161: 259?C278.

[2]      Chen, F. H., Huang, W., Jin, L. Y., et al. Characteristics of precipitation changes and their spatial differences in the arid zones of Central Asia in the context of global warming [J]. Science China: Earth Sciences, 2011, 41(11): 11?C27.

[3]      Liu, X., Wang, Y., Xin, L., et al. China??s oases have expanded by nearly 40% over the past 20 years [J]. Land Degradation and Development, 2022, 33(18): 3817?C3828.

[4]      Han, D. L. Xinjiang Artificial Oasis [M]. Beijing: China Environmental Science Press, 2001.

[5]      Gui, D. W., Zeng, F. J., Lei, J. Q., et al. Suggestions for Sustainable Development of the Oases in the South Rim of Tarim Basin [J]. Journal of Desert Research, 2016, 36(1): 6?C11.

[6]      Jia, B. Q. Exploration of some theoretical issues in oasis landscapes [J]. Arid Land Geography, 1996, 19(3): 8?C15.

[7]      Shen, Y. C., Wang, J. W., Wu, G. H., et al. China Oasis [M]. Zhengzhou: Henan University Press, 2001.

[8]      Chen, X., Luo, G. P. Research on oasis ecology in arid zones and its progress [J]. Arid Land Geography, 2008, 26(4): 7?C15.

[9]      Gui, D. W., Xue, J., Liu, Y., et al. Hotan Oasis [J/DB/OL]. Digital Journal of Global Change Data Repository, 2020. https://doi.org/10.3974/geodb.2020.09.13.V1. https://cstr.escience.org.cn/CSTR:20146. 11.2020.09.13.V1.

[10]   Gui, D. W., Li, B. F., Zhang, S. Y., et al. Water System, watershed and elevation classification dataset in the upper reaches of Hotan River, China [J/DB/OL]. Digital Journal of Global Change Data Repository, 2022. https://doi.org/10.3974/geodb.2022.07.02.V1. https://cstr.escience.org.cn/CSTR:20146.11.2022. 07.02.V1.

[11]   GCdataPR Editorial Office. GCdataPR data sharing policy [OL]. https://doi.org/10.3974/dp.policy.2014.05 (Updated 2017).

[12]   Dong, D. W. Analysis of spatial and temporal evolution of Hotan oasis in the context of urbanization [D]: Urumqi: Xinjiang University, 2019.

[13]   Yao, J., Wu, J., Xiao, C., et al. The classification method study of crops remote sensing with deep learning, machine learning, and Google Earth engine [J]. Remote Sensing, 2022, 14(12): 2758.

[14]   Cheng, Y., Wang, W., Ren, Z., et al. Multi-scale Feature Fusion and Transformer Network for urban green space segmentation from high-resolution remote sensing images [J]. International Journal of Applied Earth Observation and Geoinformation, 2023, 124: 103514.