Datasets for research on groundwater flow and its interactions with surface water in an alpine catchment on the northeastern Tibetan Plateau, China

Climate warming has significantly changed the hydrological cycle in cold regions, especially in areas with distributed permafrost. Groundwater flow and its interactions with surface water are important components of the hydrological process; however, few studies or modeling works have been based on long-term groundwater level, temperature, hydrogeochemistry or isotopic field observations 15 from boreholes due to obstacles such as remote locations, limited infrastructure, and harsh work conditions. In the Hulugou catchment located in the headwater region of the Heihe River on the northeastern Tibetan Plateau (TP), we drilled four sets of depth-specific wells and monitored the dynamic groundwater levels and temperatures at different depths. Surface water (including river water, glacier meltwater, and snow meltwater), precipitation, groundwater from boreholes, spring water, and soil water 20 samples were collected to measure the hydrochemistry, dissolved organic carbon (DOC), dissolved inorganic carbon (DIC), and stable and radioactive isotopes at 64 sites. This study provides datasets of these groundwater parameters spanning six consecutive years of monitoring/measurements. These data can be used to investigate the groundwater flow process and the interactions between groundwater and https://doi.org/10.5194/essd-2021-273 O pe n A cc es s Earth System Science Data D icu ssio n s Preprint. Discussion started: 10 September 2021 c © Author(s) 2021. CC BY 4.0 License.

surface water on the TP under global climate change. The datasets provided in this paper can be obtained 25 at https://doi.org/10.5281/zenodo.5184470 (Ma et al., 2021b) and will be subject to further updates.

Introduction
As the "Third Pole" of the Earth, the Tibetan Plateau (TP) is known as the "Water Tower of Asia" and comprises the headwater regions of many large rivers in Asia, including the Linsang, Ganges, Indus, Yellow and Yangtze Rivers (Qiu, 2008;Immerzeel et al., 2010). At present, the total amount of water 30 storage on the TP (including glacier reserves, lake water, and runoff from the outlets of major rivers) has been initially estimated to exceed nine trillion cubic meters (Li et al., 2019;Pritchard, 2019;Liu et al., 2020). Under the background of global warming, not only has the temperature of the TP gradually increased over the past 50 years (Hartmann et al., 2013;Yao et al., 2013), but the warming rate has also been slowly increasing (Chen et al., 2015;Kuang and Jiao, 2016). Therefore, the TP has been experiencing 35 a series of remarkable changes, such as permafrost degradation, continuously decreasing snow cover, and glacier and lake shrinkage (Jin et al., 2011;Yao et al., 2012;Liu et al., 2015;Huang et al., 2017;Xu et al., 2017;Bibi et al., 2018;Ran et al., 2018;Yao et al., 2019b;Bian et al., 2020). These changes have vastly impacted the hydrological system on the TP, affected the living environments of 1.7 billion people and caused economic losses of up to $12.7 trillion (Yao et al., 2019a). 40 The flow of groundwater and its interaction with surface water play important roles in regional ecological-environmental and biogeochemical cycles, especially in cold regions (Frey and Mcclelland, 2009;Prowse and Brown, 2010;Wang et al., 2010;Amanambu et al., 2020). Previous studies have shown that groundwater, the main recharge source of river water in permafrost regions, is an important component in maintaining the base flow of rivers Carey et al., 2013; resources (Green et al., 2011;O'donnell et al., 2012;Sharma et al., 2012;Cochand et al., 2019). In the case of global warming, interactions between groundwater flow and surface water may cause the release 50 of carbon trapped in permafrost and aggravate the greenhouse effect (Harlan, 1973;Solomon et al., 2007;Schaefer et al., 2011;Wisser et al., 2011;Mckenzie and Voss, 2013;Connolly et al., 2020;Behnke et al., 2021). Because of the heat released by the movement of groundwater in the subsurface layer and the thermal insulation effect of the surface ice layer, saturated sediments are not frozen year round and provide interstitial habitats for groundwater fauna (Schohl and Ettema, 1990;Clark and Lauriol, 1997;Alekseyev, 55 2015; Huryn et al., 2020;Terry et al., 2020). All these findings undoubtedly confirm the importance of groundwater flow in cold-region hydrological systems.
However, existing studies on the interactions between groundwater and surface water have mainly focused on Arctic-subarctic regions. These areas are mainly characterized by continuous permafrost, and the main groundwater recharge source in these regions is in the form of spring snowmelt (Mcclymont et 60 al., 2010). In contrast to Arctic-subarctic regions, the middle-and low-latitude catchments on the TP feature distributions of continuous permafrost, discontinuous permafrost, island permafrost and seasonally frozen ground due to the great topographical and elevational differences that exist throughout the region (Cheng and Jin, 2013;Chang et al., 2018). The Asian monsoon climate on the TP causes the main hydrological inputs to occur in summer and autumn, when the amounts of precipitation and glacier 65 meltwater are largest (Lu et al., 2004;Chang et al., 2018). Different combinations of permafrost, seasonally frozen ground and hydrological conditions lead to complex interactions between groundwater and surface water on the TP (Woo, 2012). Therefore, it is necessary to deepen the research on groundwater flow processes on the TP.
Current studies on the interactions between groundwater and surface water on the TP have focused 70 on tracing flow paths with different types of isotopic and geochemical data or building numerical waterheat coupling models to explore the influence of climate change on hydrological processes (Ma et al., 2009;Ge et al., 2011;Evans et al., 2015;Hu et al., 2019;Li et al., 2020a;Yang and Wang, 2020;Tan et https://doi.org/10.5194/essd-2021-273  al., 2021). Although great progress has been made in studying hydrological processes on the TP, including research on changing streamflow compositions under the impact of climate change and frozen soil 75 degradation Cuo et al., 2019;Xu et al., 2019;Shen et al., 2020), little is known about the groundwater system or the processes that control groundwater and surface water interactions due to challenges associated with the poor field conditions and weak infrastructure on the TP (Yao et al., 2019a).
Most existing studies have used spring water and baseflow measurements in winter to represent groundwater and have rarely directly observed groundwater indicators through boreholes (Pu et al., 2017;80 Gui et al., 2019;Li et al., 2020b;Pu et al., 2021). The coupled groundwater flow and heat transport models employed in previous studies to represent permafrost regions on the TP have mainly focused on scenarios in which the conceptual models are tested and lack verification with actual field data (Ge et al., 2011;Evans et al., 2015). To date, there is no established systematic research site considering groundwater flow or its interactions with surface water on the TP to monitor physical or chemical surface water and 85 groundwater indicators and the freeze-thaw processes of permafrost and seasonally frozen ground. To fill this gap, we established a systematic monitoring site in the upper reaches of the Heihe River on the northeastern TP.
Specifically, this paper introduces a monitoring system for the groundwater level, ground temperature, water chemistry, dissolved organic carbon (DOC) and dissolved inorganic carbon (DIC) 90 concentrations, and isotopic compositions of various water bodies in an alpine catchment. In Section 2, the general setting of the study area is introduced in detail. In Section 3, the layout of the monitoring wells, the lithology of the boreholes and the mineralogical compositions of the cores are described. In addition, groundwater level and ground temperature changes are also explained. In Section 4, the methods used for the collection, preservation, and analysis of data representing various water bodies are described. The 95 methods for obtaining raw data mentioned in this article are provided in Section 5, and future research prospects and a summary of the whole article are provided in Section 6.

Study area
The study area is located in the upper reaches of the Heihe River in the northeastern region of the 100 Tibetan Plateau (Fig. 1). The catchment occupies an area of 23.1 km 2 (99°50′37″-99°53′54″E and 38°12′14″-38°16′23″N). The elevation of the study area ranges from 2960 to 4800 m a.s.l. The study area has a continental climate, with warm, humid summers and cold, dry winters. The annual average temperature is -3.1 °C, and the annual precipitation is 403.4 mm; precipitation is mainly concentrated from July to September, with these months accounting for approximately 70% of the annual precipitation 105 (Chen et al., 2014a;Chen et al., 2014b). The water surface evaporation was 1231 mm in 2013 (Yang et al., 2013). The discharge of the catchment was approximately 3.58×10 4 m 3 /day in 2012 (Chen et al., 2014a;Chen et al., 2014b). The latest specific precipitation data, temperature data and discharge data can be downloaded from the following website: http://hhsy.casnw.net.
Two main tributaries originate from the southern glaciers in the study area (Fig. 1). These tributaries 110 are mainly fed by glacier and snow meltwater, ice lakes and springs in the high mountains (Yang et al., 2013;Chang et al., 2018;Hu et al., 2019). In the warm season of each year (May-September), these tributaries are sustained; during the rest of the year, they are intermittently dry (Yang et al., 2013;Chang et al., 2018;Hu et al., 2019). The two tributaries converge into a single channel at the northern end of a piedmont alluvial plain and finally flow into the main channel of the Heihe River. Due to the thick alluvialpluvial sediments deposited in this plain, groundwater and surface water interact significantly (Ma et al., 2017;Chang et al., 2018;Hu et al., 2019).
The region above 4200 m a.s.l. in the study area is mainly eroded by mountain glaciers. The area comprising modern glaciers and permanent snow cover is approximately 1.62 km 2 (Ma et al., 2017;Chang et al., 2018;Hu et al., 2019). Many moraine sediments are distributed on the surface in this region, 120 constituting a porous aquifer of moraine breccia with large pores and good connectivity (Chang et al., 2018). According to field investigations, the aquifer exhibits high permeability and provides a good channel for groundwater flow (Ma et al., 2017;Chang et al., 2018). In the warm season (May-September), the aquifer is recharged by meltwater from glaciers and snow and by precipitation; this water is rapidly discharged to nearby rivers and provides lateral recharge for low-elevation aquifers (Ma et al., 2017). 125 Vegetation is rare in this region, and the vegetation coverage is extremely low (Liu et al., 2012(Liu et al., , 2014. In the elevation range of 3500 to 4200 m a.s.l., permafrost is distributed discontinuously and reaches a maximum melting depth of 2 m at the end of August (Ma et al., 2017). Through analysis of drilling data, the thickness of permafrost was found to be approximately 20 m (Ma et al., 2017;Hu et al., 2019). There is a porous aquifer consisting of argillaceous gravel on the planation surface at the top of the hill in the 130 permafrost region (Ma et al., 2021a). In the warm season (May-September), the groundwater in this aquifer is mainly recharged by the infiltration of glacier and snow meltwater, precipitation, and groundwater from the higher-elevation aquifer and exists in the forms of suprapermafrost, intrapermafrost, and subpermafrost groundwater (Ma et al., 2021a). This aquifer groundwater is typically discharged to various tributaries in the form of springs near the rivers at the foot of the slope. In this region, alpine 135 meadows are the main vegetation type (Liu et al., 2012;Chen et al., 2014a;Liu et al., 2014).
Seasonally frozen soil is mainly distributed in the areas below 3500 m a.s.l. The maximum freezing depth is approximately 3 m; freezing to this depth occurs in January (Ma et al., 2017). Alluvial sediments in the piedmont plain constitute a porous aquifer containing sands and gravels (Ma et al., 2017;Chang et al., 2018). The thickness of this aquifer is 20-50 m, and it can store water in summer and maintain 140 baseflow in winter to regulate streamflows (Ma et al., 2021a). In this area, the surface water and groundwater are hydraulically well connected, and the groundwater in the aquifer is recharged by the infiltration of river water and the lateral runoff of groundwater from the adjacent mountainous areas. In addition, the groundwater in this region is discharged into lower-elevation rivers in valleys or through springs near the drainage basin outlet (Ma et al., 2017;Chang et al., 2018). The groundwater mainly flows 145 from south to north based on the terrain. There are mainly alpine grassland and herbaceous plants in the areas below 3200 m a.s.l. (Liu et al., 2012). The vegetation types in higher-elevation areas are alpine shrubs (Liu et al., 2014).

Layout of well groups
There are four well groups in the study area, among which well groups WW01, WW02 and WW03 155 are located in areas with seasonally frozen soils and well group WW04 is located in a permafrost area ( Fig. 1). The specific locations of these well groups are as follows: WW01 is located at the top of the mainstream in the northern part of the study area, near the intersection of the east and west tributaries; WW02 is located in the middle of the piedmont alluvial plain; WW03 is located at the top of the piedmont alluvial plain, near the bedrock in the southern mountainous area; and WW04 is located on the planation 160 surface of a terrace in the eastern mountainous area. The details of these well groups are listed in Table 1

Sediments and their mineral compositions in the well groups
During drilling, lithological characterization was conducted at different depths, and the results are 175 shown in Figure 2. In the seasonally frozen area of the piedmont alluvial plain, the sediments are mainly composed of mud-bearing pebble gravels that are very loose and poorly sorted (Ma et al., 2017;Chang et al., 2018). The drilling depths of the WW01 and WW02 well groups did not reach bedrock, while the drilling depth of well group WW03 reached weathered sandstone bedrock, indicating that the thickness of the aquifer located at the top of the piedmont alluvial plain is thinner than that in the middle and lower 180 locations (Fig. 2). In addition, a clay layer with a thickness of approximately 3-6 m was found in the WW01, WW02 and WW03 well groups; this layer may be continuously distributed throughout the piedmont alluvial plain (Fig. 2).
The results obtained from well group WW04, which is located on the planation surface of the permafrost region, show that the perennially frozen layer is distributed at depths between 2 and 20 m 185 underground (Fig. 2) and mainly consists of sand, gravel and ice. The active layer is ~2 m thick and is composed of clay (Fig. 2). Depths below 20-24.3 m are characterized by the unfrozen subpermafrost aquifer, which is mainly composed of sand and gravel (Fig. 2). The intrapermafrost aquifer, which is located at an underground depth of 12 m, is unfrozen within the permafrost layer and consists of a clay layer with a thickness of approximately 0.2 m, as determined in the WW04 well group (Fig. 2). The results of the mineral composition and proportion analyses of the core samples are listed in Table   2. Chlorite, illite, quartz and K-feldspar are the main minerals present in these samples, accounting for more than 80% of all minerals. Calcite and dolomite account for 0% to 10% and 2% to 25% of the total

Collection and analysis of water samples
The following section is an introduction to the collection, preservation, and analysis of various water 235 samples obtained in the study area over six years, from 2012 to 2017. Water samples were mainly collected in the following four stages: (1) frozen period (January), (2) thawing period (April and May), (3) thawed period (July and August) and (4) refreezing period (September). The locations of the sampling sites in the study area are shown in Figure 1. The methods used to obtain specific information regarding the different sampling sites and the analytical results of the samples are described in Section 5.

Sample collection and preservation
In the study area, we collected seven types of water samples, including river water, precipitation, spring water, groundwater, soil water, glacier meltwater, and snow meltwater samples. The specific sampling site locations are shown in Figure 1. During the collection of all water samples (excluding the precipitation and soil water samples), the temperature (T), pH, dissolved oxygen (DO), electrical 245 conductivity (EC), and oxidoreduction potential (ORP) of the water samples were analyzed by a portable water quality analyzer (HQ40d, Hach, USA), which was calibrated for pH daily before use. Then, the sample bottles were cleaned three times with filtered water. When collecting groundwater samples, the water was pumped for 10 minutes to drain the old water from the wells and ensure the collection of representative water samples. headspace left above the water samples when the sample bottles were sealed.

265
During the drilling process of well group WW04, soil samples were collected at different depths, and the samples were sealed and stored in boric acid in glass bottles at -10 °C. The specific borehole numbers and sample depths are listed in the corresponding dataset. Then, the soil samples were sent back to the laboratory to undergo vacuum extraction of water from the soils, and the stable isotope ratios were subsequently measured.

DOC and DIC
The

Analysis of stable isotopes
The 13

Analysis of 3 H and 14 C isotopes
The 3  were removed by filtration with glass filter paper, followed by the addition of 85% phosphoric acid to the samples. The standard samples used for 14 C analysis were prepared according to the method described by Liu et al. (Liu et al., 2019). The 14 C analysis results were expressed as percent modern carbon (pmC), and the analysis precision was 2%. The results of this analysis are listed in Table 3.  "n.d." means not determined.

Data availability
All data mentioned in this paper can be obtained at https://doi.org/10.5281/zenodo.5184470 (Ma et al., 2021b). The original data were divided into two files. The first file contains monitoring data, including 340 groundwater levels and ground temperatures. The second file includes the results of the sample analyses and the numbers and locations (longitudinal and latitudinal coordinates) of the sampling sites.

Conclusions
This study presents a systematic monitoring site scheme located in an alpine catchment in the headwater region of the Heihe River with distributed permafrost and seasonally frozen ground. Four sets 345 of clustered wells were drilled, and the groundwater levels and ground temperatures were continuously monitored for six years at 30-minute intervals. We collected samples of sediments, river water, This study is among the few attempts to obtain groundwater level, temperature, water chemistry and isotopic data representing groundwater exiting the ground from springs and directly sampled from drilled 355 boreholes; thus, these data in combination with other measurements, such as river discharge, precipitation, water chemistry and isotopic value measurements of glacier and snow meltwater can be used to investigate groundwater flow systems and groundwater and surface interactions in areas with distributed permafrost and seasonally frozen ground. The data obtained in this study can be used to explore the different kinds of groundwater flow-related issues, including but not limited to (1) the effect of the freeze-360 thaw process of soils on groundwater flow and the interactions between groundwater and surface water; (2) the interplay between permafrost degradation and groundwater flow changes; and (3) the water quality in alpine catchments under the impacts of seasonal freeze-thaw processes in soils and permafrost degradation. Based on the existing data, the following efforts should be made in the future to obtain a better understanding of the coupled hydrobiogeochemical cycles under the impacts of climate change in 365 alpine catchments: (1) coupling with other monitoring systems (meteorological monitoring systems, remote sensing monitoring systems, and hydrological monitoring systems) to obtain data representing climatic characteristics, the permafrost distribution and its changes, ecological conditions, and river discharges and levels; (2) developing combined flow-heat-solute transport models to understand the impacts of seasonal freeze-thaw processes in soils and permafrost degradation on groundwater flow, on 370 the interaction between groundwater and surface water, and on the water quality and element cycles; and (3) correctly evaluating the impacts of groundwater flow changes on the regional biogeochemical cycle