Next Article in Journal
Anaerobic Digestion for Biogas Production from Municipal Sewage Sludge: A Comparative Study between Fine Mesh Sieved Primary Sludge and Sedimented Primary Sludge
Previous Article in Journal
Putting the “Beaver” Back in Beverley Brook: Rapid Shifts in Community Composition following the Restoration of a Degraded Urban River
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Applicability of Difference in Oxygen-18 and Deuterium of Water Sources and Isotopic Hydrograph Separation in a Bamboo Catchment during Different Rainfall Types

1
College of Water Resources and Hydrology, Hohai University, Nanjing 210098, China
2
Water Resources Department of Jiangsu Province, Nanjing 210029, China
3
Ningbo Hongtai Water Conservancy Information Technology Co., Ltd., Ningbo 315000, China
4
Department of Aquatic Ecosystem Analysis and Management, Helmholtz Centre for Environmental Research-UFZ, 39114 Magdeburg, Germany
5
Department of Ecohydrology and Biogeochemistry, Leibniz Institute of Freshwater Ecology and Inland Fisheries (IGB), 12587 Berlin, Germany
*
Authors to whom correspondence should be addressed.
Water 2021, 13(24), 3531; https://doi.org/10.3390/w13243531
Submission received: 28 September 2021 / Revised: 1 December 2021 / Accepted: 6 December 2021 / Published: 9 December 2021
(This article belongs to the Section Urban Water Management)

Abstract

:
Typhoon storm and plum rain are two typical rainfall types in the lower regions of the Yangtze River Basin, which frequently cause flood disasters in China. New information in stable water isotopes offers the opportunity to advance understanding of runoff mechanisms and water source dynamics in response to these two typical rainfall types. We intensively monitored two representative rainfall events in a small bamboo forestry watershed in 2016. Results showed that precipitation isotopic variations during the event were generally larger than those of other monitored compartments (including throughfall, surface overland water, groundwater and river water) and also larger for the plum rain than for the typhoon event (δ18O varied in 5.2‰ and 3.7‰, respectively). Importantly, the differences of isotopic temporal variation between rainfall and throughfall showed significant impacts on the two-component hydrograph separation for both rainfall types (e.g., if not considered, the pre-event water fractions were 26.6% and 15.3% higher for the typhoon and plum rain events, respectively). Furthermore, we evaluated the role of soil water on the three-component isotopic hydrograph separation model; results revealed that soil water accounted for 10.9% and 28.3% of the total discharge in typhoon and plum rain events, respectively. This underpins the important role of soil water dynamics during the rainy season in this humid region.

1. Introduction

The lower region of the Yangtze River Basin, as one of the most active economic regions of China, has experienced dramatic development over recent decades; meanwhile, high risks of social-economic damages have been raised by flood-induced disasters, which might occur regularly due to the regional climate pattern (i.e., heavy rainfall events occurring during the plum rain and typhoon seasons) [1]. For precise flood predictions and optimal water resource management, it is of highly importance to advance understanding of the regional hydrological cycle and catchment mechanistic responses to the typical plum rain and typhoon events. Stable isotopes of water (δD and δ18O), as an ideal tracer, have been increasingly introduced in hydrology research to inform hydrological processes and their responses to environmental changes [2]. Among others, stable water isotopes and their temporal variations have been widely integrated into discharge hydrograph separation, revealing event-scale water source portioning and its dynamics. However, it is much less intensively studied in the lower region of the Yangtze River Basin with regular plum rain and typhoon events.
Isotopic hydrograph separation (IHS) makes use of the stable isotopes of water (δD and δ18O) as conservative tracers to trace the movement of water through catchments and to estimate how much all water sources contribute to streamflow. Existing IHS studies highlighted the importance of the old water (pre-event water) in the runoff generation [3,4], which contrasts to the traditional perception that “rainfall was considered to dominate storm hydrographs”. Many hydrologists have focused on five assumptions in the traditional two-component hydrograph separation [5,6,7] to improve the understanding of different dominant runoff components in different watersheds.
One assumption is that the isotopic composition of rain or snowmelt is constant, or its spatiotemporal variation can be accounted for [8]. Many studies have found that the isotopic composition of rainfall exhibited significant change during an event and evaluated how the isotopic temporal variation influences the results of IHS [5,9].
Additionally, isotopic composition of direct rainfall from the open sky is different from that of throughfall due to many potential reasons, for example, their different formation processes [10,11]. Some studies have found that, compared with direct rainfall, throughfall is enriched in 18O and D [12], while others have observed completely opposite results [10]. Spatiotemporal variability of throughfall isotopic composition at both the plot and catchment scale has been reported by previous studies [13], but few have explored impacts of the difference between rainfall and throughfall isotopic variations on hydrograph separation to date. This is of particular importance for forestry catchments, where canopy interception might significantly impact catchment responses to rainfall.
Another assumption is that the contribution of soil water can be ignored, or its isotopic composition is the same as that of groundwater. In the development of three-component models in hydrograph separation, interflow from soil water is assumed to be playing a key role in runoff partitioning and in regulating seasonal variations of stream water isotopes [14]. Soil water contributions are not negligible based on investigations of hydro-chemical parameters in different watersheds, and the mechanism of interflow generation is examined in experimental watersheds [15,16]. However, the temporal variability of soil water isotopic composition during rainfall events has not been intensively focused on [17], nor its influence on the results of three-component isotopic hydrograph separation.
Typhoon is a mature tropical cyclone that develops between 180° and 100° E in the Northern Hemisphere typically between July and September; plum rain is known as an annual meteorological cycle of precipitation in the East Asia monsoon region and occurs during May to July in the lower region of the Yangtze River Basin [18]. Plum rain and typhoon events result from two different climatic systems, thus exhibiting differences in runoff partitioning, as well as isotope signatures in different runoff components. Currently, transport trajectories and water sources of plum rain and typhoon have been studied in many basins [19,20], while detailed isotopic comparisons and their applications in hydrograph separation remain rare.
Overall, the isotopic composition of rainfall, throughfall and soil water might exhibit significant temporal variations during the different plum rain and typhoon rainfall events in the lower Yangtze River regions; and considering such variations in hydrograph separation may offer new insights into the regional hydrological processes and their responses to typical rainfall types. The objectives of this study are: (1) to examine the temporal variations and characteristics of isotopic composition in different water sources during different rainfall types at event scale and to illuminate the reasons for the differences; (2) to calculate the hydrograph separation based on isotopes in precipitation and throughfall, and compare the differences of using constant isotopic values and temporal variations during the event; (3) to evaluate the soil water contribution of the two different rainfall types using a three-component hydrograph separation model; and (4) to explore the runoff generation mechanism between rainfall types based on the isotopic signatures and hydrograph separation results. To achieve these objectives, we intensively monitored one typical typhoon and plum rain event in 2016 and analyzed the temporal variations of isotopic compositions in rainfall, throughfall, surface water, soil water, groundwater and river water. Hourly sampling scheme was conducted in the Hemuqiao watershed, a small bamboo watershed in the lower region of the Yangtze River Basin. Together, the stable isotope data obtained from all key hydrological compartments established the data foundation for the detailed hydrograph separation and process understanding in the study watershed.

2. Materials and Methods

2.1. Study Area

The Hemuqiao catchment is a 1.35 km2 bamboo headwater catchment (Figure 1: 30°34′05″~30°34′55″ N; 119°47′05″~119°48′20″ E), located in in the lower regions of the Yangtze River Basin. This wet subtropical monsoon climate region is one of the highest rainfall zones in China. The region has high mean annual rainfall of 1580 mm/y. Most of the rainfall occurs during the monsoon from June to September. The watershed is located in a mountain area, and the elevation varies from 150 to 600 m above sea level. The mean annual air temperature is 14.0 °C, and monthly mean minimum and maximum temperatures are 1.3 °C and 25.0 °C, respectively. Annual average water surface evaporation is 805 mm. Land cover is dominated by bamboo. The study area exhibits multiple soil types, including red soil, yellow soil, lithologic soil, fluvo-aquic soil and paddy soil. The typhoon event on the 14 September and the plum rain event on the 24 June, which occurred in 2016, were selected for study.

2.2. Methods

2.2.1. Sampling Method during the Rainfall Events

Rainfall, throughfall and meteorological data were monitored at Hemuqiao meteorological station during the two selected storms (Figure 1). Precipitation was monitored using a rain gauge with a tipping bucket. A water level sensor was deployed above a rectangular thin-wall weir at the outlet of the catchment to record the water level dynamics. Discharge was calculated using the rating formula of a right triangle thin-wall weir from the water level data.
The sampling locations are shown in Figure 1. Rainfall, throughfall, surface water and soil water sampling sites were located in the middle of the catchment. Groundwater samples were collected from the piezometer site. Stream water was sampled at the outlet of the catchment. Bulk rainwater and throughfall were collected with a 50 cm-diameter stainless steel barrel. In addition, incremental rainfall and throughfall sampling were conducted simultaneously using modified versions of a Kennedy [21] sequential rainfall sampler. At the same time, hourly sampling for rainfall was conducted using a 20 cm-diameter stainless steel barrel. Surface water samples were collected using collectors emplaced in the gully and soil water samples were collected from a soil sampler installed 50 cm beneath the surface. The temporal resolution of samples of rainfall, surface water, soil water and stream water was hourly, from 8:00 to 20:00. Groundwater was sampled every two hours during the typhoon event and was set to have the same value for the baseflow during the plum rain event.
Vacuum distillation was used to extract soil water. The collected samples were analyzed for isotopic compositions (δ18O and δD) using mass spectrometry (Thermo Fisher MAT 253) and EC (electronic conductivity) in the State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University. The isotopic compositions were reported in δ-notation as per mil (‰) relative to the standard isotope ratio of Vienna Standard Mean Ocean Water (V-SMOW). The accuracy of measurements is ±0.2‰ for oxygen-18 and ±1‰ for deuterium.

2.2.2. Hydrograph Separation Method

The two-component isotopic hydrograph separation model is based on the water and isotopic concentration balance equations, as follows:
Q t = Q o + Q n
δ t Q t = δ o Q o + δ n Q n
where Q and δ denote the discharge and the isotopic values of different water components, respectively. Subscripts t, o and n refer to the total stream water, old water (pre-event water, contributing from groundwater and soil water that are stored in watershed prior to event) and new water (event water, contributing from rainfall), respectively. The relative contributions of the old water and new water can be calculated as:
Q o Q t = δ t δ n δ o δ n ,   Q n Q t = δ t δ o δ n δ o
Considering temporal variation of rainfall (throughfall) isotopic composition during the event, an arithmetic mean isotopic value (δ) for the isotopic value can be computed as:
δ = i = 1 n δ i n
where δi denotes the isotopic value of the ith samples, and n denotes the total number of samples throughout the rainfall event.
Additionally, the isotopic hydrograph separation was also conducted using the three-component model, which explicitly separates the pre-event water sources into soil water and groundwater. Equation (5) is used to determine the relative contributions of rainfall, soil water and groundwater:
Q r / Q t Q i / Q t Q g / Q t = 1 1 1 δ 18 O r δ 18 O i δ 18 O g E C r E C i E C g 1 1 δ 18 O t E C t
where δ18O is the oxygen-18 value and EC is electronic conductivity. Subscripts t, r, i and g refer to stream water, rainfall, soil water and groundwater, respectively. If surface overland water is sampled and its isotopic signature is considered to be different from rainfall water, then r refers to surface water.

3. Results and Discussion

3.1. Characteristics of Isotopic Value and EC in Waters during Different Rainfall Type

3.1.1. Rainfall and Throughfall

Characteristics of stable isotope values and EC in the water reservoirs during different storm events at the Hemuqiao catchment are given in Table 1. In this study, rainfall and throughfall isotopic composition both exhibited significant temporal variations and the variation range of rainfall was greater than that of throughfall (for typhoon, δ18O: 3.7‰ Vs 0.6‰, δD: 33.4‰Vs 4.9‰; for plum rain, δ18O: 5.2‰ Vs 1.8‰, δD: 37.7‰ Vs 17.9‰).
The short-term rainfall isotopic fluctuation during a rainfall event can be strongly influenced by factors such as moisture source, rainfall intensity, wind velocity and cloud structure [22,23,24,25]. As a result of their different formation processes, the variation range of rainfall δD and δ18O can be greater than that of throughfall (Table 1). Typically, in forest catchments, throughfall occurs only when cumulative precipitation exceeds canopy storage capacity. This means the intercepted rainfall water will be fully mixed before dropping; meanwhile, the occurrence time of throughfall is later than that of rainfall, resulting in a more damped pattern in the throughfall isotopic composition.
Due to the finer sampling resolution, the temporal variations of rainfall isotopes during the event were nicely captured by our monitoring. In the typhoon event, rainfall δ18O values exhibited the pattern of a first enrichment and were followed by a depletion, during the heavy rainfall period (highlighted in Figure 2a), with increasing rainfall rates, the δ18O values increased from −7.28‰ to −5.58‰ and subsequently decreased to −8.43‰ as the rainfall rates decreased. In contrast, for the plum rain event, rainfall δ18O value decreased from −10.47‰ to −13.52‰ and reached the minimum value of the entire event. During the intensive rainfall period (highlighted in Figure 3a), the δ18O values increased to −8.25‰, reaching the maximum value. Generally, a plum rain event has more abundant total precipitation, a longer duration, significant isotopic fractionation and presentation of amount effect; all together typically resulting in more depleted isotopic values in rainfall. This is in line with our measurements (Table 1), that the averaged isotopic value of rainfall in typhoons (δ18O: −7.7‰; δD: −54.8‰) was higher than that in plum rain (δ18O: −11.1‰; δD: −74.2‰). In these humid regions with high relative humidity, evaporation and small temperature variation in the process of rainfall, with a decrease in temperature and evaporation of raindrops, air saturation is easy to compensate; however, with prolonged rainfall, no evaporation and mixing with surrounding moisture occurred and resulted in a depleted value of rainfall.

3.1.2. Surface Water

From Table 1 and Figure 2a and Figure 3a, it can be seen that although the surface water isotopic composition variation was smaller than that of rainfall (for typhoon, δ18O: 1.7‰ Vs 3.7‰, δD: 9‰Vs 33.4‰; for plum rain, δ18O: 3.0‰ Vs 5.2‰, δD: 8.5‰Vs 37.7‰), it still exhibited significant temporal variation (for typhoon, δ18O: −8.7‰ to −7.0‰, δD: −64.3‰ to −59.4‰; for plum rain, δ18O: −7.2‰ to −4.2‰, δD: −42.9‰ to −34.4‰). The changes in surface water isotopic values might be caused by the dynamic mixing of precipitation during the event and the water stored before the event [26,27]. The significant temporal variability of surface water isotope values emphasizes the importance of measuring the δ18O and δD of sequential surface water samples, rather than using a single fixed value for hydrograph separation [5,12,26,28,29,30,31].
Figure 3. (a) Rainfall and surface water δ18O composition during the plum rainstorm and rainfall amount. (b) δ18O values for soil water, groundwater and streamflow. (c) Discharge and results of the two-component mixing model, using δ18O for pre-event and event waters as a constant. (d) Discharge and results of the two-component mixing model, using temporal variation in δ18O in rainfall.
Figure 3. (a) Rainfall and surface water δ18O composition during the plum rainstorm and rainfall amount. (b) δ18O values for soil water, groundwater and streamflow. (c) Discharge and results of the two-component mixing model, using δ18O for pre-event and event waters as a constant. (d) Discharge and results of the two-component mixing model, using temporal variation in δ18O in rainfall.
Water 13 03531 g003

3.1.3. Soil Water

The soil water isotopic value exhibited significant temporal variation during two type rainfall events (for typhoon, δ18O: −13.7‰ to −5.7‰, δD: −106.1‰ to −54.0‰; for plum rain, δ18O: −11.4‰ to −1.0‰, δD: −89.2‰ to −46.1‰; Table 1 and Figure 2b and Figure 3b). The significant variation in soil water isotopic value reinforces the importance of soil water sampling, which will improve the accuracy of isotopic hydrograph separation. The variation range of soil water δ18O values was greater than that of rainfall (for typhoon, 8‰ Vs. 3.7‰; for plum rain, 10.4‰ Vs. 5.2‰). This indicates that soil water can be replenished not only by event precipitation but also by other water sources, such as groundwater or river water due to the rising groundwater and stream water level.
Figure 4 shows the correlations between δ18O and δD of different water source compartments. The enriched soil water isotope values, i.e., located right-below the meteoric water line, indicate that soil evaporation (in addition to transpiration) has induced fractionation effects on water in the soil zone, and that rainfall are the greatest source of recharging soil water. Some of the soil water values are higher than precipitation, and a possible reason is that soil water in certain parts of the catchment is mixed with enriched pre-event water. A similar conclusion is found in Gazis’s study that δ18O average values in precipitation are −17.2‰, compared with an enriched average in soil water (−14.0‰) [32]. The above results of isotopic soil water revealed that there is a large isotopic fluctuation in soil water, which could be regarded as a mixture of precipitation and pre-event water.

3.1.4. Groundwater

There was consistent limited change in the groundwater isotope values throughout the typhoon event (δ18O: −8.2‰ to −7.1‰, δD: −55.8‰ to −47.2‰; Table 1 and Figure 2b). Isotope values of sampled groundwater prior to the typhoon event were −7.6‰ for δ18O and −51.7‰ for δD, respectively. The δ18O value in the groundwater samples had slightly smaller variation range (1.1‰) compared to that of rainfall (3.7‰), surface water (1.7‰) and soil water (8‰) samples. The relatively constant groundwater isotopic composition supported the assumption that using the pre-event baseflow isotopic value as the initial groundwater isotopic value for hydrograph separation in the Hemuqiao catchment [16].

3.1.5. River Water

Compared with rainfall, the variation ranges of stream water isotopic composition were smaller (for typhoon, δ18O: 1.8‰ Vs. 3.7‰, δD: 6.7‰ Vs. 33.4‰; for plum rain, δ18O: 2.9‰ Vs. 5.2‰, δD: 7.8‰ Vs. 37.7‰); however, it still showed significant temporal fluctuations, suggesting that a constant isotopic value of stream water is unlikely sufficient for hydrograph separation.
In the typhoon event, the dynamic of stream water isotopic composition demonstrated a tendency of a slight decrease followed by an increase, which is opposite to the dynamics of the stream discharge (Figure 2b). At the beginning of the event, the discharge was 0.02m3/s and the average stream water δ18O value was −7.06‰; at the peak time, the discharge increased to 1.60m3/s, and the δ18O value decreased to −8.43‰. After 11h of the peak discharge, the stream discharge decreased to baseflow level (ca. 0.24m³/s), and the stream water δ18O values returned to −6.93‰.
The evaporation of river water can also potentially influence the isotopic value of river water and then affect the component of discharge hydrograph. The river water points to the dual-isotope space being distributed around the LMWL (Figure 4), indicating that precipitation water is the primary recharge source to river water; however, the slope was less than LMWL, indicating that the source waters have also undergone intense dynamic fractionation.

3.2. Isotopic Hydrograph Separation

3.2.1. Impacts of Temporal Variations of Rainfall Isotopic Composition

The results of the two-source isotopic hydrograph separation model in the typhoon event showed that during the first 10.75 h, the river water was dominated by pre-event water (Figure 2d). At the flood peak point, stream discharge was dominated by the event water (69.6%), which may be a result of saturated overland flow during the heavy rainfall period (Figure 2d). Meanwhile, there was still considerable pre-event water contributing to the peak stream water (accounting for 30.4%). In the flood recession period, the event water contribution decreased, and the pre-event water predominated again after 15.25 h of the peak discharge.
Considering temporal variations of rainfall isotopes during the typhoon event, we could observe a three-phase pattern during the rising stage of the hydrograph (Figure 2b,d): (1) the increase of flow is initially supported by pre-event water; (2) then, the event water contribution increased sharply associated with a decrease in stream isotopic values; and (3) finally, the stream water becomes predominated by the event water. Interestingly, such dynamic pattern of event water contribution was not observed when using constant δ18O value. Moreover, the event water fractions were underestimated compared to the former case (i.e., the whole event average: 54.6% Vs. 58.3%; at the discharge peak time: 47.9% Vs. 69.6%; Figure 2c,d). Continuous monitoring of stable isotope δD and δ18O are therefore necessary for capturing the dynamic variations in the rising limb of the stormflow hydrograph. This result emphasizes the importance of the consideration of temporal variation and the use of finer temporal scale [33] in δ18O in rainfall in two-component isotopic hydrograph separation.
In contrast, during the plum rain event, the stream water continued to be predominated by pre-event water (i.e., the whole event average: 77.8%; at the discharge peak time: 76.7%). Moreover, the differences of event water contribution between using the constant rainfall δ18O value and considering its temporal variation during the event were not significant in plum rain events, as compared with typhoon events (Figure 2 and Figure 3). This might have been resulted from the different formations of the two types of rainfall.

3.2.2. Impacts of using Isotopes of Rainfall and Throughfall

The different formation processes between rainfall and throughfall influenced the results of isotopic hydrograph separation for different rainfall types (Table 2; Figure 5a,b).
It can be seen from Table 2 that the pre-event water fractions were lower using throughfall than using precipitation isotopes for both types of storm event; however, the differences were larger in the typhoon event than in the plum rain event (average: 26.6% Vs. 15.3%). We also found that the difference was generally lower for flood peaks than for the whole average values for both rainfall types (typhoon: 15.8% Vs. 26.6%; plum rain: 13.8% Vs. 15.3%).
The distributions of fractions of pre-event water during two storm events are presented in Figure 5a,b. It should be noted that the pre-event water percentages exhibited a large variation in the typhoon event, using either rainfall or throughfall. On average, the pre-event water percentage was 45.4% using rainfall and 18.8% using throughfall. The average pre-event water fraction of the plum rain event was significantly higher (84.0% for rainfall and 68.7% for throughfall) than that of the typhoon event (Table 2). This discrepancy is probably caused by the higher soil moisture content during the long plum rain season from June–July.
Similar phenomena were observed for pre-event water percentages when the flood reached its peak. As the flood crested, the pre-event water percentages were 33.4% for rainfall and 17.6% for throughfall in the typhoon event, while in the plum rain event, they were 82.5% for rainfall and 68.7% for throughfall. In different stages of typhoon event hydrograph, the pre-event water accounted for higher percentages in the rising and recession limbs, while lower percentages in the flood peak, using rainfall or throughfall; the fractions of pre-event water varied more smoothly throughout the whole hydrograph of the plum rain event. Nevertheless, we can draw the conclusion that in a forested watershed, the isotopic values of throughfall and rainfall have strong impacts on hydrograph separation for both rainfall types.

3.2.3. The Contribution of Soil Water Based on the Three-Component Model

In the development of the three-component model for hydrograph separation, soil water plays an increasingly pivotal part in runoff components. Researchers have revealed the fundamental importance of soil water isotopic composition on soil water supply mechanisms, basin runoff formation, and hydrograph separation [34,35]. While the method has been extensively evaluated, the dynamics storage-flux mixing in the soil should be considered, especially when isotopic signals of soil water varied during the event as reported here. In our study, oxygen isotopes and electrical conductivity were used as hydrological tracers in the three-component hydrograph separation. The fractions of surface water, soil water and groundwater are shown in Table 2, and the separation results of the three-component model are presented in Figure 5c,d.
In the typhoon event, the three water component fractions changed prominently, which also reflected remarkable isotopic changes in the water sources. Source fractions in the plum rain event changed greatly in the earlier stage; however, this change tended to be steady, along with rainfall. Fractions of surface water and groundwater simultaneously changed in the opposite trend, while that of soil water fluctuated dynamically (Figure 5d). The average fractions of surface water and soil water were considerably lower in the typhoon event than in the plum rain event, and the fractions of groundwater were higher in the typhoon event (Table 2).
Water source fractions in different types of rainfall events had diverse trends from the beginning to the end of the event episode. From Figure 5c,d, due to the smaller rainfall amount at the beginning of the rainfall, contributions from surface water and soil water remained low, regardless of the type of rainfall and groundwater, giving priority to the discharge in the river. Along with the development of rainfall events, fractions of soil water and groundwater showed contrary trends. Soil water fractions increased first and then decreased until the flooding peaks. This demonstrates that the river was much less influenced by the soil water at the flood peak. In the late period of the event, the percentage of surface water was low and those of soil water and groundwater were high. The event-averaged percentages of surface water, soil water and groundwater were 43.8%, 10.9% and 45.4%, respectively, for the typhoon event and 62.0%, 14.0% and 24.0%, respectively, for the plum rain event. The results showed that soil water occupied a certain proportion in the water sources, indicating that soil water could not be neglected when separating hydrographs.
In the period of the flood peak, an increasing fraction of surface water was found, achieving 40~60%. This means soil water and groundwater together contributed approximately 40%. The two-component model underestimated the fraction of event water, when larger fractions of surface water were discharged during the flood. The maximum fraction of soil water usually occurred adjacent to the flood peak, instead of at the flood peak time, when the fractions of soil water were markedly lower, 14.0% for the typhoon and 20.5% for the plum rain.

4. Conclusions

The hourly isotopic data from different compartments of the catchment hydrological cycling were sampled during typical typhoon and plum rain in this study. The measurements indicated that δ18O in precipitation and throughfall exhibited significant temporal variations, and the throughfall was more enriched in heavy isotopes than the direct rainfall. Isotope values in precipitation during the typhoon event had stronger variation compared with those during the plum rain event. Compared with rainfall, isotope concentrations in surface water were more damped. The groundwater δD and δ18O remained stable during the two-type storm events. δD and δ18O of soil water had significant variations, and the variation was more significant in the typhoon event than in the plum rain event.
The two-component isotopic hydrograph separation model results revealed that: (1) pre-event fractions were higher for the plum rain than for the typhoon event; and (2) fractions in the typhoon event exhibited remarkable variation compared with the moderate variation in the plum rain. Furthermore, the pre-event water fraction was overestimated by 26.6% in typhoon storm and by 15.3% in plum rain when using rainfall isotopic value, compared to the results of using throughfall isotopic value in this watershed. This indicates that, in the forested watershed, the difference between rainfall and throughfall isotopic values should be considered in both rainfall types for hydrograph separation.
The results of the three-component isotopic hydrograph separation model in the Hemuqiao showed that fractions of surface water and groundwater had inverse dynamic patterns, and fractions of soil water varied dynamically in-between, accounting for 10.9% and 28.3% of the total stream flow in the typhoon and the plum rain events, respectively. The calculated three-source fractions in the typhoon event changed significantly during the course of rainfall, and the fractions in plum rain varied more strongly in the earlier stage of the event and tended to be steady in the following stage.
This paper comprehensively compared the differences between rainfall and throughfall and investigated the impacts of considering temporal variations of rainfall during the event on the isotopic hydrograph separation based on the two-component method. The importance of soil water isotopic values was further explored based on the three-component method. Based on our study, further studies focusing on the influence of both temporal and spatial variations of different water sources on isotopic hydrograph separation are of high interest. Furthermore, the coupling of isotope and hydrological models will improve our understanding of basin runoff generation and concentration and will provide advice for basin water resource planning and flood management.

Author Contributions

Conceptualization, Y.Y. and S.Q.; methodology, Y.W.; software, P.S.; formal analysis, Y.J.; resources, Q.Y.; data curation, P.S.; writing—original draft preparation, Y.Y. and S.Q.; writing—review and editing, X.Y. All authors have read and agreed to the published version of the manuscript.

Funding

The study is financially supported by the National Key Research and Development Program of China (2019YFC0409000), the National Natural Science Foundation of China (52179011), the Fundamental Research Funds for the Central Universities (2019B41014).

Data Availability Statement

The data used in this paper will be made available upon request; please send a request to [email protected] for data.

Conflicts of Interest

The authors declare no competing interests.

References

  1. Cui, L.; Wang, L.; Lai, Z.; Tian, Q.; Liu, W.; Li, J. Innovative trend analysis of annual and seasonal air temperature and rainfall in the Yangtze River Basin, China during 1960–2015. J. Atmos. Solar-Terr. Phys. 2017, 164, 48–59. [Google Scholar] [CrossRef]
  2. Zhang, Y.; Song, C.; Zhang, K.; Cheng, X.; Band, L.E.; Zhang, Q. Effects of land use/land cover and climate changes on terrestrial net primary productivity in the Yangtze River Basin, China, from 2001 to 2010. J. Geophys. Res. Biogeosci. 2014, 119, 1092–1109. [Google Scholar] [CrossRef]
  3. Buttle, J.M. Isotope hydrograph separations and rapid delivery of pre-event water from drainage basins. Prog. Phys. Geogr. Earth Environ. 1994, 18, 16–41. [Google Scholar] [CrossRef]
  4. Gibson, J.J.; Edwards, T.W.D.; Birks, S.J.; St Amour, N.A.; Buhay, W.M.; McEachern, P.; Peters, D.L. Progress in isotope tracer hydrology in Canada. Hydrol. Process. 2005, 19, 303–327. [Google Scholar] [CrossRef]
  5. Mcdonnell, J.J.; Bonell, M.; Stewart, M.K.; Pearce, A.J. Deuterium variations in storm rainfall: Implications for stream hydrograph separation. Water Resour. Res. 1990, 26, 455–458. [Google Scholar] [CrossRef]
  6. Ogunkoya, O.O.; Jenkins, A. Analysis of storm hydrograph and flow pathways using a three-component hydrograph separation model. J. Hydrol. 1993, 142, 71–88. [Google Scholar] [CrossRef]
  7. Pionke, H.B.; Gburek, W.J.; Folmar, G.J. Quantifying stormflow components in a Pennsylvania watershed when 18 O input and storm conditions vary. J. Hydrol. 1993, 148, 169–187. [Google Scholar] [CrossRef]
  8. Sklash, M.G.; Farvolden, R.N. The role of groundwater in storm runoff. J. Hydrol. 1979, 43, 45–65. [Google Scholar] [CrossRef]
  9. Laudon, H.; Hemond, H.F.; Krouse, R.; Bishop, K.H. Oxygen 18 fractionation during snowmelt: Implications for spring flood hydrograph separation. Water Resour. Res. 2002, 38, 40–41. [Google Scholar] [CrossRef] [Green Version]
  10. Xu, X.; Guan, H.; Deng, Z. Isotopic composition of throughfall in pine plantation and native eucalyptus forest in South Australia. J. Hydrol. 2014, 514, 150–157. [Google Scholar] [CrossRef]
  11. Ikawa, R.; Yamamoto, T.; Shimada, J.; Shimizu, T. Temporal variations of isotopic compositions in gross rainfall, throughfall, and stemflow under a Japanese cedar forest during a typhoon event. Hydrol. Res. Lett. 2011, 5, 32–36. [Google Scholar] [CrossRef]
  12. Allen, S.T.; Brooks, J.R.; Keim, R.F.; Bond, B.J.; Mcdonnell, J.J. The role of pre-event canopy storage in throughfall and stemflow by using isotopic tracers. Ecohydrology 2014, 7, 858–868. [Google Scholar] [CrossRef]
  13. Allen, S.T.; Keim, R.F.; Mcdonnell, J.J. Spatial patterns of throughfall isotopic composition at the event and seasonal timescales. J. Hydrol. 2015, 522, 58–66. [Google Scholar] [CrossRef]
  14. Grieu, P.; Lucero, D.W.; Ardiani, R.; Ehleringer, J.R. The mean depth of soil water uptake by two temperate grassland species over time subjected to mild soil water deficit and competitive association. Plant Soil 2001, 230, 197–209. [Google Scholar] [CrossRef]
  15. Klaus, J.; Mcdonnell, J.J. Hydrograph separation using stable isotopes: Review and evaluation. J. Hydrol. 2013, 505, 47–64. [Google Scholar] [CrossRef]
  16. Mcdonnell, J.J.; Stewart, M.K.; Owens, I.F. Effect of Catchment—Scale Subsurface Mixing on Stream Isotopic Response. Water Resour. Res. 1991, 27, 3065–3073. [Google Scholar] [CrossRef]
  17. Carey, S.K.; Quinton, W.L. Evaluating runoff generation during summer using hydrometric, stable isotope and hydrochemical methods in a discontinuous permafrost alpine catchment. Hydrol. Process. 2005, 19, 95–114. [Google Scholar] [CrossRef]
  18. Ding, Y.; Wang, Z.; Sun, Y. Inter-decadal variation of the summer precipitation in East China and its association with decreasing Asian summer monsoon. Part I: Observed evidences. Int. J. Climatol. 2008, 28, 1139–1161. [Google Scholar] [CrossRef]
  19. Wenguang, Z.; Jingyi, M.; Bo, L.; Shichun, Z.; Jing, Z.; Ming, J.; Xianguo, L. Sources of monsoon precipitation and dew assessed in a semiarid area via stable isotopes. Hydrol. Process. 2017, 31, 1990–1999. [Google Scholar] [CrossRef]
  20. Qu, S.; Wang, Y.; Zhou, M.; Liu, H.; Shi, P.; Yu, Z.; Xiang, L. Temporal 18O and deuterium variations in hydrologic components of a small watershed during a typhoon event. Isot. Environ. Health Stud. 2016, 53, 172–183. [Google Scholar] [CrossRef]
  21. Kennedy, V.C.; Zellweger, G.W.; Avanzino, R.J. Variation of rain chemistry during storms at two sites in northern California. Water Resour. Res. 1979, 15, 687–702. [Google Scholar] [CrossRef]
  22. Celle-Jeanton, H.; Travi, Y.; Blavoux, B. Isotopic typology of the precipitation in the Western Mediterranean Region at three different time scales. Geophys. Res. Lett. 2001, 28, 1215–1218. [Google Scholar] [CrossRef]
  23. Johnson, K.R.; Ingram, B.L. Spatial and temporal variability in the stable isotope systematics of modern precipitation in China: Implications for paleoclimate reconstructions. Earth Planet. Sci. Lett. 2004, 220, 365–377. [Google Scholar] [CrossRef] [Green Version]
  24. Liu, Z.; Tian, L.; Chai, X.; Yao, T. A model-based determination of spatial variation of precipitation δ 18 O over China. Chem. Geol. 2008, 249, 203–212. [Google Scholar] [CrossRef]
  25. Pionke, H.B.; Dewalle, D.R. Intra- and inter-storm 18 O trends for selected rainstorms in Pennsylvania. J. Hydrol. 1992, 138, 131–143. [Google Scholar] [CrossRef]
  26. Hu, H.; Bao, W.; Wang, T.; Qu, S. Isotopic exchange of soil water in water cycle. In Advances in Water Resources and Hydraulic Engineering—Proceeding of 16th IAHR-APD Congress and 3rd Symposium of IAHR-ISHS; Zhang, C.T.H., Ed.; Springer: Berlin/Heidelberg, Germany, 2009; pp. 188–192. [Google Scholar]
  27. Dinçer, T.; Payne, B.R.; Florkowski, T.; Martinec, J.; Tongiorgi, E. Snowmelt runoff from measurements of tritium and oxygen-18. Water Resour. 1970, 6, 110–124. [Google Scholar] [CrossRef]
  28. Iorgulescu, I.; Beven, K.J.; Musy, A. Flow, mixing, and displacement in using a data-based hydrochemical model to predict conservative tracer data. Water Resour. Res. 2007, 43, 223–224. [Google Scholar] [CrossRef]
  29. Lyon, S.W.; Desilets, S.L.E.; Troch, P.A. Characterizing the response of a catchment to an extreme rainfall event using hydrometric and isotopic data. Water Resour. Res. 2008, 44, 214. [Google Scholar] [CrossRef]
  30. Segura, C.; James, A.L.; Lazzati, D.; Roulet, N.T. Scaling relationships for event water contributions and transit times in small-forested catchments in Eastern Quebec. Water Resour. Res. 2012, 48, 7502. [Google Scholar] [CrossRef]
  31. Weiler, M.; Mcglynn, B.L.; Mcguire, K.J.; Mcdonnell, J.J. How does rainfall become runoff? A combined tracer and runoff transfer function approach. Water Resour. Res. 2003, 39, 57–76. [Google Scholar] [CrossRef] [Green Version]
  32. Robertson, J.A.; Gazis, C.A. An oxygen isotope study of seasonal trends in soil water fluxes at two sites along a climate gradient in Washington state (USA). J. Hydrol. 2006, 328, 375–387. [Google Scholar] [CrossRef]
  33. Tweed, S.; Munksgaard, N.; Marc, V.; Rocket, N.; Bass, A.; Forsythe, A.J.; Bird, M.I.; Leblanc, M. Continuous monitoring of stream δ18O and δ2H and stormflow hydrograph separation using laser spectrometry in an agricultural catchment. Hydrol. Proc. 2016, 30, 648–660. [Google Scholar] [CrossRef]
  34. Araguás-Araguás, L.; Rozanski, K.; Gonfiantini, R.; Louvat, D. Isotope effects accompanying vacuum extraction of soil water for stable isotope analyses. J. Hydrol. 1995, 168, 159–171. [Google Scholar] [CrossRef]
  35. Gazis, C.; Feng, X. A stable isotope study of soil water: Evidence for mixing and preferential flow paths. Geoderma 2004, 119, 97–111. [Google Scholar] [CrossRef]
Figure 1. Location of the Hemuqiao catchment and sampling sites for different water sources.
Figure 1. Location of the Hemuqiao catchment and sampling sites for different water sources.
Water 13 03531 g001
Figure 2. (a) Rainfall and surface water δ18O composition during the typhoon storm and rainfall amount. (b) δ18O values for soil water, groundwater and streamflow. (c) Discharge and results of the two-component mixing model, using δ18O for pre-event and event waters as a constant. (d) Discharge and results of the two-component mixing model, using temporal variation in δ18O in rainfall.
Figure 2. (a) Rainfall and surface water δ18O composition during the typhoon storm and rainfall amount. (b) δ18O values for soil water, groundwater and streamflow. (c) Discharge and results of the two-component mixing model, using δ18O for pre-event and event waters as a constant. (d) Discharge and results of the two-component mixing model, using temporal variation in δ18O in rainfall.
Water 13 03531 g002
Figure 4. All δ18O and δD data of rainfall, surface water, soil water, groundwater, and stream water relative to the local (LMWL) and global (GMWL) meteoric water lines. The LWML is calculated from rainfall collected in all rainfall events between 2015 and 2016 in the Hemuqiao catchment. (a) Typhoon event and (b) plum rain event.
Figure 4. All δ18O and δD data of rainfall, surface water, soil water, groundwater, and stream water relative to the local (LMWL) and global (GMWL) meteoric water lines. The LWML is calculated from rainfall collected in all rainfall events between 2015 and 2016 in the Hemuqiao catchment. (a) Typhoon event and (b) plum rain event.
Water 13 03531 g004
Figure 5. Results of isotopic hydrograph separation: (a) two-source model for typhoon, (b) two-source model for plum rain, (c) three-component model for typhoon, and (d) three-component model for plum rain.
Figure 5. Results of isotopic hydrograph separation: (a) two-source model for typhoon, (b) two-source model for plum rain, (c) three-component model for typhoon, and (d) three-component model for plum rain.
Water 13 03531 g005
Table 1. Characteristics of stable water isotopes and EC in each hydrological compartment during the typhoon and plum rain events at the Hemuqiao catchment.
Table 1. Characteristics of stable water isotopes and EC in each hydrological compartment during the typhoon and plum rain events at the Hemuqiao catchment.
Water SourceRainfall Typeδ18O(‰)δD(‰)EC(μS/cm)
MinMaxMeanMinMaxMeanMinMaxMean
RainfallTyphoon−9.1−5.4−7.7−64.3−30.9−54.820.258.643.7
Plum rain−13.5−8.3−11.1−95.3−57.6−74.230.470.254.1
ThroughfallTyphoon−8.7−8.1−8.3−64.3−59.4−62.047.450.949.2
Plum rain−13.0−11.2−11.9−89.9−72.0−79.342.469.254.7
Surface waterTyphoon−8.7−7.0−8.0−57.5−48.5−53.870.498.484.7
Plum rain−7.2−4.2−6.3−42.9−34.4−40.233.157.944.2
Soil waterTyphoon−13.7−5.7−8.3−106.1−54.0−71.457.679.367.6
Plum rain−11.4−1.0−7.2−89.2−46.1−67.343.188.167.1
GroundwaterTyphoon−8.2−7.1−7.6−55.8−47.2−51.7141.7179.9157.3
Plum rain−4.5−4.5−4.5−36.9−36.9−36.943.843.843.8
Stream waterTyphoon−8.7−6.9−7.9−56.6−49.9−53.689.6136.7105.2
Plum rain−7.5−4.6−6.7−44.7−36.9−41.546.978.860.2
Table 2. Results of isotopic separation model in the Hemuqiao watershed.
Table 2. Results of isotopic separation model in the Hemuqiao watershed.
Event TypePre-Event Water Fraction Based on Two-Source IHS ModelThree-Component IHS Model
Peak/%Average/%Peak Fraction/%Average Fraction/%
PrecipitationThroughfallPrecipitationThroughfallSurface WaterSoil WaterGroundwaterSurface WaterSoil WaterGroundwater
Typhoon33.417.645.418.862.014.024.043.810.945.4
Plum rain82.568.784.068.742.820.536.843.828.328.4
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

You, Y.; Qu, S.; Wang, Y.; Yang, Q.; Shi, P.; Jiang, Y.; Yang, X. Applicability of Difference in Oxygen-18 and Deuterium of Water Sources and Isotopic Hydrograph Separation in a Bamboo Catchment during Different Rainfall Types. Water 2021, 13, 3531. https://doi.org/10.3390/w13243531

AMA Style

You Y, Qu S, Wang Y, Yang Q, Shi P, Jiang Y, Yang X. Applicability of Difference in Oxygen-18 and Deuterium of Water Sources and Isotopic Hydrograph Separation in a Bamboo Catchment during Different Rainfall Types. Water. 2021; 13(24):3531. https://doi.org/10.3390/w13243531

Chicago/Turabian Style

You, Yang, Simin Qu, Yifan Wang, Qingyi Yang, Peng Shi, Yuxun Jiang, and Xiaoqiang Yang. 2021. "Applicability of Difference in Oxygen-18 and Deuterium of Water Sources and Isotopic Hydrograph Separation in a Bamboo Catchment during Different Rainfall Types" Water 13, no. 24: 3531. https://doi.org/10.3390/w13243531

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop