Characteristics and Influencing Factors of Dissolved Organic Matter in Cave Drip Water—A Case Study from Furong Cave, Southwest China

Abstract

Understanding the hydrological processes of dissolved organic matter (DOM) in the surface karst zone is crucial for the utilization and safety of groundwater resources. However, research on DOM in drip water from karst caves is limited. In this study, continuous monitoring was conducted for four years at four drip water monitoring sites (MP1, MP2, MP3, and MP9) in Furong Cave, Southwest China. The three-dimensional fluorescence excitation–emission spectroscopy (3D-EEM) and fluorescence region integration (FRI) methods were employed, along with correlation analysis, to investigate the spectral characteristics, composition, sources, and influencing factors of the DOM in the drip water of Furong Cave. The results indicated that (1) the three-dimensional fluorescence peaks in the drip water were classified into six categories: A, B, C, T, T_i, and M. (2) The dominant source of the drip water DOM is endogenous organic matter. (3) The FRI analysis indicates a relatively high proportion of Type II substances in the drip water, predominantly composed of tryptophan-like substances. (4) The DOM in the drip water of Furong Cave was influenced by various factors, including the mixing effect of “new” and “old” water, water residence time in karst systems, and water–rock interactions (WRI), resulting in the complex responses of drip rates and DOM to surface precipitation and temperature at different drip water sites. This study provides a reference for comparative research on DOM in cave drip water in karst regions, which contributes to a better understanding of the migration mechanism of DOM in karst aquifers under different climate and karst ecological conditions.

1. Introduction

The survival and development of over 25% of the global population depend on water resources in karst aquifers [1]. The unique geological structures in karst regions facilitate the infiltration of surface pollutants into the underlying aquifers through the overlying soil or fractures, resulting in the contamination of groundwater resources [2]. The transport of solutes is typically controlled by hydrological processes in karst areas, highlighting the significance of understanding hydrological processes within karst regions [3].

Furthermore, utilizing organic proxies in stalagmites and conducting multi-proxy comparisons can enhance the accuracy of the paleoclimate interpretation of stalagmites. Compared to traditional proxies such as oxygen isotopes and trace elements in stalagmite paleoclimate reconstructions, organic matter in stalagmites has the ability to record information associated with ecosystem and environmental changes. In recent decades, a variety of organic proxies in stalagmites have been applied in paleoclimate reconstruction studies. These organic proxies include organic carbon isotopes [4], GDGTs [5,6], branched fatty acids, and hydroxy fatty acids [7,8]. Research on fluorescent organic matter in stalagmites has primarily focused on utilizing organic luminescence laminae as a research tool; the application of stalagmite organic matter fluorescence as a proxy has been employed to reconstruct paleoenvironmental and paleoclimatic changes [9,10,11]. Despite the significant achievements of organic matter in stalagmites in the field of paleoclimate reconstruction, the wide applicability and uncertainties of the proxy indicators in stalagmite-based paleoclimate reconstructions still require validation. There is an urgent need to continuously develop new proxies and for the exploration of novel testing methods to enhance the field of stalagmite paleoclimate reconstructions. Similar to the proxies of oxygen and carbon isotopes and trace elements in stalagmites, understanding the significance of organic matter in stalagmite paleoclimate reconstruction through modern cave monitoring and deciphering the signal transmission and mechanisms of organic matter in karst systems is particularly important. Long-term monitoring studies in caves have primarily focused on stable isotopes (δ¹⁸O, δ¹³C) [12,13,14,15,16] and trace elements (Mg, Ca Ba Sr) as representatives of inorganic geochemical processes with climatic implications [17,18,19,20]; relatively less attention has been paid to organic geochemical processes [3,21,22,23,24]. Currently, the modern monitoring of organic matter in caves has been limited in duration, with the majority of studies focused on investigating the mechanisms behind the formation of fluorescent luminescence laminae in stalagmites [10,25,26]. The contemporary processes of organic matter from drip water to stalagmites in karst regions, aiming to gain insights into the output mechanisms and key controlling factors of colloidal organic matter (COM) in the karst vadose zone, as well as its response to external climate dynamics [22,27]. Furthermore, the evolution patterns, sources, and controlling factors of dissolved organic matter (DOM) in drip water in the southwestern region of China remain unclear.

COM encompasses organic particulate matter with particle sizes larger than 0.22 μm [22]. While DOM represents a heterogeneous mixture of aromatic and aliphatic organic compounds [28,29,30,31,32] that plays a crucial role in global carbon cycling and ecosystem functioning [33]. DOM originates from various sources, including soil and soluble microbial products [34,35,36,37,38]. Conducting a systematic investigation of the dynamic changes and controlling factors of DOM in karst regions contributes to the rational management and utilization of water resources in karst areas [39,40,41,42].

Previous studies have been conducted on rainwater DOM in karst and non-karst areas [3,43,44], as well as the DOM in drip water [3,27,45,46]. However, the transport mechanism of DOM is especially intricate due to the influence of hydrogeological processes on its composition [45]. Research in karst regions on DOM and hydrochemical characteristics primarily focuses on the underground rivers in karst areas and the soil–karst groundwater systems that are sensitive to infiltration events [47]. Currently, there is limited research on the migration mechanisms of DOM in cave drip water over long-term continuous monitoring in karst areas [23,39]. The long-term temporal variations and the interannual-scale controlling factors of DOM in drip water remain unclear. Strengthening research on cave drip water DOM in karst regions is of the utmost importance in assessing the impact of climate change on the process of organic matter export. This study delved into the transport mechanisms and controlling factors of DOM in karst systems, contributing to a comprehensive understanding of the climate and environmental implications of organic proxies in geological carriers, such as stalagmites.

Over the past few decades, fluorescence spectroscopy has emerged as a potent tool for revealing the chemical characteristics of DOM and has been widely employed in DOM analysis. Three-dimensional excitation–emission matrix (3D-EEM) spectroscopy allows for the acquisition of spectral information from different fluorescent groups, enabling the determination of DOM composition and identification of its sources [48,49]. Parallel factor analysis (PARAFAC) is utilized to qualitatively and quantitatively extract information from excitation–emission matrices (EEM) [50,51]. By coupling three-dimensional excitation–emission matrices with fluorescence regional integration (3D-EEM-FRI), spectral information identification and the characterization of multi-component fluorescence spectra can be achieved [3,52].

There remains a lack of comprehensive comparisons regarding the transport characteristics and main controlling factors of DOM in karst systems under different climatic and geological backgrounds. To further investigate the transport characteristics of DOM in karst aquifers, this study specifically aims to (1) compare the chemical composition and sources of DOM among different monitoring sites. (2) Investigate the transport mechanisms of DOM and the influence of aquifer processes on DOM composition. This study provides new reference data on the environmental significance and comprehensive comparison of DOM in karst waters from various regions in the monsoon zone of China, based on continuous cave monitoring data and employing the excitation–emission matrix fluorescence region integration (EEM-FRI) technique.

2. Materials and Methods

2.1. Study Site

Furong Cave (29°13′44″ N, 107°54′13″ E, and 480 m a.s.l) is located in the southwestern region of China and represents a characteristic karst valley area (Figure 1a). It has developed within the geological formations of Cambrian limestone and dolomite [53,54]. The region exhibits a subtropical humid monsoon climate characterized by hot and wet summers, and cold and dry winters. The annual average temperature is 18.3 °C, and the annual average precipitation ranges from 1000 to 1200 mm. The period from May to October accounts for 70–80% of the annual rainfall [54] (Figure 1b). The region above the cave is characterized by lush vegetation, predominantly composed of trees and shrubs. The soil thickness ranges from approximately 30 to 90 cm, with the dominant soil types being yellow soil and yellow-brown soil [55]. The cave maintains a consistent annual mean temperature of around 16–16.3 °C [54], and the relative humidity of the cave air approaches 95–100%.

2.2. Drip Water Monitor Sites in the Cave

Four drip water monitoring sites (MP1, MP2, MP3, MP9) were set up in Furong Cave (Figure 1c) [54]. The drip heights at MP1, MP2, and MP9 are estimated to be approximately 22 m, 27 m, and 42 m, respectively. Each water droplet disperses into multiple droplets as it falls from the cave ceiling [56]. Drip site MP3 was located on the cavity wall, and the drip height was approximately 0.3 m. MP2 and MP9 exhibit year-round dripping. There is a good correlation between the soil moisture content above the cave and precipitation [57]. During periods of low precipitation, the retention time of infiltrated water in soil pores increases, leading to enhanced mineralization processes. Conversely, with increased precipitation, the amount of infiltrated water transported into the soil also increases [17].

2.3. Cave Monitoring and Collection of Samples

Temperature data outside Furong Cave were recorded using temperature data loggers (HOBO U12-011, Onset, MA, USA) set to automatically record air temperature every 2 h. The measurement range of the loggers was −40–70°C, with an error margin of ±0.1 °C [58]. Precipitation measurements were recorded using an automated rain gauge (RG3-M, Onset, MA, USA) at a frequency of one measurement every 2 min. The rain gauge has a precision of ±1.0%. The loggers of temperature and precipitation were installed 1.5 m above the ground [56].

We conducted monthly on-site monitoring of the cave environment near the drip sites in Furong Cave, including cave air temperature, humidity, pCO₂, as well as drip rate and pH values of the drip water. The water samples were analyzed on-site using a portable multi-parameter testing device, the Multi350i (WTW Co., Munich, Germany) for measurements of conductivity (EC), pH, and water temperature. The measurement accuracies for EC, pH, and water temperature were ±0.25%, ±0.01pH, and ±0.1 °C, respectively. The variations in pCO₂ concentration in the cave air were monitored near each drip water site using the Testo 535 carbon dioxide monitor (Testo Co., Black Forest, Germany) [58]. The measurement range of the monitor was 0–9999 ppm, with a resolution of 1 ppm and an error of 2% [59]. Bicarbonate (HCO₃⁻) content was determined by on-site titration using alkalinity reagent kits manufactured by Merck KGaA Co., Darmstadt, Germany, with a measurement accuracy of 0.1 mmol/L. All the mentioned parameters were recorded during the monitoring process.

Water samples were collected from four drip monitoring sites (MP1, MP2, MP3, and MP9) at approximately every 30 days intervals during the period from July 2019 to June 2023. Transparent glass bottles with a volume of 10 mL were used for collecting the drip water samples. The glass bottles were soaked in a 5% nitric acid solution for 6 h prior to use, rinsed three times with ultrapure water, and allowed to dry naturally. The pre-cleaned glass bottles were placed underneath each drip site to collect the drip water samples, with collection times ranging from 1 to 3 h depending on the drip rate. After collecting the drip water samples, the bottles were sealed with sealing film, transported to the laboratory, and stored in a refrigerator at 4 °C for subsequent testing. Due to the COVID-19 pandemic, samples were not collected during certain time periods, including January 2020 to May 2020, December 2021 to March 2022, and November 2022 to January 2023 for all sites, as well as June 2021 for MP1, March 2021 to April 2021 for MP3, and November 2020 for MP9. During monthly cave sampling, a pre-cleaned 20 mL graduated cylinder, which had been rinsed three times with ultrapure water, was placed underneath each monitoring site to collect and measure the drip water. The drip rate (mL/min) was determined by measuring the volume of drip water collected in the graduated cylinder within a 1 min interval using a stopwatch.

2.4. Sample Analysis

The study employed the Horiba Jobin Yvon Aqualog three-dimensional fluorescence spectrometer at the State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences (Wuhan), to perform three-dimensional excitation–emission matrix (3D-EEM) analysis on drip water samples collected from July 2019 to June 2023. The testing and analysis procedures were conducted at room temperature, with deionized water (Milli-Q, Merck KGaA Co., Darmstadt, Germany) used to monitor the instrument’s stability and precision [22]. The instrument parameters were set as follows: an integration time of 0.5 s, an excitation wavelength measurement range of 240–600 nm with a step size of 3 nm, and an emission wavelength measurement range of 246–824 nm with a step size of 3 nm. To eliminate instrument-specific biases, inner-filter effect correction was applied to each data matrix, and the in situ data processing software of Aqualog V3.6 was utilized for Rayleigh scatter correction [60,61].

In this study, fluorescence region integration (FRI) was employed for quantitative analysis of the three-dimensional fluorescence spectra, where the integrated standardized volume of each region obtained through FRI represents the relative content of corresponding DOM regions [52]. The FRI method divided the three-dimensional fluorescence spectra into five regions (Figure 2), each representing different substances (Table S1) [52]. The cumulative fluorescence intensity (Φ_i) of organic matter with similar properties within specific fluorescence regions was calculated using Equation (1) [52]:

$${\Phi }_i={\displaystyle \int {}_{\mathrm{ex}}}{\displaystyle \int {}_{\mathrm{em}}}I({\lambda }_{\mathrm{ex}}{\lambda }_{\mathrm{em}})d{\lambda }_{\mathrm{ex}}d{\lambda }_{\mathrm{em}}$$

(1)

$${\Phi }_{i,n}=M{F}_i{\Phi }_i$$

(2)

$${\Phi }_{T,n}={\displaystyle \sum _{i=1}^5{\Phi }_{i,n}}$$

(3)

$$P_{i,n}={\Phi }_{i,n}/{\Phi }_{T,n}\times 100\%$$

(4)

In the equation, Φ_i,n represents the normalized integrated volume of fluorescence region _i, Φ_i represents the integrated volume of fluorescence region _i, λ_ex and λ_em denote the excitation and emission wavelengths, respectively, I(λ_exλ_em) represents the fluorescence intensity corresponding to the excitation–emission wavelength pair, Φ_T,n represents the normalized integrated volume of the total fluorescence region, P_i,n represents the proportion of the normalized integrated volume of fluorescence region _i to the total normalized integrated volume, MF_i is the multiplication factor, which is the reciprocal of the proportion of the integrated area of fluorescence region _i to the total integrated area of all fluorescence regions [52].

2.5. Fluorescence Spectroscopic Parameters

The definitions of the three-dimensional fluorescence spectroscopic parameters can be found in Table S2. The fluorescence index (FI) represents the ratio of fluorescence intensities at emission wavelengths of Em = 450 nm and Em = 500 nm when excited at a wavelength of Ex = 370 nm. It characterizes the humic substance sources of DOM in drip water [62,63]. Studies have shown that when FI = 1.4, the fluorescence components in DOM primarily originate from terrestrial sources, while an FI value of 1.9 indicates a microbial source [63]. BIX, the biological index, indicates the ratio of fluorescence intensities at emission wavelengths of Em = 380 nm and Em = 430 nm when excited at a wavelength of Ex = 310 nm [64,65]. BIX reflects the contribution of protein-like DOM fractions in the water, and a higher BIX value suggests a larger contribution of protein-like DOM [66]. HIX, the humification index, indicates the ratio of the integrated fluorescence intensities in the range of emission wavelengths Em = 435–480 nm to the integrated fluorescence intensities in the range of emission wavelengths Em = 300–345 nm when excited at a wavelength of Ex = 254 nm.

3. Results

3.1. Climate Variation

During the monitoring period, the monthly rainfall ranged from 10 to 405 mm, with higher rainfall occurring during the summer months (June to August) and lower rainfall during the winter months (December to February) (Figure 3a). The annual rainfall varied between 874 mm and 1613 mm, with an average annual rainfall of 1295 mm (Figure 3a). The monthly average temperature ranged from 5.7 to 32.1 °C, with the lowest temperatures observed in December, January, and February, and the highest temperatures recorded in July and August (Figure 3b).

3.2. Chemical Characteristics of Drip Water

During the monitoring period, the EC variations in the drip water at four monitoring sites (MP1, MP2, MP3, MP9) in Furong Cave ranged from 345 to 560 μs/cm, 420 to 607 μs/cm, 401 to 571 μs/cm, and 306 to 596 μs/cm, with mean values of 449 μs/cm, 508 μs/cm, 493 μs/cm, and 481 μs/cm, respectively (Figure 3e). The pH values exhibited ranges of 6.80~8.29, 6.60~8.30, 6.60~8.10, and 6.30~8.30, with mean values of 7.56, 7.66, 7.25, and 7.58, respectively (Figure 3d). The variations in HCO₃⁻ concentrations ranged from 4.9 to 6.9 mmol/L, 5.2 to 6.6 mmol/L, 4.8 to 7.8 mmol/L, and 4.7 to 7.1 mmol/L, with mean values of 5.5 mmol/L, 6.1 mmol/L, 5.9 mmol/L, and 6.0 mmol/L, respectively (Figure 3f).

From January 2019 to June 2023, the drip rates at MP1, MP2, MP3, and MP9 were in the ranges of 0.8–13.0 mL/min, 16.0–86.0 mL/min, 0.3–6.9 mL/min, and 1.8–26.0 mL/min, respectively. The corresponding mean values were 3.2 mL/min, 40.7 mL/min, 3.1 mL/min, and 10.8 mL/min (Figure 3c and Table S3). Among the four monitoring sites, MP2 exhibited the fastest drip rate and the greatest variability. The drip rate at MP2 showed a good correspondence with precipitation, indicating its sensitivity to external rainfall [59]. Due to the thick overlying bedrock of 300–500 m above Furong Cave, there is a time lag for precipitation to infiltrate and form drips inside the cave, resulting in no significant seasonal variations in the drip rates at each drip site [54,58].

3.3. Three-Dimensional Fluorescence Spectroscopy Analysis

Three-dimensional excitation–emission matrix (3D-EEM) was utilized to obtain the fluorescence spectral information and analyze the composition and sources of dissolved organic matter (DOM) in the drip water. The three-dimensional fluorescence spectra of the drip water samples from four monitoring sites (MP1, MP2, MP3, MP9) were obtained (Figure 2). The identification of the fluorescence components was based on the peak values corresponding to the excitation wavelength (EX) and emission wavelength (EM) [67], revealing six main peaks, namely A, B, C, T, Ti, and M (Figure 2) [68,69,70]. Peak A (EX: 260 nm/EM: 380–460 nm) belongs to the ultraviolet region of humic acids [71], representing refractory DOM with a predominantly terrestrial origin [72]. Peaks B and T are classified as protein-like fluorescence peaks associated with microbial processes, which can originate from exogenous sources or microbial activities [73]. Peak B (EX: 225~230 nm/EM: ~300 nm) represents low-excitation wavelength fulvic acid-like substances, indicative of tyrosine-like materials [74,75,76]. Peak C (EX: ~380 nm/EM: ~480 nm) corresponds to heteroaromatic structures of humic substances derived from biomass containing two or more microbial or aromatic rings [74,77]. The two peaks T and Ti (T: EX: ~230 nm/EM: ~280 nm and Ti: EX: ~280 nm/EM: ~340 nm) represent tryptophan-like compounds, which are endogenous organic matter associated with microbial activity [71,78]. Previous studies have found that the fluorescence signal at the peak Ti can indicate biologically available natural organic matter [73]. The peak M (EX: 330–350 nm/EM: 380–420 nm) belongs to microbial sources or humic acid-like components in the marine environment [79].

3.4. Fluorescence Spectral Parameters of Drip Water

Various fluorescence parameters can be derived from the fluorescence spectra, such as the fluorescence intensity, fluorescence index (FI), biological index (BIX), humification index (HIX), and others. These parameters are calculated based on the fluorescence intensity at specific excitation and emission wavelengths and can be used to assess the origin, humification degree, and relative content of the DOM components in cave drip water samples. The fluorescence parameters indicate a mixture of endogenous newly formed organic matter and exogenous terrestrial organic matter in the drip water DOM of Furong Cave. The definitions of the three-dimensional fluorescence spectral parameters can be found in Table S2. The FI values in the drip water of Furong Cave ranged from 0.99 to 3.78, with an average value of 1.82, indicating the combined influence of endogenous and terrestrial sources (Table S4). At the monitoring sites MP1, MP2, MP3, and MP9, the FI values ranged from 1.49 to 2.42, 1.14 to 2.40, 1.44 to 2.28, and 0.99 to 3.78, respectively, with average values of 1.80, 1.76, 1.80, and 1.90 (Table S4). Among the four monitoring sites, the mean FI values of MP1, MP2, and MP3 (1.80, 1.76, and 1.80, respectively) were slightly lower than the average FI value of 1.90 at MP9. This suggests that MP9 has a relatively higher proportion of endogenous DOM components compared to the other monitoring sites.

The BIX values in the drip water of Furong Cave range from 0.47 to 7.58, with a mean value of 1.46. The majority of the BIX values were distributed above 0.8, indicating a predominant contribution of endogenous newly formed organic matter to the drip water DOM. Specifically, at the monitoring sites MP1, MP2, MP3, and MP9, the BIX values ranged from 0.82 to 4.85, 0.59 to 5.20, 0.49 to 2.25, and 0.47 to 7.58, respectively, with mean values of 1.45, 1.33, 1.06, and 2.00. These results suggest that the drip water DOM is primarily influenced by endogenous newly formed organic matter (Table S4 and Figure 3i).

The HIX serves as an indicator of the humification level of DOM in the drip water [65,80]. A higher HIX value indicates a higher degree of humification and stability of DOM in the drip water [80]. The HIX values in the drip water of Furong Cave ranged from 0.06 to 0.94, with an average value of 0.59, indicating a weak degree of humification predominantly originating from endogenous organic matter source (Table S4 and Figure 3h). Furthermore, there is a significant negative correlation between the BIX and HIX indices in all four drip water samples (R² = 0.68; p < 0.01 for MP1, R² = 0.65; p < 0.01 for MP2, R² = 0.33; p < 0.01 for MP3, R² = 0.72; p < 0.01 for MP9) (Figure 4).

3.5. The Fluorescence Region Integration (FRI) Analysis

The proportions (%) (P_i,n) of specific structural organic compounds in five regions at the four monitored drip water sites were determined by conducting Fluorescence Region Integration (FRI) analysis (Table S5). Based on the classification criteria provided in Table S1, the substances represented by the five integration regions were determined. Region I corresponds to tryptophan-like substances, while Region II represents tyrosine-like substances. Both Region I and II exist in the form of free molecules and belong to aromatic proteinaceous compounds [81], Region III corresponds to fulvic acid substances, Region IV represents soluble microbial metabolites, and Region V corresponds to humic acid-like substances (Figure 5).

In the drip water samples from MP1, MP2, MP3, and MP9, the proportion of substances in Region I range from 9.22% to 43.03%, Region II ranges from 13.81% to 57.40%, Region III ranges from 1.50% to 25.68%, Region IV ranges from 10.31% to 33.83%, and Region V ranges from 0.68% to 34.20% (Table S5). The proportions of each substance in the drip water samples at each monitoring site have been presented in Figure 5 and Table S5 in detail. The percentage distribution of P_i,n for the various substances in the drip water samples from the four monitoring sites (MP1, MP2, MP3, and MP9) of Furong Cave shows a relatively high proportion of substances in Region II, ranging from 38% to 42%, indicating a higher concentration of tyrosine-like compounds (Figure 5). The proportions of substances in Region III and Region V are relatively low, suggesting a lower content of fulvic acid-like substances in the ultraviolet and visible light regions (Figure 5).

4. Discussion

4.1. Sources and Variations of Dissolved Organic Matter (DOM)

Previous studies have indicated that an FI value of ~1.4 represents terrigenous organic matter, while an FI value of ~1.9 indicates microbial-derived organic matter [82]. During the dry season (November to April of the following year), the mean FI values at the four drip water monitoring sites ranged from 1.66 to 1.78. The mean FI values for MP1, MP2, MP3, and MP9 were 1.68, 1.66, 1.78, and 1.77, respectively (Table S6), indicating a terrigenous organic matter source. In contrast, during the rainy season (May to October), the mean FI values at the four drip water monitoring sites ranged from 1.81 to 1.96. The mean FI values for MP1, MP2, MP3, and MP9 were 1.86, 1.82, 1.81, and 1.96, respectively (Table S6), indicating a predominance of microbial-derived organic matter. The concentration of terrigenous organic matter in cave drip water increases during the dry season, while microbial activity decreases.

In the Furong Cave drip water, the presence of a mixture of endogenous and exogenous organic matter is the main characteristic, with a predominance of endogenous small molecular substances. The endogenic and newly formed autochthonous sources are relatively high, while the humification degree (HIX) is relatively low. Overall, the average HIX values for MP1, MP2, MP3, and MP9 during the dry season were 0.65, 0.66, 0.65, and 0.50, respectively (Table S6), while during the rainy season, they were 0.60, 0.62, 0.64, and 0.47, respectively (Table S6). These HIX values indicate an endogenic nature, with a low humification degree of dissolved organic matter (DOM) due to dilution effects from rainfall [3]. During the dry season, microbial activity tends to decompose small organic molecules in DOM, leading to an accumulation of larger organic molecules [42,83,84], resulting in a similar level of humification degree between the dry and rainy seasons for DOM.

The average BIX values for MP1, MP2, MP3, and MP9 during the dry season were 1.12, 1.04, 0.93, and 1.18, respectively (Table S6). During the rainy season, the average BIX values for these monitoring sites were 1.63, 1.48, 1.12, and 2.16, respectively (Table S6). The BIX values for the DOM in the drip water at all four monitoring sites were consistently higher than 0.8 (dry season: 1.07; rainy season: 1.58) (Table S6), indicating a relatively high endogenic nature [42]. The average BIX value for the four monitoring sites in Furong Cave was 1.46 (Table S4), corresponding to BIX values (>1) associated with the input of newly formed autochthonous DOM into the water [64,65]. Similar to the average BIX value (1.01) observed in Daxiao Cave, indicating that the dominant source of dissolved organic matter (DOM) in cave drip water is endogenous organic matter [45], the fluorescence parameter characteristics in the drip water of Daxiao Cave are similar to those of Heshang Cave (HIX = 0.73; BIX = 1.24) [22]. The HIX values of the DOM at the four monitoring sites were higher in summer compared to other months (Figure 3h), indicating a higher degree of humification for the DOM during the period June to August. The HIX and BIX indices of the Furong Cave drip water exhibited a range of variation (HIX: 0.06–0.94; BIX: 0.47–7.58), and they showed an inverse correlation (Figure 4), indicating that the organic matter transported from the overlying soil to the cave drip water undergoes microbial processes within the vadose zone.

The primary sources of organic matter in drip water in karst regions are attributed to two pathways: (1) the direct transport of soil organic matter (SOM) by moisture from the overlying soil, and (2) the aggregation of previously deposited organic matter in the aquifer at the drip water outlets. In the case of Furong Cave, the fluorescence parameters in the drip water predominantly originate from endogenous sources, particularly during the dry season [85]. As precipitation decreases, the microbial activity in the overlying soil correspondingly increases, leading to an elevated production of DOM in the soil, which serves as a source for DOM in drip water [45]. Additionally, the increased residence time of the infiltrating water in the overlying aquifer allows for the transportation of organic matter generated through microbial decomposition processes. The variation ranges of the HIX and BIX indices in the Furong Cave drip water (HIX: 0.06–0.94; BIX: 0.47–7.58) and their significant negative correlation (Figure 4) clearly indicate that the DOM primarily originates from endogenous and microbial sources (with average BIX values consistently >0.8). Therefore, it can be inferred that the organic matter in the cave drip water undergoes microbial processes during its transportation from the overlying soil system to the cave drip sites. Consequently, the DOM in the drip water is influenced by the residence time in the aquifer, with the primary source being the overlying soil system.

4.2. The Changes in DOM and Its Relationship with Precipitation

Assuming that the production rate of DOM in the overlying soil is consistent with the precipitation amount and intensity, the primary controlling factor influencing the concentration changes in the individual drip water DOM is the processes occurring in the overlying aquifer [47,86]. Factors such as rock thickness, the development of fractures, and permeation pathways affect the retention time, mixing degree, response rate to the external environment, and microbial mineralization degree of karst water, ultimately determining the variations in the characteristics and concentration of DOM in each drip water site [47,78,87]. The seasonal variation in the drip rate observed in Furong Cave (Figure 3c) is attributed to the rock layers spanning a depth range of 300 to 500 m above the cave. Notably, these variations exhibit distinct differences from the pronounced seasonal patterns observed in surface precipitation. Furthermore, the seasonal characteristics of the drip rate are not apparent. The response of the drip rate to fluorescence intensity exhibits a lag effect (Figure 3c and Figure 6), attributed to the long transport pathways and complex mixing effects experienced by atmospheric precipitation as it passes through the thick overlying rock layers of Furong Cave [54,88]. Furthermore, there can be significant variations in the drip rate magnitude among different drip water sites, with differences of up to tenfold (Figure 3c), primarily due to variations in the thickness of the underlying bedrock and the pathways and channels through which the karst water flows [59]. These factors may ultimately result in a lack of a clear correlation between the DOM concentration in the drip water and monthly precipitation (Figure 7). For example, in this study, the average drip rates at the four monitoring sites were MP2 > MP9 > MP1 > MP3 (Table S3), while the average fluorescence intensities were MP3 > MP1 > MP2 > MP9 (Table S3). The longer residence time of the DOM in the overlying aquifer at MP2, influenced by the mixing of “old” and “new” water in the aquifer, obscures the seasonal signals and leads to a lower concentration of DOM in its drip water.

The variation in drip rate in response to precipitation is more sensitive to changes in drip water pressure compared to the response of DOM concentration to precipitation [71,86]. During the rainy season, when precipitation is abundant, the increased water pressure leads to a faster drip rate (Figure 3a,c). In the dry season, when precipitation is scarce, the drip water supply at each drip water site primarily relies on the retained water in the bedrock, resulting in a relatively constant drip rate (Figure 3a,c). Among the four monitoring sites, MP2 exhibits the largest range of drip rate variations (16 to 82 mL/min) and the highest mean value (41.8 mL/min) (Figure 3c and Table S3), indicating the most pronounced influence from external precipitation processes. During rainfall events, as the amount of rainwater increases, it washes over the soil and bedrock, eventually infiltrating into the unsaturated zone of the karst aquifer. Within this zone, the water is subjected to comprehensive regulation through various processes, including soil leaching, microbial activity, and rock adsorption. Based on limited monitoring data and the heterogeneity of karst hydrology, compared to the MP2 point, the drip rate variations and mean values at MP1, MP2, and MP9 are smaller (Figure 3c and Table S3), and there is no significant correlation between the drip rate and fluorescence intensity (Figure 6). The hydrological processes in the karst area, along with the influence of the aquifer on mineral-mediated DOM adsorption, provide favorable conditions for DOM filtration and mineral adsorption in the underlying bedrock fractures [27,78,86,87]. In other words, during the dry season, the mixing of soil moisture and retained water in the karst aquifer results in a combined seasonal signal carried by cave drip water, leading to a lower DOM content (or fluorescence intensity) in the drips (Figure 3g) [89].

In summary, the concentration variations of the DOM in cave drip water are influenced and controlled by the DOM production in the overlying soil, precipitation amount and intensity, as well as processes within the karst aquifer [47,90,91]. The mixing, lagging, and water–rock interaction (WRI) of the karst water in soil and bedrock, along with the depletion of soil moisture in the overlying area, can result the discharge of karst water and the concentrations of DOM lagging the changes in precipitation [27,85,86,87,92,93,94,95].

Previous studies conducted in the relevant field have investigated the dissolved organic matter (DOM) and chemical characteristics in water bodies. It has been found that chemical indicators such as Na, K, and SO₄²⁻ can reflect the hydrogeochemical processes and features of the aquifers in karst regions [23]. Furthermore, these chemical indicators can provide corresponding indications of mineralization processes in regional water bodies [23]. Microbial processes in karst regions are influenced by the residence time and pathway mechanisms of infiltrating water in the overlying soil and bedrock fractures [45]. Additionally, the mineralization process of karst groundwater is constrained by these factors [27]. These influences result in varying concentrations of dissolved organic matter (DOM) and corresponding 3D-EEM spectral characteristics in drip water within caves [45], thereby leading to a lack of significant correlation between the DOM, drip rate, and external climatic factors such as precipitation and temperature (Figure 6, Figure 7 and Figure 8). Through this study, a deeper understanding was gained regarding the sources, compositional structure, migration mechanisms, and controlling factors of dissolved organic matter (DOM) in cave drip water in karst regions. This study also provided a valuable reference for comparative analysis in related research. Consequently, further advancement is necessary to delve into the research on DOM in water systems within karst regions.

4.3. Temperature and DOM

The two main factors influencing the transport of DOM in karst aquifer drip water and its concentration are temperature and rainfall [27,91]. A monitoring study conducted in Brazilian caves revealed that an increase of 5.5 °C in the average winter temperature of the previous year resulted in interannual variations in the DOM concentration in drip water, indicating that the changes in DOM are influenced and controlled by climate parameter fluctuations [91]. During the monitoring period, the Heshang caves in Central China experienced significant fluctuations between dry and wet years, with substantial differences in annual rainfall variations that may mask the temperature response signal of Soil Organic Matter (SOM) decomposition [96]. At the interannual scale, the fluorescence signal of DOM in drip water shows low sensitivity to temperature response [27]. The thick bedrock overlying Furong Cave, ranging from 300 to 500 m, introduces a lagged relationship between meteorological factors and variations in organic matter concentration in drip water, leading to a noticeable lag in the fluorescence intensity response to the monthly mean temperature at each drip site. The fluorescence signals in drip water collected from various monitoring sites were observed to be insensitive to temperature changes, showing no significant correlation (Figure 8).

5. Conclusions

This study conducted a 4-year monitoring program at four drip sites within Furong Cave in Southwestern China. It utilized fluorescence spectral parameters to reflect the characteristics of the drip water DOM and employed EEM-FRI for the quantitative analysis of DOM composition. By combining meteorological and environmental data from inside and outside the cave, this study investigated the sources and the spatial–temporal variations of DOM in drip water. The DOM in the cave drip water is influenced by processes such as microbial activities in the overlying soil, organic matter productivity in the soil, and hydrological processes, including karst water recharge methods; infiltration rates; recharge amounts; and groundwater flow paths. The dominant input of the drip water DOM is a mixture of endogenous and exogenous sources, with a relatively low degree of humification. The concentration fluctuations of the dissolved organic matter (DOM) in the drip water exhibit no significant correlation with regional climatic factors, such as precipitation and temperature. This phenomenon is controlled by the substantial thickness of the bedrock overlaying the cave, which leads to a time lag in the infiltration of water through the overlying aquifer. Consequently, the organic matter signals carried by the drip water display a mixed effect, necessitating further in-depth research and an exploration of the specific influencing processes. Within the same cave, the diverse responses of the drip rate and DOM concentration at different drip sites to surface precipitation and temperature emphasize the importance of long-term cave monitoring efforts. This, in turn, facilitates the investigation of the spatial variations and primary controlling factors of DOM in drip water within a specific cave. This also indicates the uncertainty of reconstructing surface climate and karst environmental changes using a single indicator, emphasizing the necessity of comprehensive studies using multiple indicators.

The findings of this study contribute to the analysis and understanding of the response relationship between the transport of dissolved organic matter (DOM) in drip water within karst aquifers and regional climatic factors at the interannual scale. Furthermore, it highlights the need for the qualitative and quantitative analysis of the time lag between the DOM in drip water and climatic factors (precipitation and temperature) in future research. This analysis will accurately assess the impact mechanisms of regional climate change on the output processes of DOM in water systems within karst regions. Moreover, the research conducted in Furong Cave represents a localized area and serves as a single-case cave monitoring study. In future research, it is essential to continue long-term cave monitoring and conduct comparative studies with different karst water systems in various regions. These studies should explore the dynamic impact processes of climate on the DOM in karst aquifers, taking into account the geographical and environmental differences among different regions.