Impact of elevated soil temperatures on disinfection byproduct dynamics and coagulation efficiency in soil runoff as a drinking water source

Extreme thermal events, characterized by frequent, intense, and prolonged periods of high temperatures, are projected to increase due to climate change (Fischer et al. 2021). The impact of climate change on these temperature extremes has been well-documented (Min et al. 2011). As global warming progresses, the frequency and intensity of heat waves and other extreme weather events escalate, leading to higher soil temperatures. Under adverse weather conditions, soil's surface temperature can become exceptionally high as absorbed solar radiation surpasses the energy dissipated into the atmosphere (Onwuka 2018; O'Brien & Hatfield 2020). For example, a peak bare soil surface temperature of 56 °C was recorded in Iowa in 2018 (O'Brien & Hatfield 2020).

Rising temperatures stimulate soil microbial activity (Karhu et al. 2014) and increase the production of dissolved organic matter (DOM), which subsequently enters freshwater systems from soils and wetlands (Lipczynska-Kochany 2018). Microbial adaptations to higher temperatures include changes in lipid compositions to stabilize cell membranes and the expression of heat shock proteins (Jansson & Hofmockel 2019). In addition, warming accelerates soil heterotrophic microbial respiration rates, enhancing CO₂ emissions (Bond-Lamberty et al. 2018) and reducing soil organic matter (SOM), thereby potentially degrading soil fertility, which is critical for maintaining soil structure and water-holding capacity (Cavicchioli et al. 2019).

Soil also contains a natural organic matter (NOM), a complex mixture of decomposing plants and animals, which plays a pivotal role in environmental chemistry and microbiology (Huang et al. 2020). DOM, a part of NOM, serves as the primary precursor for disinfection byproducts (DBPs) in drinking water (Bolan et al. 2011), which are associated with potential health risks, including carcinogenic and mutagenic effects (Richardson et al. 2007). The quantity and composition of NOM influence the types and concentrations of DBPs produced (Li et al. 2017). Extreme weather conditions may significantly affect the generation of DBPs in natural water environments (Chang et al. 2020).

Coagulation is an important process in the treatment of water containing dissolved DOM (Edzwald & Tobiason 1999). This process effectively removes a significant proportion of DOM, particularly humic substances, primarily hydrophobic (Yu et al. 2022). Humic substances, a key component of SOM, can dissolve into water through various environmental processes (Li et al. 2020a). Furthermore, soil temperature has been identified as a significant factor influencing DOM characteristics (Zhao et al. 2023), which, in turn, can affect the speciation of DBPs during water treatment (Qadafi et al. 2021).

Previous studies have extensively explored the impact of soil temperature on DOM characteristics. For example, Zhao et al. (2023) investigated how elevated soil temperatures accelerate soil organic carbon decomposition, leading to increased production of DOM. Wang et al. (2022) investigated the influence of heat waves on the interactions between DOM and microbial communities, highlighting how extreme temperature events modify these interactions in the presence of microplastics. Additionally, Liu et al. (2019) analyzed how interactions between soil temperature and revegetation processes affect DOM dynamics, emphasizing that vegetation recovery can alter the temperature sensitivity of DOM production and composition. However, this study needs to discuss the impact of soil characteristics on DBP formation. Several studies have been conducted on the correlation between soil condition and DBP formation, including the impact of drought (Yuan et al. 2024), sub-tropical forest soils (Zhang et al. 2009), long-term nitrogen deposition (Li et al. 2020a), thermally altered soils (Cawley et al. 2017), and wildfire (Wang et al. 2015). However, information on the specific impact of soil temperature and the coagulation process on DBP formation is still lacking. This research aims to investigate the impact of short-term extreme temperatures (up to 65 °C) on soil runoff properties, including soil respiration, DOC, and DON, and their role in DBP precursor formation. It evaluates how enhanced coagulation mitigates DBPs at extreme temperatures. Additionally, the study examines the cytotoxicity potency of DBPs under extreme conditions, providing insights into the interplay between temperature, microbial activity, and DBP dynamics to inform better water treatment strategies.

For the extreme temperature experiment, amber bottles were meticulously prepared to eliminate potential sources of contamination. The bottles were rinsed with a detergent solution, followed by an acid wash to remove residual organic material. Subsequent rinsing with deionized (DI) water ensured the bottles were devoid of detergent and acid residues. Soil samples were collected from the local Xingnong Group Yumei Biotechnology Co., Ltd market brand, with active ingredients 80% of organic matter, 1.5% total nitrogen, and 1.0% total phosphate. Soil samples were then carefully loaded into the prepared bottles (360 ± 22 g or 10.9 ± 0.2 cm), and the experiment encompassed three temperature conditions: 25, 45, and 65 °C. Soil temperature is most influenced by air temperature at shallower depths, typically up to about 20 cm (approximately 8 inches) below the surface (Islam et al. 2015). Deeper layers, reaching 10 m, tend to remain relatively stable but are subject to long-term warming trends and moisture interactions due to climatic changes (García-García et al. 2023). The chosen experimental soil depth reflects these real-world conditions by representing the upper soil layers where temperature fluctuations due to air temperature are most significant.

Daily monitoring involved weighing each bottle to track changes in soil weight. Sterile DI water was added as needed to maintain consistent moisture levels. This process was sustained over a 7-day incubation period. The 7-day duration was selected to approximate real-world heat wave conditions. In Europe, heat waves with extreme temperatures between 1979 and 2016 lasted an average of 5 days (Zschenderlein et al. 2019). Conducting the experiment for 7 days provides a slightly extended timeframe to capture the potential prolonged effects of heat waves on soil respiration and associated processes. This approach ensures a more comprehensive understanding of the system's response to such events. At the end of the experiment, soil samples were collected from each bottle for the soil runoff experiment and water quality analysis. Two measurements were taken from the same bottle to ensure analytical accuracy rather than independent duplicates.

Briefly, after 7 days of incubation, 90 g of soil was directly added to 900 mL of distilled water (1:10 w/v soil to solution ratio) in a pre-acid wash 2,000 mL flask, and the mixture was shaken at 100 rpm for 1 h (Jones & Willett 2006). The soil extracts were then screened by sieving 0.275 mm to remove soil particles. A jar test was used to conduct the coagulation process. The pH of the samples was adjusted by HCl 0.1 M to pH 5–6 as the optimum pH for coagulation with high organic matter (Sun et al. 2019). The aluminum sulfate (alum) doses applied to the samples were 0, 100, 300, 500, 700, 900, 1,100, and 1,300 mg/L. The samples were mixed rapidly at a speed of 220 rpm in 1 min, and slowed down and the speed was to 60 rpm for 14 min. After that, samples were settled for 30 min. Each sample's final turbidity and pH were measured to choose the optimum coagulant dose. The use of an extreme alum dose of up to 1,300 mg/L was necessary to achieve a turbidity level below 5 NTU, as required by drinking water standards.

Before chemical characterization, DOC extraction was collected through vacuum filtration using Pall Supor^®-450 membranes (pore size 0.45 μm). Six grams of soil samples were taken for each soil sample to measure the moisture (24 h in a 105 °C oven). The moisture of the sample was used to measure the dry weight of the soil. Dissolved organic carbon (DOC) was measured using 1030 Aurora Total Organic Carbon Analyzer, OI Analytical, Texas, method US EPA 413.5. Dissolved organic nitrogen (DON) was measured by subtracting total dissolved nitrogen (TDN) from ammonia (NH₄-N), nitrate (NO₃-N), and nitrite (NO₂-N).

TDN analysis using persulfate oxidation followed by Hagedorn & Schleppi (2000). Nitrate and nitrite were measured by ion chromatography (IC) (DIONEX ICS-5000, Thermo Scientific, USA). Ammonia was measured by IC (DIONEX AQUION, Thermo Scientific, USA). UVA was measured using a UV–vis spectrophotometer (UV-1280, Shimadzu, Japan) at wavelength of 254 nm. Specific UV absorbance (SUVA) (L/mg.m) was calculated as a ratio of the UV absorbance at 254 nm (au/mL) with DOC (mg/L).

This study examined the generation of DBPs that often occur during the process of chlorination. The uniform formation condition test was performed to mimic the chlorination of drinking water (Summers et al. 1996). Briefly, all samples were diluted to a DOC concentration of 1 mg/L, buffered by pH 8 borate buffer, and chlorinated with freshly prepared NaOCl/pH 6.7 borate buffer solution (pH 8.0) at 25 °C in the dark for 24 h without headspace. Following the process, residual chlorine was quenched with NH₄Cl solution, and DBPs were extracted and measured using a GC-ECD (Agilent 6890N) under strict adherence to EPA Methods 551.1 and 552.3. Twenty DBPs were measured: four trihalomethanes (THMs; including trichloromethane (TCM), tribromomethane (TBM), bromodichloromethane (BDCM), and dibromo-chloromethane (DBCM)), nine HAAs (including chloroacetic acid (MCAA), bromoacetic acid (MBAA), dichloroacetic acid (DCAA), trichloroacetic acid (TCAA), bromochloroacetic acid (BCAA), dibromoacetic acid (DBAA), bromodichloroacetic acid (BDCAA), chlorodibromoacetic acid (CDBAA), and tribromoacetic acid (TBAA)), four HANs including dichloroacetonitrile (DCAN), trichloroacetonitrile (TCAN), bromochloroacetonitrile (BCAN), dibromoacetonitrile (DBAN)), two haloketones (HKs; including 1,1-di- chloro-2-propanone (DCP), 1,1,1-trichloro-2-propanone (TCP)), and chloropicrin or trichloronitromethane (TCNM).

Cytotoxicity assays are laboratory tests used to evaluate the toxic effects of substances, such as DBPs, on living cells. These assays typically involve mammalian cells, such as mouse or human cells cultured in vitro or alternative models like invertebrates (e.g., Artemia salina, Daphnia magna, and Tetrahymena pyriformis) (Solis et al. 1993; Altomare et al. 2012). In the case of DBP research, cytotoxicity is frequently assessed using a single Chinese hamster ovary (CHO) cell line, a well-established system in toxicological studies. The CHO cell assay evaluates the reduction in cell density as a function of DBP concentration over a 72-h period (Plewa et al. 2002).

In this study, direct cytotoxicity assays were not performed. Instead, cytotoxicity potency was calculated using existing reference data, where molar concentrations of each DBP were combined with their respective LC50 values (the concentration lethal to 50% of cells), as reported by Wagner & Plewa (2017).

Table 1 summarizes the LC50 values (M) of DBPs tested in this research, based on data reported by Wagner & Plewa (2017). The table also includes each compound's relative molar mass (Mr), which was used to convert concentrations from mg/L to molar units (M). However, LC50 values for certain HK, such as DCP and TCP, are currently unavailable in the literature.

Table 1

A summary of DBPs was analyzed using the CHO chronic cytotoxicity assay (Wagner & Plewa 2017)

Analyte .	LC50 standard (M) .	Relative molar mass (Mr) (dimensionless) .
Halomethane
TCM		119.38
BDCM		163.80
DBCM		208.28
TBM		253.74
Haloacetic acid
MCAA		94.50
MBAA		138.95
DCAA		128.94
TCAA		163.38
BCAA		173.39
BDCAA		207.83
DBAA		217.84
CDBAA		252.29
TBAA		296.74
Haloacetonitrile
TCAN		144.38
DCAN		109.94
BCAN		154.39
DBAN		198.84
TCNM		164.38
Haloketone
DCP
TCP

Analyte .	LC50 standard (M) .	Relative molar mass (Mr) (dimensionless) .
Halomethane
TCM		119.38
BDCM		163.80
DBCM		208.28
TBM		253.74
Haloacetic acid
MCAA		94.50
MBAA		138.95
DCAA		128.94
TCAA		163.38
BCAA		173.39
BDCAA		207.83
DBAA		217.84
CDBAA		252.29
TBAA		296.74
Haloacetonitrile
TCAN		144.38
DCAN		109.94
BCAN		154.39
DBAN		198.84
TCNM		164.38
Haloketone
DCP
TCP

View Large

The cytotoxicity potency is widely used in the environmental field and allows for the estimation of cytotoxicity based on known experimental data without conducting new in vitro assays (Chuang & Mitch 2017; Liu et al. 2018; Kozari et al. 2024). By adopting this approach, the study integrates existing toxicological data to assess the potential health impacts of DBPs practically and reliably.

DOM characterization was measured by measuring EEM fluorescence followed by parallel factor analysis (PARAFAC). Soil runoff was diluted into DOC value 1 mg/L using 0.01 mol/L KCl and acidified to pH 3 with HCl to minimize the complexation of metals with DOC (Drewes et al. 2006). The samples were measured in a 1 cm cuvette using a fluorescence spectrophotometer (F7100, Hitachi, Japan). The scanned areas for excitation were 220–400 nm and 290–550 nm for emission. The scanning interval for excitation was 5 nm and for emission, 1 nm, with the bandwidths set to 5 nm for both excitation and emission. Raman calibration is also performed by normalizing the area under the Raman peak (Ex = 350 nm, Em = 381–426 nm) (Xu et al. 2021). The sweep speed was 1,200 nm/min. Data analysis following PARAFAC by Matlab software and DOMFluor toolbox according to previous studies (Chen et al. 2003).

The highest CO₂ production rate at 65 °C can be explained by the stimulation of thermophilic microorganisms and enhanced enzymatic activity, which accelerate the decomposition of organic matter at elevated temperatures. These results are in line with previous research that mentioned maximum extracellular enzyme activity in soils was observed at temperatures in the thermophilic microorganism (55–75 °C) (Gonzalez et al. 2023). However, prolonged exposure to such extreme conditions may also lead to microbial stress or community shifts that could affect activity. The observed increase in the CO₂ production rate from day 0 to day 7 indicates cumulative microbial activity and organic matter mineralization over time (Bradford 2013). This trend highlights the temperature-dependent dynamics of soil microbial communities and their role in carbon cycling under heat wave conditions.

The effect of temperature on microbial activity has been a topic of interest in various scientific studies. Soil microorganisms decompose SOM and release CO₂ through metabolism. Soil microbial communities can give positive climate feedback by releasing more greenhouse gas (CO₂) that can cause the atmosphere temperature to increase (Cavicchioli et al. 2019). This study finds that increasing atmospheric temperature can increase CO₂ gas production in soil, which aligns with several previous studies (Pietikäinen et al. 2005; Chow et al. 2006; Tang et al. 2018). However, the highest temperature used in the previous research is 45 °C. Most of the research on soil respiration often focuses on more moderate temperature ranges relevant to the ecosystems, even though the current situation shows that some countries in Asia, such as China and South Korea, are experiencing heat waves with temperatures above 50 °C (García-García et al. 2023).

The results highlight the interactive effects of temperature on CO₂ gas production in soil. The CO₂ gas produced in this research is predicted to be biotic combined with an abiotic process. Settling up control experiments of soil without living organisms can help establish a baseline for abiotic CO₂ (Tang et al. 2018). Such controls could help separate microbial contributions from purely chemical or physical CO₂ emissions. However, this study did not conduct control experiments specifically designed to differentiate between these processes. Instead, CO₂ measurements were taken immediately after the soil was placed in the bottle to determine the initial CO₂ concentration, which could originate from the blank space within the bottle. These measurements showed non-detectable CO₂ levels, suggesting that initial contributions from abiotic sources, such as trapped gases in the headspace, were minimal. Moreover, one method to establish a control experiment involves sterilizing with an autoclave and dividing the soil sample into two portions. One portion remains sterilized to represent abiotic conditions, while the other is supplemented with an inoculum, comprising non-sterilized soil to reintroduce microbial communities from the original soil (Fanin & Bertrand 2016). This approach allows for a more apparent distinction between biotic and abiotic contributions. However, sterilization using an autoclave is unsuitable for this experiment since the autoclave will change the DOM composition. Autoclave sterilization modifies DOM composition by chemically degrading compounds and altering molecular structures through processes such as hydrolysis and denaturation of various substances and colloids. (Andersson et al. 2018) (Dill & Shortle 1991, Druart & De Wulf 1993), and the result of the DBP precursor will not portray the actual condition. In this experiment, CO₂ gas production was measured directly from unaltered soil samples under controlled temperature conditions (25, 45, and 65 °C) over a seven-day incubation period. This approach preserved the DOM composition and microbial communities, ensuring that the results accurately reflect natural conditions. Maintaining the soil in its original state allowed observation of the combined effects of biotic and abiotic processes on CO₂ production and DBP precursor generation, providing insights relevant to real-world environmental impacts.

One of the most important contributions of the soil microbiota is the decomposition of organic matter and an increased warming effect that promotes a higher organic matter decomposition rate (Bradford et al. 2016; Aamir et al. 2019). However, several factors influence microbial respiration, such as carbon use efficiency (CUE), microbial biomass carbon (MBC), enzyme activity, and microbial community (You et al. 2019). The mechanisms for the extreme temperature effect on soil respiration by the abovementioned factors have been studied.

CUE reflects the efficiency of microorganisms in converting carbon into biomass rather than CO₂, with higher CUE indicating reduced respiration rates and enhanced carbon retention. Environmental factors such as temperature and nutrient availability significantly influence CUE; optimal conditions boost it, while stressors like extreme temperatures lower it, increasing CO₂ release as microbes rely on stored biomass for energy (Auffret et al. 2016; Dacal et al. 2019; Anjileli et al. 2021). Elevated temperatures, in particular, divert carbon from biomass formation to respiration, reducing CUE and exacerbating soil carbon losses during heat events.

MBC represents the total carbon in microbial cells and indicates microbial activity and health. Higher MBC often correlates with increased respiration due to greater biomass and metabolic activity, though this depends on microbial community composition and metabolic rates of specific groups (Auffret et al. 2016; Zhao et al. 2016). Extreme heat can initially boost MBC through higher growth rates but may later reduce it due to stress or mortality in sensitive microbes. This shift affects respiration rates and contributes to soil carbon loss (Auffret et al. 2016; Nyberg & Hovenden 2020).

Microbial enzymes are key to breaking down organic matter and driving respiration. Enzyme activity increases with substrate availability, boosting decomposition and CO₂ emissions. Factors like moisture and temperature significantly influence this activity, for example, higher moisture levels can enhance enzyme production and microbial respiration (Auffret et al. 2016; Dacal et al. 2019). Enzymatic activity drives SOM decomposition and influences soil respiration. Extreme temperatures can boost enzyme activity up to a threshold, after which denaturation or reduced microbial viability may lower efficiency. Higher temperatures generally accelerate SOM breakdown and CO₂ release, though this depends on enzyme types and substrate availability (Anjileli et al. 2021).

Microbial community composition influences metabolic capabilities and responses to environmental changes. Species vary in substrate use and efficiency, with shifts from competition to facilitation altering respiration dynamics. Changes in dominant species traits can increase or decrease respiration rates, while diverse communities may utilize a broader substrate range, enhancing ecosystem respiration (Auffret et al. 2016; García et al. 2023). Microbial community composition influences responses to temperature changes. Heat stress can shift communities toward thermophilic species that efficiently decompose organic matter, enhancing soil respiration. Diverse communities may better withstand temperature fluctuations, stabilizing respiration rates compared to less diverse, heat-sensitive groups. (Auffret et al. 2016; Dacal et al. 2019).

The result is aligned with previous research that DOC in the soil extracts increased with the increasing incubation temperature (Martínez et al. 2003). The increasing DOC concentration is due to the new carbon produced mainly from microbe growth alongside decomposition processes (Hu et al. 2024). Thermophilic microorganisms grow optimally at high temperatures because their protein structures and cell membranes are specifically adapted to withstand such environments. These microorganisms initiate the catabolism of organic materials by producing specific enzymes that work more efficiently at elevated temperatures. Hydrolytic enzymes, such as cellulase, lipase, and protease, play key roles in breaking down complex organic materials. Cellulase breaks down cellulose into glucose, lipase breaks down lipids into glycerol and fatty acids, and protease breaks down proteins into amino acids (Blumer-Schuette 2020). At high temperatures, around 65 °C, hydrolysis reactions occur more rapidly due to the increased kinetic energy of substrate molecules and enhanced enzyme efficiency. Hydrolysis products, such as glucose or fatty acids, are used by microorganisms as energy sources. In aerobic respiration, glucose is metabolized through glycolysis, the electron transport chain, and the Krebs cycle, producing CO₂ as a byproduct. In anaerobic respiration, glycolysis is followed by fermentation or reducing other compounds, excluding O₂, to produce energy for the microorganisms (Blumer-Schuette 2020). These processes highlight how thermophilic microorganisms adapt to high temperatures to support their efficient metabolic activities. In addition, higher temperatures increased the relative concentration of high molecular weight soluble organics (Martínez et al. 2003). Higher temperatures can accelerate polymerization reactions, where small molecules (monomers) combine to form larger, more complex molecules. The additional energy from heat increases the activity of the molecules, leading to more frequent collisions and a higher likelihood of bonding. This process facilitates the formation of molecules with higher molecular weights (Li et al. 2020b).

The DON analysis in Figure 2(a) also demonstrated distinct differences in organic nitrogen content among the temperature treatments. Increasing temperature leads to increasing DON. The concentration of DON in soil runoff at 25, 45, and 65 °C is 0.62, 0.84, and 1.81 mg N/g soil, respectively. The DON decreases after coagulation to 0.15, 0.36, and 0.69 mg N/g soil for 25, 45, and 65 °C, respectively.

These results are similar to those reported by Dai et al. (2020), which show increasing DON and NH4 + concentration by the elevated temperature. In the previous research (Øygarden et al. 2014), increasing concentration of nitrogen in soil runoff has been associated with extreme events. Increasing temperature in the soil increases the nitrogen mineralization rate through the increase in microbial activity, leading to enhanced N₂O emission to the atmosphere (Lu & Xu 2014; Dai et al. 2020).

Elevated temperatures significantly increase DON concentrations in soils by several possible mechanisms in the nitrogen cycle explained by Dai et al. (2020). Higher temperatures enhance the microbial decomposition of SOM, leading to greater release of DON into the soil solution. Additionally, warming reduces microbial nitrogen immobilization, causing less nitrogen to be stored in microbial biomass and leaving more nitrogen, including DON, in the dissolved phase. This reduction in nitrogen immobilization reflects the increased maintenance energy demands of microbial communities under warming conditions. Moreover, elevated temperatures accelerate nitrogen turnover, including mineralization and nitrification (Davidson & Janssens 2006; Dai et al. 2020), further contributing to nitrogen accumulation in dissolved forms. These shifts highlight how temperature-driven changes in microbial activity and nitrogen cycling dynamics can lead to increased DON concentrations, reflecting the broader impacts of climate warming on soil ecosystems. Although the enzyme deactivation may occur at temperatures more than 40 °C that will be applied to non-thermophilic microbes, the higher DON at 65 °C could be contributed by soil thermophilic microbes (Dai et al. 2020; Wang et al. 2023). Moreover, the thermal response is more sensitive to warming at higher nutrients (Hu et al. 2024).

Figure 2(b) revealed the slightly increasing SUVA₂₅₄ value in the elevated temperature in the soil runoff. This finding is aligned with previous research (Weishaar et al. 2003; O'Donnell et al. 2016) which indicates that the decomposition in higher temperatures generates greater aromaticity of the DOM fractions, leading to greater SUVA₂₅₄ values. Zhao et al. (2024) suggest that high-temperature soil can modify the microbial community, allowing only heat-tolerant microorganisms to perform the decomposition process. In addition, those microbes specifically break down plant-derived aromatic components and accumulate other components (Han et al. 2022). In the present study, the coagulation process reduced the aromaticity in DOM, indicated by a lower SUVA254 value. After coagulation treatment, we obtained SUVA254 values of 3.08, 2.79, and 2.88 L/mg m for 25, 45, and 65 °C, respectively. SUVA254 values of more than 4 L/mg m imply an abundance of humic content with large molecular weight and high hydrophobicity (Abeynayaka 2012). However, SUVA less than 2 L/mg m indicates a low humic component with low molecular weight, which will affect the coagulation performance with alum due to interrupted surface properties and degree of crystalization of the primary nanoparticles (Abeynayaka 2012; Zhang et al. 2021).

The EEM spectra indicated increases in regions III and V for samples at 65 °C compared to those at 25 °C, suggesting a higher presence of FA and HA precursors. This increase signifies the leaching of aromatic carbons, which contributes to the elevated specific ultraviolet absorbance (SUVA) values observed in Figure 2. These findings align with previous studies that high SUVA, C-rich NOM produced high yields of C-DBPs (Hua et al. 2020) emphasize the significance of molecular weight and chemical functional properties in understanding DBP formation from DOM precursors (Hua et al. 2017; Hua et al. 2018; Hua et al. 2020).

After 7 days of incubation, the enhanced fluorescence intensity in these regions indicates a greater production of FA- and HA-like substances. Following chlorination, these organic matter types were identified as primary contributors to DBP formation (Chang et al. 2020). However, a study using intracellular algogenic organic matter (IOM) demonstrates that while small molecular weight fractions of IOM, HA- and FA-like fluorescence, preferentially tend to form THMs, large fractions of IOM, aromatic protein- and soluble microbial product-like fluorescence, are favorable for the formation of HAA (Hua et al. 2019). Increasing HA (region V) in higher temperatures leads to increasing HANs formation but not for TCNM as another N-DBP because HA is also found as a major precursor of HANs but not a significant precursor for TCNM (Zhang et al. 2020; Kozari et al. 2024). TCNM precursors are primarily micromolecular, polar, and non-positively charged organics (Zhang et al. 2020). Phenolic compounds, especially those with meta-substituents and electron-withdrawing groups, facilitate TCNM precursor formation during UV/monochloramine treatment (Zhang et al. 2022). Ammonia can be a nitrogen source for TCNM during UV/chlorine processes, with monochloramine photolysis producing reactive nitrogen species that contribute to TCNM formation (Zhou et al. 2020).

Unexpectedly, the intensity of peaks after coagulation is higher than before coagulation. This result is in contrast with previous research that showed coagulation can reduce 27% of humic- and fulvic-like compounds based on the EEM-PARAFAC (Oloibiri et al. 2017). The possible reason is that the coagulation processes, often used in water treatment, can alter soil DOM's physical and chemical properties. During coagulation, particulate organic matter can aggregate and settle, which may release previously bound or adsorbed DOM into the solution. This release is particularly evident when coagulation agents like aluminum sulfate are used, as they can disrupt existing soil aggregates and mobilize organic compounds (Andersson et al. 2024).

This temperature-induced increase in DBP concentrations raises concerns about the potential risk to human health as the drinking water quality might have a significant impact. The findings suggest that soil surface temperature plays a crucial role in shaping the dynamics of DBP formation and release into the runoff, emphasizing the need for careful consideration of temperature variations in water resource management. It is necessary to conduct more studies to clarify the underlying mechanisms causing this temperature-sensitive behavior and to develop mitigation techniques for the effects of the temperature of the soil surface and chlorination soil runoff.

As alum concentration increased, there was a consistent decrease in pH for all temperature conditions. Adjustment of pH in coagulation is important. A low pH makes the humus in the water a negatively charged HA colloid that is easily removed; a high pH causes the humus to become a humate, which decreases its water repellence and increases its hydrophilicity, thereby reducing the removal rates (Sun et al. 2019).

Notably, to achieve a turbidity level within the optimal range of 2.5–4.0 NTU, the soil runoff samples after incubation at 25 and 45 °C required an alum concentration of 500 mg/L. In contrast, the 65 °C sample necessitated a higher alum concentration of 900 mg/L to achieve similar turbidity levels. The increased alum demand at higher temperatures, particularly at 65 °C, suggests that the coagulation process becomes more challenging under warmer conditions. This phenomenon may be attributed to changes in the composition and characteristics of the soil runoff, impacting the efficiency of alum in particle removal.

Figure 6 shows the effect of coagulation on potential DBP formation and speciation, including THM, HAA, HAN, HK speciation, and TCNM.

Figure 6(a) reveals that there has been a slight fluctuation of THM/DOC in various temperatures. The THM speciation slightly increases at 45 °C but decreases at 65 °C compared to 25 °C. The increase in THM speciation at 45 °C suggests that elevated temperatures may enhance the transformation of organic precursors into THMs. The increase could be attributed to accelerated chemical reactions under higher temperature conditions. Conversely, the decrease in THM speciation at 65 °C implies a different mechanism of THM precursor to increasing temperature. The possible reason is that aromatic-like molecules, as precursors of THM, are more favorable for decomposition as temperature warms and could thus be readily decomposed by microbes (William & Plante 2018). The aromatic-like molecules, as warm-depleting molecules, were more thermodynamically favorable for oxidation (Hu et al. 2024). This finding indicates that extreme temperature can decrease THM. The result aligned with Wang et al. (2013), which shows decreasing THM concentration between unburned and burned detritus up to 63.6%. Understanding these temperature-specific effects is crucial for predicting and managing THM levels in water treatment processes.

The coagulation process can reduce potential THM formation by around 41–47%. The result is comparable with a previous study that mentioned that alum coagulation can reduce potential THM formation by 42% (Wang et al. 2021). This underscores the importance of enhanced alum coagulation as an effective treatment strategy for mitigating the formation of THMs in water.

Figure 6(b) indicates a substantial rise in HAA/DOC ratios as the temperature increases. After 7 days of incubation, the HAA/DOC for 65 °C is increased by 20–27% compared to 25 °C. This observation suggests a strong correlation between higher temperatures and the enhanced formation of HAAs in the presence of DOC. The substantial increase in HAA speciation at higher temperatures, particularly at 65 °C, suggests that the temperature plays a pivotal role in forming HAAs. The increasing temperature might increase the readily oxidizable functional groups, i.e. carbon–carbon double bond, alcohol, and amine, as a precursors of HAA, especially DCAA and TCAA (Hua et al. 2019). A previous study demonstrated that the phenolic hydroxyl group is more likely to produce TCAA during chlorination, whereas the carboxyl and alcoholic hydroxyl groups tend to form DCAA (Li et al. 2011). Although high aromatic molecules such as humic species tend to be a primary precursor of THMs and HAAs (Bond et al. 2012), the formation of D-HAA5 is more strongly associated with hydrophilic DOM (Hua & Reckhow 2005; Hong et al. 2013) that may be increased at higher temperatures. The increase in hydrophilic DOM at higher temperatures was inferred from the EEM spectra (Figure 3), where Region III (FA-like, hydrophilic fractions of humic substances resulting from organic matter decomposition) showed greater intensity compared to samples at lower temperatures. This suggests that elevated temperatures enhance the production of hydrophilic DOM fractions, possibly due to microbial activity and enhanced breakdown of complex organic matter into smaller, more soluble components. Previous studies have shown that hydrophobic DOM, as THM precursors, tends to decrease under extreme temperatures (60–90 °C) (Huang et al. 2024; Yuan et al. 2024), aligning with our findings of increased HAA formation and a slight decrease in THM formation at higher temperatures. These trends suggest temperature-driven shifts in DOM composition, favoring hydrophilic over hydrophobic fractions.

The dominant HAAs in the speciation profile include DCAA and TCAA, with smaller percentages of BCAA, and BDCAA. The prevalence of DCAA and TCAA as dominant species in the HAA speciation profile aligns with known patterns in water treatment. Understanding the distribution of specific HAAs is crucial for targeted treatment strategies, as different HAAs have varying health implications and regulatory limits. Coagulation by the alum process can reduce potential HAA formation by around 47–50%. The result is comparable with a previous study that demonstrated decreasing of potential HAA formation with alum coagulation by 41% (Wang et al. 2021).

Figure 6(c), examining the HAN speciation at different temperatures, reveals specific trends: At 45 °C, there is a slight increase in HAN speciation compared to the baseline at 25 °C. At 65 °C, a sharp and notable increase (36–39%) in HAN speciation compared to 25 °C. HAN formation increased significantly due to elevated DON production from enhanced microbial decomposition at higher soil temperatures (Dai et al. 2020). Since DON is a primary precursor for HANs (Lee et al. 2007), this explains its increase.

In contrast, THMs and HAAs are predominantly derived from carbon-rich precursors such as humic and fulvic acids. While higher temperatures also increased DOC production (a precursor for THMs and HAAs), the pathways for these compounds are less directly tied to nitrogen cycling. Furthermore, THMs and HAAs require different precursor characteristics, and the formation efficiencies can vary depending on the balance of hydrophilic versus hydrophobic DOM fractions. Additionally, while HAN formation increased, its relative formation was lower compared to THMs and HAAs. This is likely due to the instability of HANs, as they can degrade and transform into other DBPs, such as HAAs, under experimental conditions (Zhang et al. 2009). This transformation process aligns with the observation of significant HAA formation at higher temperatures.

Following the reduction in DON (Figure 2(a)), alum coagulation can decrease the potential HAN formation concentration by 22–28%. The results are comparable with a previous study that mentioned a reduction of potential HAN formation with alum coagulation by 23% (Wang et al. 2021) and 11–64% (Włodarczyk-Makuła & Nowacka-Klusek 2019). The predominant haloacetonitrile species in the speciation profile is DCAN, constituting more than 97% of the HAN concentration. This result aligns with previous research that showed that during the chlorination process, tyrosine and tryptophan, as HAN precursors, produced the greatest amount of DCAN and generated a small amount of TCAN (Jia et al. 2016).

Figure 6(d) revealed that after the incubation period of 7 days, the concentration of HK/DOC at 65 °C showed a significant rise of 36–39% compared to that at 25 °C. Interestingly, coagulation procedures increased the HK/DOC concentration by 47% (from 0.15 to 0.22 μg/mg-C) in the soil sample at 25 °C and by 54% (from 0.25 to 0.38 μg/mg-C) in the soil sample at 65 °C. However, at 45 °C, a slight decrease in HK concentration after coagulation (17%) was observed. Similar to HAN, HK formation was influenced by the presence of nitrogen (Li et al. 2020a). HK was also an intermediate byproduct that can rapidly be formed during chlorination and converted to THMs and HAAs due to reaction with residual chlorine or hydrolysis process (Fu et al. 2017). The coagulation process commonly removes HPOA molecules such as humic substances but has a low impact on hydrophilic substances (Notodarmojo et al. 2023; Qadafi et al. 2023). The presence of ketoacid in 3-oxopentanedioic acid, which appears to be a hydrophilic or transphilic fraction, has been reported to be an HK precursor in water (Bond et al. 2012). In addition, the study mentioned that the precursor of HK was not effectively controlled by alum coagulation (Peterson et al. 2022). Removing the hydrophobic fraction may shift the chlorine reactivity to the HK precursor, increasing HK formation. Like HAN, HK is an intermediate product that can be mineralized into THMs and HAAs. The result is similar to the study that treatment of HKs (1,1-dichloropropanone, 1,1,1-trichloropropanone) in wastewater with alum coagulation produces an increase of those species up to 183% (from 15.4 to 43.7 nM). In addition, the study mentioned that the precursor of HKs was not effectively controlled by alum coagulation (Peterson et al. 2022). However, another study showed different results; removing potential HKs formation with alum coagulation can reach 51–89% (Włodarczyk-Makuła & Nowacka-Klusek 2019). These differences underscore the importance of optimizing treatment conditions, such as adjusting pH and understanding water composition, to enhance HK precursor removal efficacy. This suggests that treatment outcomes for HK precursors are highly dependent on the operational and water matrix-specific parameters.

Based on Figure 6(e) the highest concentration of TCNM/DOC was observed at 25 °C, followed by a noticeable decrease of 30% at 45 °C and a further reduction of 24% at 65 °C. The decrease in TCNM/DOC concentrations as temperatures increase – 30% at 45 and 24% at 65 °C – may be attributed to altered reaction kinetics or the thermal degradation of precursors. TCNM, as N-DBP, shows a different trend with HAN. The possible reason is the precursor difference between TCNM and DCAN (as the dominant analyte in HAN). Most nitrogen in TCNM originated from glycine, and DCAN originated from pyrrole (Yang et al. 2012), tyrosine, and tryptophan (Jia et al. 2016). The chlorination environment also impacts the N-DBP speciation. Chlorination under neutral conditions tends to produce DCAN than TCNM. The higher Cl2: DON ratio could also increase the DCAN formation faster than TCNM (Jia et al. 2016).

The coagulation processes proved effective, reducing potential TCNM formation concentrations by 49–61%. A previous study showed the removal of potential TCNM formation with alum coagulation by 69% (Włodarczyk-Makuła & Nowacka-Klusek 2019).

The highest contributor to the cytotoxicity potency of chlorinated soil runoff is DCAN, with 59–66% of total cytotoxicity potency. Even though, in mass weight, DCAN only contributes 2% of the total concentration of DBPs. A plausible explanation of this phenomenon is the escalation of DON as temperature increases (Øygarden et al. 2014), leading to increasing concentrations of DCAN precursors such as pyrrole (Yang et al. 2012), tyrosine and tryptophan (Jia et al. 2016). Based on Table 1, DCAN has a lower LC50 value than other DBP speciation measured (except MBAA, BCAN, and DBAN), indicating that DCAN has higher toxicity potency than other DBP (Wagner & Plewa 2017). However, in this case, the mass concentration of BCAN and DBAN is much lower compared to DCAN, and no MBAA is found in the sample, which is why DCAN is the highest contributor to the cytotoxicity potency of chlorinated soil runoff.

Understanding the effect of extreme soil temperature on DOM in soil runoff and chlorinated water allows us to predict the potential deterioration in water quality caused by DBP. By understanding the potential toxicity of DBP speciation, we can prepare the best mitigation and technology to avoid future adverse effects on human health, such as using preozonitation and ozonation integrated with biological activated carbon filtration (O3/BAC) (Zhang et al. 2021).

In conclusion, this study elucidates the intricate interactions between extreme temperatures and enhanced coagulation on soil runoff, mainly how they influence the production and cytotoxicity of DBP precursors. Short-term exposure to elevated temperatures, especially at 65 °C, significantly increases soil respiration, DOC, and DON in soil runoff. Correspondingly, the concentrations of HAAs, HANs, and HKs levels increased with temperature, highlighting the temperature-associated dynamics of DBP formation. Enhanced coagulation effectively reduced the concentrations of THM, HAA, HAN, and TCNM, but unexpectedly, it increased HK formation, especially at 65 °C, which warrants further investigation.

Moreover, the study demonstrates that extreme temperatures significantly enhance the cytotoxicity of DBP speciation, underscoring potential health and environmental concerns. The findings deepen our understanding of the complex relationships among soil surface temperature, microbial activities/soil respiration, and DBP dynamics. This knowledge is crucial for developing targeted water treatment strategies to safeguard human health and the environment.

We greatly appreciate the continued analytical and technical support provided by Prof. Gen-Shuh Wang from the School of Public Health at National Taiwan University and Prof. Yi-Hsueh Chuang at National Yang Ming Chiao Tung University.

This study was supported by the Taiwan National Science and Technology Council project # NSTC 112-2221-E-002-063-MY3 and National Taiwan University Core Consortiums project (NTU-CC-113L894404) within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan.

N.N.A., and H.T. conceptualized the study, wrote, reviewed, and edited the article. M.Q. reviewed the article.

All relevant data are included in the paper or its Supplementary Information.

The authors have no conflicts of interest to declare.