Method for evaluating and tracing wastewater comprehensive toxicity based on multi-index response of algal cells

Abstract

This invention discloses a comprehensive toxicity evaluation and source tracing method for wastewater based on multi-indicator responses of algal cells. The method includes: adding the wastewater to be tested into an algal cell culture system for exposure and obtaining algal cell viability index values ​​at multiple exposure times; determining the pollution toxicity type of the wastewater based on a machine learning model and / or a multi-algal species combination detection method, and tracing the pollution source based on the pollution toxicity type; and evaluating the toxicity and risk of the wastewater based on the pollution toxicity type and algal cell viability index values ​​using a comprehensive toxicity evaluation system. This invention can identify toxicity types based on algal cell viability indicators and construct a comprehensive toxicity assessment system to comprehensively evaluate the toxicity and risk of wastewater, solving three major problems of existing technologies: static endpoint determination, neglect of species sensitivity differences, and missed detection of chronic toxicity.

Description

Technical Field

[0001] This invention relates to the field of water quality risk control and environmental monitoring technology, specifically to a method for comprehensive toxicity evaluation and source tracing of wastewater based on multi-indicator responses of algal cells. Background Technology

[0002] With my country's rapid economic development, the discharge of industrial wastewater and domestic sewage is constantly increasing, containing a large number of known or unknown toxic and harmful pollutants. These pollutants may pose potential risks to aquatic ecosystems and human health after discharge. Currently, the assessment of complex pollutants in wastewater mainly relies on traditional chemical analysis methods, focusing on the qualitative and quantitative analysis of specific pollutants. However, wastewater contains a wide variety of pollutants, many of which are difficult to identify through chemical analysis, increasing the uncertainty of water quality assessment. Furthermore, different pollutants may exhibit synergistic, antagonistic, or cumulative effects. Therefore, relying solely on the chemical analysis of a single component is insufficient to comprehensively reflect the actual harm of complex pollutants to the ecological environment and human health. Compared to chemical analysis, biotoxicity detection methods can integrate the interactions of multiple pollutants, establishing a dose-response relationship between pollutant concentration and biological effects, thus directly reflecting the comprehensive toxicity of polluted water to organisms.

[0003] Microalgae, as typical model organisms for aquatic ecotoxicology, are widely used in studies of the toxic effects of pollutants due to their direct exposure to the aquatic environment, short generation cycles, and sensitive environmental responses. Among them, *Cercospora spp.* is one of the species recommended by the International Organization for Standardization (ISO) for ecotoxicological bioassays; this single-celled green algae is generally more sensitive to various pollutants than other species. Furthermore, *Cercospora spp.* not only has broad adaptability to freshwater environments but also possesses advantages in experimental biology such as simple culture conditions and a short generation cycle.

[0004] Currently, microalgal toxicity bioassay systems primarily use algal growth inhibition rate, cell viability, photosynthetic efficiency, and changes in physiological indicators as toxicity endpoints. Among these, the half-maximal effective concentration (WIF) and toxicity units (TUDs) determined by algal growth inhibition experiments are the most commonly used indicators for assessing the toxicity level and risk level of water bodies. However, algal growth inhibition experiments are time-consuming (typically 72 or 96 hours) and cannot be performed with high-throughput detection, limiting their application in practical wastewater comprehensive toxicity assessments. Existing technologies have developed rapid detection methods based on algal cell viability indicators (such as Chinese patent application CN120594781A), which, through FDA fluorescent staining and high-throughput detection using an enzyme-linked immunosorbent assay (ELISA) reader, can rapidly evaluate the comprehensive toxicity of wastewater within 20–48 hours.

[0005] However, existing technologies still have the following shortcomings: First, most existing algal toxicity detection methods only focus on the toxicity intensity of a single endpoint such as photosynthetic activity or growth inhibition, and cannot provide information on the types of potential toxins. As a result, when toxicity exceeds the standard, it is difficult to further guide pollution source tracing and process optimization, and the function is relatively simple. Second, the response of a single algal species to mixed pollutants has an "additive" characteristic, making it difficult to analyze the toxicity contribution of each component in complex wastewater and to identify the main toxic factors. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a comprehensive wastewater toxicity evaluation and source tracing method based on multi-index responses of algal cells. This method rapidly evaluates the comprehensive toxicity of wastewater while simultaneously enabling rapid attribution of toxicity types and quantitative analysis of major toxic factors.

[0007] A comprehensive toxicity assessment and source tracing method for wastewater based on multi-indicator responses of algal cells includes the following steps: The wastewater to be tested was added to an algal cell culture system for exposure and toxin treatment, and algal cell viability index values ​​were obtained at multiple exposure times. Based on machine learning models and / or multi-algae combination detection methods, the pollution toxicity type of the wastewater to be tested is determined, and pollution source tracing is carried out based on the pollution toxicity type; Based on the pollution toxicity type and algal cell viability index of the wastewater to be tested, a comprehensive toxicity evaluation system is used to evaluate the toxicity and risk of the wastewater to be tested. The comprehensive toxicity evaluation system includes: evaluating the toxicity level of the wastewater to be tested based on the acute toxicity intensity and damage reversibility of algal cells, and evaluating the risk level of the wastewater to be tested based on the instantaneous inhibition rate and cumulative toxicity load of algal cells.

[0008] Note: The above method evaluates toxicity intensity through algal cell viability indicators, identifies toxicity types through two methods, and analyzes the toxicity contribution through multi-algal species combinations. Based on this, a multi-dimensional comprehensive toxicity assessment system is innovatively constructed. Through dynamic assessment of effect and recovery, adaptive level correction of species sensitivity weight, and time-effect cumulative toxicity load model, it solves three major problems of existing technologies: static endpoint determination, neglect of species sensitivity differences, and missed detection of chronic toxicity.

[0009] Furthermore, the ammonia nitrogen concentration of the wastewater to be tested is ≤30 mg / L; the algal cell viability index value is determined and calculated by the FDA / PI dual fluorescence staining method.

[0010] Note: The above method controls the exposure concentration range by limiting the ammonia nitrogen concentration, avoiding acute toxic interference to algal cells caused by high concentrations of ammonia nitrogen, and ensuring the accuracy and repeatability of the experiment.

[0011] Furthermore, the type of pollution toxicity of the wastewater to be tested is determined based on a machine learning model, including: Chlorophyll fluorescence induction kinetic curves of algal cells during exposure to toxins were collected, and characteristic parameters in the chlorophyll fluorescence induction kinetic curves were extracted. The feature parameters are input into the machine learning model, and the pollution toxicity type of the wastewater to be tested is output. The characteristic parameters include at least one of initial fluorescence intensity, maximum fluorescence intensity, fluorescence decline rate, steady-state time, and half-inhibition time; the toxicity type of the wastewater to be tested is any one of heavy metals, organic toxins, surfactants, or inorganic salts. The machine learning model is a random forest classification model.

[0012] Note: The above method can quickly and accurately output the toxicity type of the wastewater to be tested by collecting the chlorophyll fluorescence induction kinetic curve of algal cells during the exposure to toxicity. Compared with traditional chemical analysis methods or single biological tests, this method does not require prior knowledge of the pollutant composition, has a sensitive response and reliable classification, and the random forest model has the ability to resist overfitting and handle nonlinear relationships, which can effectively distinguish different types of toxic mechanisms, providing clear and quantifiable toxicity type basis for subsequent pollution source tracing and comprehensive toxicity assessment.

[0013] Furthermore, the pollution toxicity type of the wastewater to be tested is determined based on the multi-algae combination detection method, including: Multiple sensitive algal species were selected and combined with the algal cells to form a multi-algal species detection plate; The wastewater to be tested was subjected to parallel exposure to the multi-algae detection plate for 20-48 hours; then, based on the pattern recognition analysis method, the toxicity contribution rate of each type of pollution was obtained. The type of pollution with the highest contribution rate is taken as the type of pollution toxicity of the wastewater to be tested.

[0014] Explanation: The above method utilizes the varying sensitivities of different algal species to different types of pollutants by selecting multiple sensitive algal species and target algal cells to form a multi-species detection plate for parallel exposure to toxicity. This allows for the identification of toxicity sources in mixed pollution systems. Furthermore, based on pattern recognition analysis, the contribution rate of each pollutant toxicity type is quantitatively calculated, and the species with the highest contribution rate is identified as the primary toxicity type of the wastewater being tested. This method overcomes the limitations of single-species testing, which can only provide comprehensive toxicity intensity but cannot distinguish toxicity types. It provides more refined and reliable toxicity classification results for complex wastewater samples, effectively supporting subsequent pollution source tracing and risk assessment.

[0015] Furthermore, based on machine learning models and multi-algae species combination detection methods, the pollution toxicity type of the wastewater to be tested is determined, including: The pollution toxicity type of the wastewater under test was output by using machine learning models and multi-algae combination detection methods, respectively. The results are considered reliable when the output of the machine learning model matches the output of the multi-algae species combination detection method. When the output of the machine learning model is inconsistent with the output of the multi-algae combination detection method, the parallel experiment of the multi-algae combination detection method is reset or the detection conditions are adjusted according to the output of the machine learning model. The multi-algae combination detection method is then tested again, and the output of the multi-algae combination detection method is taken as the pollution toxicity type.

[0016] Furthermore, the pollution source tracing based on pollution toxicity type includes: A toxicity type-characteristic pollutant mapping library is established, which includes pollutants and emission sources corresponding to heavy metals, pollutants and emission sources corresponding to organic toxins, pollutants and emission sources corresponding to surfactants, and pollutants and emission sources corresponding to inorganic salts. The determined pollution toxicity type is matched with the mapping library to identify characteristic pollutants and emission sources, which are then identified as the pollution sources of the wastewater to be tested.

[0017] Explanation: The above method establishes a toxicity type-characteristic pollutant mapping library covering heavy metals, organic toxins, surfactants, and inorganic salts in advance. It associates each toxicity type with specific characteristic pollutants and their typical emission sources. Thus, after determining the toxicity type of the wastewater to be tested, it can quickly and intuitively match and identify possible characteristic pollutants and pollution sources, realizing reverse tracing from toxic effects to pollution sources.

[0018] Furthermore, based on the pollution toxicity type and algal cell viability index value of the wastewater to be tested, a comprehensive toxicity evaluation system is used to evaluate the toxicity and risk of the wastewater to be tested; including: The toxicity level of the wastewater to be tested is determined based on the type of pollution toxicity, acute toxicity intensity, and reversibility of damage. The risk level of the wastewater to be tested is determined based on the algal cell viability index, the instantaneous inhibition rate of algal cells, and the cumulative toxicity load.

[0019] Note: The above method combines the type of pollution toxicity with the intensity of acute toxicity and the reversibility of damage to determine the toxicity level. At the same time, it determines the risk level based on algal cell viability, instantaneous inhibition rate and cumulative toxicity load. It realizes a multi-dimensional comprehensive evaluation from the perspective of toxicity type, acute effect, reversibility and cumulative risk, which makes up for the shortcomings of traditional single indicators that cannot distinguish between short-term impact and long-term chronic harm. It provides a more scientific and comprehensive decision-making basis for wastewater toxicity classification and risk warning.

[0020] Furthermore, determining the toxicity level of the wastewater to be tested based on its pollution toxicity type, acute toxicity intensity, and reversibility of damage includes: Using the acute toxicity intensity and damage reversibility of algal cells as a two-dimensional matrix, multiple toxicity levels are generated: Level I (low toxicity - reversible), Level II (low toxicity - partially reversible), Level III (low toxicity - irreversible), Level IV (moderate toxicity - reversible), Level V (moderate toxicity - partially reversible), Level VI (moderate toxicity - irreversible), Level VII (high toxicity - reversible), Level VII (high toxicity - partially reversible), Level IX (high toxicity - irreversible), Level X (extremely high toxicity - reversible), Level XI (extremely high toxicity - partially reversible), and Level XII (extremely high toxicity - irreversible). The acute toxicity intensity is determined based on the initial inhibition rate of the algal cells, and the damage reversibility is the difference between the initial inhibition rate and the post-recovery inhibition rate. The post-recovery inhibition rate is the inhibition rate measured after transferring exposed algal cells to a standard culture medium without the test wastewater and culturing for 24 hours. Obtain EC values ​​for the toxicity of various algae to each type of pollution. 50 ratio; Based on the type of pollution toxicity of the wastewater to be tested and the species of algae cells, the corresponding EC is selected. 50 The ratio is used as the sensitivity weighting coefficient of the wastewater to be tested; the sensitivity weighting coefficient is multiplied by the initial inhibition rate of algal cells to obtain the equivalent inhibition rate; The toxicity level of the wastewater to be tested is determined by comparing the equivalent inhibition rate with the two-dimensional matrix.

[0021] Explanation: The above method classifies toxicity levels by constructing a two-dimensional matrix with acute toxicity intensity and damage reversibility as dimensions. It introduces the EC50 ratio based on pollutant toxicity type and algal species characteristics as a sensitivity weighting coefficient. The initial inhibition rate is corrected to an equivalent inhibition rate before comparison with the matrix, effectively eliminating evaluation bias caused by inherent sensitivity differences among different algal species to different types of pollutants. This makes the determination of toxicity levels more accurate and objective, avoiding misjudgments or omissions due to inappropriate algal species selection. Furthermore, this two-dimensional matrix breaks through the limitations of traditional single-threshold linear classification, achieving a leap from "static endpoint determination" to "dynamic damage assessment." Taking a certain industrial wastewater as an example: if IR0=65% (high toxicity) and IDI=5% (reversible), it is classified as "high toxicity-reversible" (Level VII), indicating that although short-term toxicity is high, the ecosystem has self-repair capabilities, and its emergency response priority is lower than that of "high toxicity-irreversible" Level IX wastewater. This mechanism provides environmental managers with a more refined basis for risk classification. Regarding the equivalent inhibition rate, taking ammonia nitrogen as an example, the sensitivity weighting coefficient of *Scenedesmus* to ammonia nitrogen is 2.3 (i.e., the EC50 of *Scenedesmus* is 1 / 2.3 of that of *Creepingea spp.*). If a wastewater is identified as having "ammonia-dominant toxicity," and the measured inhibition rate of *Scenedesmus* is 40%, the equivalent inhibition rate = 40% × 2.3 = 92%, corresponding to a "high toxicity" level. However, if *Creepingea spp.* were used directly, the inhibition rate might only be "moderate toxicity." This correction mechanism effectively avoids the underestimation of ecological risks due to differences in algal species sensitivity.

[0022] Furthermore, the determination of the risk level of the wastewater to be tested based on the algal cell viability index, the instantaneous inhibition rate of algal cells, and the cumulative toxicity load includes: Based on the algal cell viability index values ​​under multiple exposure times, the instantaneous inhibition rate of algal cells under multiple times was calculated using formula (1). (1); In equation (1), denoted as , where is the instantaneous inhibition rate of algal cells; a represents the viability index value of algal cells on standard culture medium; b represents the viability index value of algal cells exposed to the toxin. Construct the time-inhibition rate curve and calculate the area under the curve of the time-inhibition rate curve; The cumulative toxicity load of the wastewater to be tested is determined based on the time distribution of the area under the curve. The risk level of the wastewater to be tested is determined based on the initial inhibition rate and cumulative toxicity load. The risk levels of the wastewater to be tested include: low risk (IR0≤4% and AUC≤480%·h), medium risk (IR0≤4% and AUC>480%·h), high risk (IR0>4% and AUC≤480%·h) and very high risk (IR0>4% and AUC>480%·h).

[0023] Explanation: The above method dynamically monitors algal cell viability at multiple exposure times and calculates the instantaneous inhibition rate to construct a time-inhibition rate curve. Then, it uses the area under the curve to quantify the cumulative toxic load and combines it with the initial inhibition rate to classify wastewater risk into four types: low risk, medium risk, high risk, and extremely high risk. This enables a joint assessment of the short-term impact and long-term chronic toxicity of wastewater, providing a more refined and time-dimensional scientific basis for risk warning and graded management.

[0024] The beneficial effects of this invention are: First, this invention constructs a dynamic two-dimensional assessment system based on "effect-recovery." Traditional methods only measure the inhibition rate at the exposure endpoint, failing to distinguish between the reversibility and irreversibility of toxic effects. This invention, by measuring the recovery rate a second time after the recovery period and calculating the irreversible damage index, reconstructs the toxicity level assessment into a two-dimensional matrix of "acute toxicity intensity × damage reversibility," generating a 12-level fine classification. This achieves a leap from static endpoint assessment to dynamic damage evaluation, providing environmental managers with a more refined basis for risk classification and avoiding the underestimation of irreversible ecological damage.

[0025] Second, this invention establishes an adaptive grading correction mechanism for species sensitivity weights. Different algal species exhibit significant differences in their sensitivity to different types of pollutants (e.g., Scenedesmus is approximately 2.3 times more sensitive to ammonia nitrogen than Crested Algae). Traditional methods use fixed thresholds to uniformly determine the grading of all wastewater, which can easily lead to an underestimation of the ecological risk of specific types of pollutants. Based on the theory of species sensitivity distribution, this invention pre-constructs a database of standard algal species sensitivity for different toxicity types. It dynamically retrieves weight coefficients based on the identified toxicity type, converts the measured inhibition rate into the equivalent inhibition rate of the standard species, and then determines the grading. An automatic grading adjustment mechanism is also included, effectively avoiding the underestimation of risk due to differences in algal species sensitivity.

[0026] Third, this invention introduces a time-effect cumulative toxicity load model. Traditional methods only measure the instantaneous inhibition rate at the exposure endpoint, failing to reflect the dynamic action of pollutants within the exposure period and lacking sufficient ability to distinguish between fast-acting and slow-acting toxins. This invention, by setting multiple time points for continuous measurement within the exposure period, calculates the area under the time-effect curve as an indicator of cumulative toxicity load and establishes a dual-threshold cross-validation system of "instantaneous inhibition rate threshold + cumulative toxicity load threshold," which can effectively identify potential chronic risks of "instantaneous compliance but cumulative exceedance" (e.g., classified as "medium risk"), thus overcoming the deficiency of single-endpoint detection in missing chronic toxicity.

[0027] Fourth, this invention organically integrates the above three dimensions to form a multi-dimensional dynamic toxicity level determination system of "acute toxicity intensity + damage reversibility + species sensitivity correction + cumulative toxicity load", which comprehensively outputs 12 levels of fine toxicity level and 4 levels of ecological risk level, providing more scientific and comprehensive technical support for wastewater toxicity supervision, pollution source tracing and ecological risk assessment. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating an embodiment of the present invention. Detailed Implementation

[0029] To further illustrate the methods and effects of this invention, the technical solution of this invention will be clearly and completely described below in conjunction with experiments.

[0030] Example 1: A comprehensive toxicity assessment and source tracing method for wastewater based on multi-index responses of algal cells, comprising the following steps: S101. The wastewater to be tested was added to the algal cell culture system for exposure and toxin treatment, and the algal cell viability index values ​​were obtained at multiple exposure and toxin treatment times. The ammonia nitrogen concentration of the wastewater to be tested is ≤30 mg / L; the algal cell viability index value is determined and calculated by FDA / PI dual fluorescence staining method; Specifically, high concentrations of ammonia nitrogen have significant acute toxicity to algal cells, which can mask the true toxic effects of other pollutants. Based on ammonia nitrogen tolerance studies of *Crescentia serratifolia*, an ammonia nitrogen concentration of 30 mg / L was determined as the critical concentration. Before measuring the ammonia nitrogen level in the wastewater, the wastewater was filtered through a 0.22 μm aqueous filter membrane. The method for determining whether the wastewater needs dilution based on the ammonia nitrogen concentration includes: when the ammonia nitrogen concentration in the water sample is >30 mg / L, the sample is diluted with ultrapure water to a concentration ≤30 mg / L; when the ammonia nitrogen concentration in the water sample is ≤30 mg / L, no dilution is required. For example, the algal cells are *Creeping echinocandus* in the logarithmic growth phase, and the cell concentration in the algal cell culture system is 10⁵~10⁶ cells / mL; the method for determining the algal cell viability index is a cell viability staining assay, specifically including: adding fluorescein diacetate to the wastewater to be tested or diluted to the wastewater to be tested to a concentration of 25 μM, incubating at room temperature and in the dark for 30~60 min, and acquiring the fluorescence intensity using a multifunctional microplate reader; the algal cell viability index is then determined. The following formula (1-1) is used to calculate: (1) In the formula, This refers to the measured values ​​of algal cell viability. This includes the measurement values ​​of algal cell viability indicators in the wastewater to be tested without dilution. Algal cell viability index measured in diluted wastewater Two values; To expose the average fluorescence intensity of algal cells after exposure to the virus, The average fluorescence intensity of algal cells that have not undergone exposure to the toxin; The measured values ​​based on algal cell vitality indicators Determine the algal cell viability index value of the wastewater to be tested. The method is as follows: When the wastewater to be tested does not require dilution, the algal cell viability index value of the wastewater to be tested... = ; When the wastewater to be tested needs to be diluted, the algal cell viability index value of the wastewater to be tested is... The following formula (1-2) is used to calculate: (1-2) In the formula, The value represents the algal cell viability index of the wastewater to be tested. This represents the dilution factor.

[0031] S102. Determine the pollution toxicity type of the wastewater to be tested based on the machine learning model and / or the multi-algae combination detection method, and trace the source of pollution based on the pollution toxicity type. I. In some embodiments of the present invention, determining the pollution toxicity type of the wastewater to be tested based on a machine learning model includes: ① Collect chlorophyll fluorescence induction kinetic curves of algal cells during exposure to the toxin, and extract characteristic parameters from the chlorophyll fluorescence induction kinetic curves; The method for acquiring the chlorophyll fluorescence induction kinetic curve includes: during the exposure process, using a chlorophyll fluorometer or a multifunctional microplate reader equipped with a chlorophyll fluorescence detection module, at an excitation light intensity of 100~500 μmol·m -2 ·s -1 Under these conditions, chlorophyll fluorescence intensity change data were continuously collected over 10–30 min to generate fluorescence induction kinetic curves.

[0032] ② Input the feature parameters into the machine learning model and output the pollution toxicity type of the wastewater to be tested; The characteristic parameters include at least one of initial fluorescence intensity, maximum fluorescence intensity, fluorescence decline rate, steady-state time, and half-inhibition time; the pollution toxicity type of the wastewater to be tested is any one of heavy metals, organic toxins, surfactants, or inorganic salts. The machine learning model is a random forest classification model; this model is established by pre-determining the kinetic characteristic parameters of four typical toxic substances. The four typical categories of toxic substances include: heavy metal toxic substances, organic toxic substances, surfactant toxic substances, and inorganic salt toxic substances; Among them, the kinetic characteristics of heavy metal toxins are: they act rapidly on the photosynthetic system of algae, inhibit chlorophyll fluorescence, and exhibit the characteristics of "rapid inhibition and stabilization within 10 minutes"; The kinetic characteristics of organic toxins are: by interfering with the photosynthetic electron transport or metabolic processes of algae, the toxic response is delayed, exhibiting the characteristics of "continuous decline, without reaching a plateau within 30 minutes"; The kinetic characteristics of surfactant-type toxins are: rapid destruction of algal cell membrane structure, resulting in a fluorescence signal that "decreases rapidly in the initial stage and then tends to level off". The kinetic characteristics of inorganic salt toxins are as follows: by affecting the molecular transport and activity of cytochrome c oxidase and altering extracellular osmotic pressure, algal cells lose or absorb water, resulting in photosynthetic activity exhibiting a characteristic of "slow inhibition, continuous decline, and failure to reach a plateau within 30 minutes." The random forest classification model described above achieves a classification accuracy of ≥85% for the four typical toxins.

[0033] In other embodiments of the present invention, the pollution toxicity type of the wastewater to be tested is determined according to the multi-algae combination detection method, including: ① Select multiple sensitive algal species and the algal cells to form a multi-algal species detection plate; The multi-algae combination detection plate includes at least three algae species with significantly different sensitivities to different pollutants. The algae species are selected from two or more of the following: *Creeping echinocandus*, which is sensitive to heavy metals and organic pesticides; *Scenedesmus*, which is sensitive to ammonia nitrogen; *Chlorella*, which is sensitive to surfactants; and *Dunaliella salina*, which is sensitive to inorganic salts. The differentiated response mode normalizes the cell viability index values ​​of each algae species to construct a toxicity effect spectrum vector, which is then compared with a preset single pollutant effect spectrum library to analyze the toxicity contribution of each component in the mixed pollutant.

[0034] ② The wastewater to be tested was subjected to parallel exposure to the multi-algae detection plate for 20-48 hours; then, based on the pattern recognition analysis method, the toxicity contribution rate of each type of pollution was obtained. ③The type of pollution with the highest contribution rate is taken as the type of pollution toxicity of the wastewater to be tested.

[0035] In other embodiments, machine learning models and multi-algal species combination detection methods are combined for determination; The results are considered reliable when the output of the machine learning model matches the output of the multi-algae species combination detection method. When inconsistencies occur, the parallel experiment is reset or the detection conditions are adjusted based on the output of the machine learning model, or a mixed toxicity warning is triggered (output when the confidence level is below 90% and the residual of the effect spectrum is greater than the preset threshold); the multi-algal combination detection method is performed again, and the pollution toxicity type output by the multi-algal combination detection method is taken as the standard.

[0036] II. The pollution source tracing based on pollution toxicity type includes: A toxicity type-characteristic pollutant mapping library is established, which includes pollutants and emission sources corresponding to heavy metals, pollutants and emission sources corresponding to organic toxins, pollutants and emission sources corresponding to surfactants, and pollutants and emission sources corresponding to inorganic salts. The determined pollution toxicity type is matched with the mapping library to identify characteristic pollutants and emission sources, which are then identified as the pollution sources of the wastewater to be tested.

[0037] S103. Based on the pollution toxicity type and algal cell viability index value of the wastewater to be tested, a comprehensive toxicity evaluation system is used to evaluate the toxicity and risk of the wastewater to be tested; the comprehensive toxicity evaluation system includes: evaluating the toxicity level of the wastewater to be tested based on the acute toxicity intensity and damage reversibility of algal cells, and evaluating the risk level of the wastewater to be tested based on the instantaneous inhibition rate and cumulative toxicity load of algal cells.

[0038] A comprehensive toxicity evaluation system is constructed based on the pollution toxicity type of the wastewater to be tested and the algal cell viability index value; including: (1) Determine the toxicity level of the wastewater to be tested based on its pollution toxicity type, acute toxicity intensity, and reversibility of damage; including: A multi-level toxicity rating is generated using a two-dimensional matrix of acute toxicity intensity and damage reversibility of algal cells. The acute toxicity intensity is determined based on the initial inhibition rate of the algal cells, and the damage reversibility is the difference between the initial inhibition rate and the post-recovery inhibition rate of the algal cells. The post-recovery inhibition rate is the inhibition rate measured after the algal cells exposed to the toxicity are transferred to a standard culture medium without the test wastewater and cultured for 24 hours. Obtain EC values ​​for the toxicity of various algae to each type of pollution. 50 ratio; Based on the type of pollution toxicity of the wastewater to be tested and the species of algae cells, the corresponding EC is selected. 50 The ratio is used as the sensitivity weighting coefficient of the wastewater to be tested; the sensitivity weighting coefficient is multiplied by the initial inhibition rate of algal cells to obtain the equivalent inhibition rate; The toxicity level of the wastewater to be tested is determined by comparing the equivalent inhibition rate with the two-dimensional matrix.

[0039] Specifically, it includes the following steps one and two; Step 1: Two-dimensional dynamic determination of effect - recovery At the exposure toxicity endpoint (20 - 48 h), measure the initial inhibition rate IR0 of algal cell viability; subsequently, transfer the algal cells to fresh medium for 24 h of recovery culture, and measure the inhibition rate IR after recovery 24 ; Calculate the irreversible damage index IDI = IR0 - IR 24 .

[0040] It should be understood that IDI reflects the reversibility degree of toxic effects. IDI ≤ 10% indicates that the damage is basically reversible (the metabolic activity of algal cells can be self - repaired during the recovery period); 10% < IDI ≤ 30% indicates partial reversibility (part of the damage is repaired, and the long - term exposure effect needs to be concerned); IDI > 30% indicates irreversible damage (the functions of algal cells are severely damaged, indicating difficulties in the recovery of the ecosystem).

[0041] Reconstruct the toxicity level determination into a two - dimensional matrix: the vertical axis is the acute toxicity intensity (low / medium / high / extremely high), which is divided according to the initial inhibition rate IR0 according to the following criteria: IR0 ≤ 27% is low toxicity (corresponding to the original f1 > 73%); 27% < IR0 ≤ 52% is medium toxicity (corresponding to 73% ≥ f1 > 48%); 52% < IR0 ≤ 96% is high toxicity (corresponding to 48% ≥ f1 > 4%); IR0 > 96% is extremely high toxicity (corresponding to f1 ≤ 4%). The horizontal axis is the damage reversibility (reversible / partially reversible / irreversible), and cross - positioning generates a 12 - level fine toxicity level.

[0042] This two - dimensional matrix breaks through the limitation of traditional single - threshold linear classification, and realizes the leap from "static endpoint determination" to "dynamic damage assessment". Taking a certain industrial wastewater as an example: if IR0 = 65% (high toxicity) and IDI = 5% (reversible), it is determined as "high toxicity - reversible" (Grade VII), indicating that although the short - term toxicity is strong, the ecosystem has self - repair ability, and the emergency response priority is lower than that of the Grade IX wastewater with "high toxicity - irreversible". This mechanism provides a more refined risk classification basis for environmental managers.

[0043] Step 2: Adaptive level correction of species sensitivity weights The specific method of the adaptive level correction of species sensitivity weights is as follows: Based on the species sensitivity distribution theory, pre - construct a standard algal species sensitivity database for different toxicity types, and record the EC50 ratios of Scenedesmus obliquus, Chlorella vulgaris, Scenedesmus quadricauda, and Dunaliella salina to various types of pollutants (heavy metals, organic poisons, surfactants, inorganic salts) as sensitivity weight coefficients.

[0044] Exemplarily, the pre - experimental measurement results are shown in Table 1 below (taking Scenedesmus obliquus as the reference species, weight coefficient = 1.00): Table 1 Pre - experimental measurement results

[0045] In actual judgment, the corresponding sensitivity weight coefficient is dynamically retrieved based on the identified toxicity type, and the equivalent inhibition rate of the standard species is calculated using the following formula: IR equivalent = IR measured × weight coefficient; the IR equivalent is then substituted into the grade threshold for judgment. When a highly sensitive algal species is detected to show significant inhibition while a low-sensitivity algal species shows a mild response, the grade adjustment mechanism is automatically triggered.

[0046] Taking ammonia nitrogen as an example, the sensitivity weighting coefficient of *Scenedesmus* to ammonia nitrogen is 2.3 (i.e., the EC50 of *Scenedesmus* is 1 / 2.3 of that of *Creepingea spp.*). If a wastewater is identified as having "ammonia-dominant toxicity," and the measured inhibition rate of *Scenedesmus* is 40%, the equivalent inhibition rate = 40% × 2.3 = 92%, corresponding to a "high toxicity" level. However, if *Creepingea spp.* were used directly, the inhibition rate might only be "moderate toxicity." This correction mechanism can effectively avoid the underestimation of ecological risks due to differences in algal species sensitivity.

[0047] (2) Determine the risk level of the wastewater to be tested based on the algal cell viability index, the instantaneous inhibition rate of algal cells, and the cumulative toxicity load. This includes: Based on the algal cell viability index values ​​under multiple exposure times, the instantaneous inhibition rate of algal cells under multiple times was calculated using formula (1). (1); In equation (1), denoted as , where is the instantaneous inhibition rate of algal cells; a represents the viability index value of algal cells on standard culture medium; b represents the viability index value of algal cells exposed to the toxin. Construct the time-inhibition rate curve and calculate the area under the curve of the time-inhibition rate curve; The cumulative toxicity load of the wastewater to be tested is determined based on the time distribution of the area under the curve. The risk level of the wastewater to be tested is determined based on the initial inhibition rate IR0 and the cumulative toxicity load. The risk level of the wastewater to be tested includes low risk (IR0≤4% and AUC≤480%·h), medium risk (IR0≤4% and AUC>480%·h), high risk (IR0>4% and AUC≤480%·h) and very high risk (IR0>4% and AUC>480%·h).

[0048] Specifically, algal cell viability indicators were continuously measured at at least six time points (e.g., 0, 4, 8, 12, 20, 30, 40, and 48 h) within a 20–48 h exposure period. A time-effect curve was constructed with time on the x-axis and inhibition rate on the y-axis. The area under the curve (AUC) was calculated using the trapezoidal method: AUC = Σ[(t i+1 -t i )×(IRi +IR i+1 [2] The concept of "toxicokinetics" is introduced, and the exposure mode is distinguished by analyzing the time distribution characteristics of AUC: rapid-acting (AUC concentrated in 0~8 h, such as heavy metal toxins, showing rapid inhibition followed by stabilization) and slow-acting (AUC uniformly distributed or continuously rising, such as organic toxins, showing continuous decline without reaching a plateau). A dual-threshold judgment system is set up: the instantaneous inhibition rate threshold (IR0>4.00%) is used for initial screening, and the cumulative toxic load threshold (AUC>480%·h) is used for final judgment.

[0049] Cross-validation of the two methods generates an "acute trigger-chronic accumulation" composite grading: When IR0≤4.00% but AUC≤480%·h, it is judged as "low risk" and long-term monitoring and early warning are not triggered; When IR0 ≤ 4.00% but AUC > 480%·h, it is judged as "medium risk" and triggers long-term monitoring and early warning. When IR0 > 4.00% but AUC ≤ 480%·h, it is judged as "high risk".

[0050] When IR0 > 4.00% and AUC > 480%·h, it is judged as "extremely high risk"; For example, taking a certain organically polluted wastewater as an example, the instantaneous inhibition rate IR0 after 48 hours of exposure is ≥35% (below the toxicity threshold of 48%). However, the time-effect curve shows that the inhibition rate continuously increases from 5% at the 4th hour to 35% at the 48th hour, and the calculated AUC value is 520%·h > 480%·h, triggering an "extremely high" warning. If only the instantaneous inhibition rate is relied upon, the wastewater would be judged as "toxic," leading to a missed detection of chronic ecological risk. This model effectively compensates for the deficiency of single endpoint detection in missing chronic toxicity.

[0051] Based on the above examples, in this embodiment of the invention, after the wastewater sample was filtered through a 0.22 μm aqueous filter membrane, the ammonia nitrogen concentration of the sample was measured to be 18.50 mg / L, which is less than the limit of 30 mg / L, and the sample did not require dilution. Using this sample, *Crescentia esculenta* cells in the logarithmic growth phase were exposed to the toxic substance for 24 h. The FDA fluorescence staining method was used to determine the algal cell viability index, and the calculated value was f1 = 41.70%, corresponding to an initial inhibition rate IR0 = 58.30%.

[0052] During the exposure and poisoning process, the chlorophyll fluorescence induction kinetic curve was collected, and the characteristic parameters were extracted and input into the random forest classification model. The model output showed that the toxicity type was "heavy metal poison", and the confidence level was 87%. A multi-algae species combined detection plate (including Selenastrum capricornutum, Chlorella vulgaris, Scenedesmus quadricauda, and Dunaliella salina) was used for parallel exposure and poisoning to measure the cell viability index values of each algae species, construct a toxicity effect spectrum vector, and analyze the toxicity contribution of each component as follows: heavy metals contributed approximately 72%, organic pollutants contributed approximately 14%, ammonia nitrogen contributed approximately 8%, and inorganic salts contributed approximately 6%. The heavy metal with the highest contribution rate was used as the final toxic factor.

[0053] (a)Two-dimensional determination of effect and recovery: The initial inhibition rate IR0 at the exposure endpoint was 58.30% (belonging to the "high toxicity" range, 52% < IR0 ≤ 96%). The algae cells after exposure and poisoning were transferred to fresh medium for recovery culture for 24 h, and the inhibition rate IR 24 after recovery was measured again to be 12.50%. The irreversible damage index IDI was calculated as IDI = 58.30% - 12.50% = 45.80% > 30%, and it was determined as "irreversible damage". Two-dimensional matrix cross-positioning: "High toxicity" on the vertical axis × "irreversible" on the horizontal axis = toxicity level XII (the highest).

[0054] (b)Species sensitivity weight correction: According to the identified toxicity type of "heavy metals" above, the weight coefficients were retrieved from the sensitivity database: Chlorella vulgaris 1.15, Scenedesmus quadricauda 0.92, Dunaliella salina 0.78. Based on the measured inhibition rate of 58.30% of Selenastrum capricornutum, the equivalent inhibition rates of each algae species were calculated and none triggered an increase in the level, confirming the reliability of the high toxicity determination.

[0055] (c)Cumulative toxicity load model: Eight time points (0, 4, 8, 12, 20, 30, 40, 48 h) were set during the exposure period to continuously measure the inhibition rate and construct a time-effect curve. The calculated AUC = 520%·h > 480%·h, and it was determined as "extremely high risk".

[0056] Comprehensive determination result: The toxicity level of this water sample is "high toxicity - irreversible" (level XII), the ecological risk level is extremely high risk, the exposure mode is rapid action type (heavy metal characteristics), and the main toxic factor is heavy metal (contribution rate 72%). It is recommended to immediately take emergency control measures.

[0057] Example 2: In the embodiment of the present invention, the application object is an anaerobic digestion effluent sample from a certain livestock and poultry breeding wastewater treatment plant; The measured ammonia nitrogen concentration was 215 mg / L > 30 mg / L, and it was diluted 8 times with pure water to an ammonia nitrogen concentration of 26.9 mg / L. The diluted water sample was used for exposure and toxicity testing for 24 h. The cell viability index value of Scenedesmus obliquus f1 = 52.30% (corresponding to the initial inhibition rate IR0 = 47.70%, belonging to the "toxic" range). The chlorophyll fluorescence kinetic characteristic parameters were input into the random forest model, and the output toxicity type was "inorganic salts (mainly ammonia nitrogen)", with a confidence level of 92%. The multi-algae combination detection plate was used to analyze the toxicity contribution: ammonia nitrogen contributed approximately 68%, organic matter contributed approximately 18%, heavy metals contributed approximately 10%, and others contributed approximately 4%.

[0058] Toxicity level determination: (a)Two-dimensional determination of effect - recovery: After 24 h of recovery culture, IR 24 = 32.50%, IDI = 47.70% - 32.50% = 15.20% (10% < IDI ≤ 30%, partially reversible). The two-dimensional matrix was initially positioned as "toxic - partially reversible".

[0059] (b)Species sensitivity weight correction (key step): According to the toxicity type identified by the machine learning method as "inorganic salts (ammonia nitrogen)", the sensitivity weight coefficient of Scenedesmus quadricauda to ammonia nitrogen was retrieved from the sensitivity database as 2.3 (i.e., the sensitivity of Scenedesmus quadricauda to ammonia nitrogen is 2.3 times that of Scenedesmus obliquus). The measured inhibition rate of Scenedesmus quadricauda was 68.50%, and the calculated standard species equivalent inhibition rate = 68.50% × 2.3 = 157.55%, far exceeding the toxicity threshold (> 96%), automatically triggering the grade up - adjustment mechanism. The toxicity level was upgraded from "toxic" to "highly toxic".

[0060] (c)Cumulative toxicity load model: AUC = 610%·h > 480%·h, determined as "extremely high risk".

[0061] Comprehensive determination result: After sensitivity weight correction, the toxicity level of this water sample is "highly toxic - partially reversible" (XI) level, and the ecological risk level is "extremely high risk". The main toxic factor is ammonia nitrogen (contribution rate 68%). If sensitivity correction is not performed (directly using the inhibition rate of Scenedesmus obliquus 47.70%), this water sample will be misjudged as "toxic".

[0062] Example 3: The application object of this example is the influent sample of a chemical industrial park sewage treatment plant.

[0063] The ammonia nitrogen concentration was measured to be 15.60 mg / L ≤ 30 mg / L, requiring no dilution. After 48 h of exposure to the original water sample, the cell viability index of *Crassula ovata* was f1 = 62.50% (corresponding to an initial inhibition rate IR0 = 37.50%, falling within the "moderately toxic" range). Chlorophyll fluorescence kinetics showed a continuous decline, failing to reach a plateau within 30 minutes. The random forest model output toxicity type as "organic toxin" with a confidence level of 91%. Analysis of the toxicity contribution from a multi-algal species combination detection panel showed: organic pollutants contributed approximately 78%, heavy metals approximately 12%, ammonia nitrogen approximately 5%, and inorganic salts approximately 5%. Comprehensive toxicity assessment: (a) Effect-recovery dual-dimensional assessment: IR was measured 24 h after recovery culture. 24 =28.30%, IDI=37.50%-28.30%=9.20% ≤ 10%, judged as "reversible damage". The two-dimensional matrix initially locates it as "toxic-reversible".

[0064] (b) Species sensitivity weight correction: The weight coefficients of organic toxins do not deviate significantly and no level adjustment is required.

[0065] (c) Cumulative toxicity load model: The inhibition rate was continuously measured at 8 time points during the exposure period: 8.5% at 4 h, 15.2% at 8 h, 22.8% at 12 h, 29.5% at 20 h, 34.2% at 30 h, 36.8% at 40 h, and 37.5% at 48 h. The calculated AUC = 495%·h > 480%·h, indicating that the cumulative load exceeds the standard, and the ecological risk level is "extremely high risk", triggering long-term monitoring and early warning.

[0066] Overall assessment results: The water sample's toxicity level is "moderate to reversible", but its ecological risk level is "extremely high risk" (requiring long-term monitoring). The main toxic factor is organic pollutants (contributing 78%).

Claims

1. A comprehensive toxicity assessment and source tracing method for wastewater based on multi-index responses of algal cells, characterized in that, Includes the following steps: The wastewater to be tested was added to an algal cell culture system for exposure and toxin treatment, and algal cell viability index values ​​were obtained at multiple exposure times. Based on machine learning models and / or multi-algae combination detection methods, the pollution toxicity type of the wastewater to be tested is determined, and pollution source tracing is carried out based on the pollution toxicity type; Based on the pollution toxicity type and algal cell viability index of the wastewater to be tested, a comprehensive toxicity evaluation system is used to evaluate the toxicity and risk of the wastewater to be tested. The comprehensive toxicity assessment system includes: evaluating the toxicity level of the wastewater to be tested based on the acute toxicity intensity and reversibility of damage to algal cells, and evaluating the risk level of the wastewater to be tested based on the instantaneous inhibition rate and cumulative toxicity load of algal cells.

2. The wastewater comprehensive toxicity evaluation and source tracing method based on multi-index response of algal cells as described in claim 1, characterized in that, The ammonia nitrogen concentration of the wastewater to be tested is ≤30 mg / L; the algal cell viability index value is determined and calculated by FDA / PI dual fluorescence staining method.

3. The wastewater comprehensive toxicity evaluation and source tracing method based on multi-index response of algal cells as described in claim 1, characterized in that, The toxicity type of the wastewater to be tested is determined based on a machine learning model, including: Chlorophyll fluorescence induction kinetic curves of algal cells during exposure to toxins were collected, and characteristic parameters in the chlorophyll fluorescence induction kinetic curves were extracted. The feature parameters are input into the machine learning model, and the pollution toxicity type of the wastewater to be tested is output. The characteristic parameters include at least one of initial fluorescence intensity, maximum fluorescence intensity, fluorescence decline rate, steady-state time, and half-inhibition time; the toxicity type of the wastewater to be tested is any one of heavy metals, organic toxins, surfactants, or inorganic salts. The machine learning model is a random forest classification model.

4. The wastewater comprehensive toxicity evaluation and source tracing method based on multi-index response of algal cells as described in claim 1, characterized in that, The pollution toxicity type of the wastewater to be tested was determined based on a multi-algal combination detection method, including: Multiple sensitive algal species were selected and combined with the algal cells to form a multi-algal species detection plate; The wastewater to be tested was subjected to parallel exposure to the multi-algae detection plate for 20-48 hours; then, based on the pattern recognition analysis method, the toxicity contribution rate of each type of pollution was obtained. The type of pollution with the highest contribution rate is taken as the type of pollution toxicity of the wastewater to be tested.

5. The wastewater comprehensive toxicity evaluation and source tracing method based on multi-index response of algal cells as described in claim 1, characterized in that, Based on machine learning models and multi-algal species combination detection methods, the pollution toxicity types of the wastewater to be tested are determined, including: The pollution toxicity type of the wastewater under test was output by using machine learning models and multi-algae combination detection methods, respectively. When the output of the machine learning model is consistent with the output of the multi-algae combination detection method, the pollution toxicity type of the wastewater to be tested is confirmed. When the output of the machine learning model is inconsistent with the output of the multi-algae combination detection method, the parallel experiment of the multi-algae combination detection method is reset or the detection conditions are adjusted according to the output of the machine learning model. The multi-algae combination detection method is then tested again, and the output of the multi-algae combination detection method is taken as the pollution toxicity type of the wastewater to be tested.

6. The wastewater comprehensive toxicity evaluation and source tracing method based on multi-index response of algal cells as described in claim 1, characterized in that, The pollution source tracing based on pollution toxicity type includes: A toxicity type-characteristic pollutant mapping library is established, which includes pollutants and emission sources corresponding to heavy metals, pollutants and emission sources corresponding to organic toxins, pollutants and emission sources corresponding to surfactants, and pollutants and emission sources corresponding to inorganic salts. The determined pollution toxicity type is matched with the mapping library to identify characteristic pollutants and emission sources, which are then identified as the pollution sources of the wastewater to be tested.

7. The wastewater comprehensive toxicity evaluation and source tracing method based on multi-index response of algal cells as described in claim 1, characterized in that, The toxicity and risk of the wastewater under test are evaluated using a comprehensive toxicity assessment system based on the type of pollution toxicity and algal cell viability index values; including: The toxicity level of the wastewater to be tested is determined based on the type of pollution toxicity, acute toxicity intensity, and reversibility of damage. The risk level of the wastewater to be tested is determined based on the algal cell viability index, the instantaneous inhibition rate of algal cells, and the cumulative toxicity load.

8. The wastewater comprehensive toxicity evaluation and source tracing method based on multi-index response of algal cells as described in claim 7, characterized in that, The determination of the toxicity level of the wastewater to be tested based on its pollution toxicity type, acute toxicity intensity, and reversibility of damage includes: A multi-level toxicity rating is generated using a two-dimensional matrix of acute toxicity intensity and damage reversibility of algal cells. The acute toxicity intensity is determined based on the initial inhibition rate of the algal cells, and the damage reversibility is the difference between the initial inhibition rate and the post-recovery inhibition rate of the algal cells. The post-recovery inhibition rate is the inhibition rate measured after the algal cells exposed to the toxicity are transferred to a standard culture medium without the wastewater to be tested and cultured for 24 hours. Obtain EC values ​​for the toxicity of various algae to each type of pollution. 50 ratio; Based on the type of pollution toxicity of the wastewater to be tested and the species of algae cells, the corresponding EC is selected. 50 The ratio is used as the sensitivity weighting coefficient of the wastewater to be tested; the sensitivity weighting coefficient is multiplied by the initial inhibition rate of algal cells to obtain the equivalent inhibition rate; The toxicity level of the wastewater to be tested is determined by comparing the equivalent inhibition rate with the two-dimensional matrix.

9. The wastewater comprehensive toxicity evaluation and source tracing method based on multi-index response of algal cells as described in claim 8, characterized in that, The risk level of the wastewater to be tested is determined based on the algal cell viability index, the instantaneous inhibition rate of algal cells, and the cumulative toxicity load, including: Based on the algal cell viability index values ​​under multiple exposure times, the instantaneous inhibition rate of algal cells under multiple times was calculated using formula (1). （1）； In equation (1), denoted as , where is the instantaneous inhibition rate of algal cells; a represents the viability index value of algal cells on standard culture medium; b represents the viability index value of algal cells exposed to the toxin. Construct the time-inhibition rate curve and calculate the area under the curve of the time-inhibition rate curve; The cumulative toxicity load of the wastewater to be tested is determined based on the time distribution of the area under the curve. The risk level of the wastewater to be tested is determined based on the initial inhibition rate and cumulative toxicity load; the risk level of the wastewater to be tested includes low risk, medium risk, high risk and very high risk.

Images