Assessment of water quality risk

Abstract

Provided herein are systems and methods that include, for example, using historical water data to advantageously assess environmental risk to a selected location or area at a present or future time created by the presence of one or more pollutants or other materials. In some aspects, this comprises receiving information regarding the selected location or area, receiving a first type of historical water quality data associated with the selected location or area, receiving a second type of historical water quality data associated with the selected location or area, wherein the second type of historical water quality data is different than the first type of historical water quality data, and providing an indication of environmental risk for a selected time or time period based on, at least in part, the first type of historical water quality data and the second type of historical water quality data.

Description

ASSESSMENT OF WATER QUALITY RISKCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application 63 / 668737. filed July 8, 2024, the entire contents of which are incorporated by reference herein.BACKGROUND

[0002] Knowledge of water quality' can be important to the health of humans, animals, and plants dependent on that water, whether the water is for drinking, manufacturing, farming and the like. For many industries, water quality can have significant consequences for public and private health of humans and the broader ecosystem. Accordingly, water quality risk modeling, monitoring and proactive warning technologies are highly desirable or even necessary' to offset environmental risk, irrespective of the source of water (e.g., lake, river, sea or ocean, wastewater stream, drinking water source, etc ).SUMMARY

[0003] To date, water risk modeling has been primarily focused on the topic of water availability and water usage. Although this is appropriate and helpful for a wide variety of industrial users and instances (e.g., mining, bottling and beverage industries, commercial farms and agriculture more generally, other industries where the primary focus is the availability' of water, etc.). However, the topic of w ater quality risk modeling has received far less attention and can be helpful to fill a much-needed gap. Several embodiments disclosed herein advantageously provide improved technologies related to understanding and / or addressing water quality' risk and related concepts. The technologies disclosed herein are configured to provide to a user a cumulative risk of an occurrence of a particular adverse event (e.g., an anomalous event at a particular location and / or time or time period (e.g., present time, future time)). In some embodiments, such an event (e.g., an anomalous or adverse event or occurrence) includes, but is not limited to. the likelihood or risk related to industrial pollution, a violation of a law, rule and / or other threshold or limit (e.g., governmental, private, selfimposed, etc.), risk to a particular geographical area from one or more contaminants of interest (e.g., lead, PFAS, other forever chemicals, etc.).

[0004] According to some embodiments, a method of assessing environmental risk to a selected location or area comprises receiving information regarding the selected locationor area, receiving a first type of historical water quality data associated with the selected location or area, receiving a second type of historical water quality data associated with the selected location or area, wherein the second type of historical water uality data is different than the first type of historical water quality data, and providing an indication of environmental risk for a selected time or time period based on, at least in part, the first type of historical water quality data and the second type of historical water quality data, wherein environmental risk is related to risk created by at least one contaminant, and wherein the first type of historical water quality data comprises data related to a violation of a regulatory threshold.

[0005] According to some embodiments, a method of assessing environmental risk to a selected location or area comprises or consists essentially of receiving information regarding the selected location or area, receiving a first type of historical water quality data associated with the selected location or area, receiving a second type of historical water quality data associated with the selected location or area, wherein the second type of historical water quality data is different than the first type of historical water quality data, and providing an indication of environmental risk for a selected time or time period based on, at least in part, the first type of historical water quality data and the second type of historical water quality data.

[0006] According to some embodiments, the method further comprises or consists essentially of providing instructions to perform at least one action, wherein environmental risk is related to risk created by at least one contaminant, wherein the at least one action is configured to environmentally impact the selected location or area, wherein the first type of historical water quality data comprises data related to a violation of a regulatory threshold, and wherein providing an indication of environmental risk comprises providing a score.

[0007] According to some embodiments, the at least one action comprises or consists essentially of manipulating at least one flow control device (e.g., valve, other flow control device, system or component, etc.). In some embodiments, the at least one action comprises performing at least partial treatment on a fluid (e.g., chemical additives, pH control, biological treatment, chemical treatment, thermal treatment, diversion, storage, etc.).

[0008] According to some embodiments, the second type of historical water quality data comprises or consists essentially of actual measured data of at least one contaminant. In some embodiments, the first type of historical water quality data comprises data related to a violation of a regulatory7threshold (e.g., US EPA standard, other regulatory7standard or rule, etc.).

[0009] According to some embodiments, each of the first type and the second type of historical water quality data comprises or consisting essentially of at least one of: datarelated to violations of a law, a rule a permit or another instrument that includes a threshold or limit, concentration data for at least one contaminant of interest; and data related to water quality relevant to the selected location or area; data related to water flow and distribution relevant to the selected location or area.

[0010] According to some embodiments, receiving information regarding the selected location or area comprises receiving geographical data related to an area of a map. In some arrangements, the map comprises an interactive map selected by a user.

[0011] According to some embodiments, the first type of historical water quality data or the second type of historical water quality data comprises data stored on a database (e.g., a government database). In some embodiments, at least one of the first type and the second type of historical water quality data comprises data obtained and / or maintained by a governmental agency or body.

[0012] According to some embodiments, environmental risk is related to risk created by at least one contaminant. In some embodiments, the at least one contaminant comprises a forever chemical (e.g., PF AS. PFOS, etc.). In some embodiments, the at least one contaminant comprises lead.

[0013] According to some embodiments, providing an indication of environmental risk comprises providing a score (e.g., an alphanumeric score, a score ranging from 0 to 1, 0 to 10 or 0 to 100, a rating, etc.).

[0014] According to some embodiments, providing an indication of environmental risk comprises providing a graphical output. In some embodiments, the graphical output comprises a darkness or color scale.

[0015] According to some embodiments, the selected time or time period is a present time. In other arrangements, the selected time or time period is a future time or a future time period.

[0016] According to some embodiments, the method further comprises providing instructions to perform at least one action. In some embodiments, the at least one action comprises manipulating at least one flow control device (e.g., valve, other flow control device, an inlet or an outlet, etc.). According to some embodiments, the at least one action comprises performing at least partial treatment on a fluid. In some embodiments, the at least one action comprises providing an alert.

[0017] According to some embodiments, a method of assessing environmental risk to a selected location or area is provided, the method comprising or consisting essentially of receiving a first type of historical water quality data associated with the selected location orarea, receiving a second type of historical water quality data associated with the selected location or area, herein the second type of historical water quality’ data is different than the second type of historical water quality data, and providing an indication of environmental risk for a selected time or time period based on, at least in part, the first type of historical water quality data and the second ty pe of historical water quality' data.

[0018] According to some embodiments, at least one of the first type and the second type of historical water quality data comprises or consists essentially of data related to violations of a law, a rule a permit or another instrument that includes a threshold or limit, concentration data for at least one contaminant of interest; and data related to water quality relevant to the selected location or area; data related to water flow and distribution relevant to the selected location or area and / or the like.

[0019] According to some embodiments, the at least one contaminant of interest comprises PF AS and / or another forever chemical. In some embodiments, the at least one contaminant of interest comprises lead.

[0020] According to some embodiments, at least one of the first type and the second ty pe of historical water quality data comprises data obtained and / or maintained by a governmental agency or body.

[0021] According to some embodiments, providing an indication of environmental risk comprises providing a score (e.g.. an alphanumerical score (e.g., 0 to 1, 0 to 10, 0 to 100, A to Z. A to F. etc.), a graphical output (e.g., a darkness or color scale).

[0022] According to some embodiments, the selected time or time period is a present time. In some embodiments, the selected time or time period is a future time or a future time period.

[0023] According to some embodiments, the method further comprises or consists essentially of identifying one or more actions based on the indication of environmental risk. In some embodiments, the one or more actions comprises an action to mitigate or lower environmental risk.

[0024] According to some embodiments, the method further comprises or consists essentially of instructing the one or more actions based on identifying one or more actions. In some embodiments, the one or more actions comprises at least one of the following: informing (e.g., automatically) one or more individual or entities, closing a valve, adding at least one chemical to a fluid source, requesting assistance from an agency or another entity, diverting a fluid, and treating a fluid.

[0025] According to some embodiments, historical data (e g., water quality, contaminant data, key violation data, etc.) are used to create mathematical models for assessing, predicting and / or otherwise evaluating water quality risk. In some embodiments, said historical data are indexed by time and / or location.

[0026] According to some embodiments, a water quality anomaly detection system comprising or consisting essentially of data processing hardware, and memory in communication with the data processing hardware, the memory storing instructions, wherein execution of the instructions by the data processing hardware causes the data processing hardware to: obtain an input indicating a first location, in response to obtaining the input, identity' historical water data associated with one or more second locations and timing data, wherein the one or more second locations are associated with the first location, wherein the historical water data is associated with a plurality of first data formats, normalize the historical water data to obtain normalized water data, wherein the normalized water data is associated with a second data format, provide the normalized water data to a model, wherein the model is configured to predict a likelihood of occurrence of one or more anomalous water quality events, detect an anomalous water quality event associated with water located at the first location based on an output of the model; and

[0027] provide, to a computing system, an indication of a likelihood of occurrence of the anomalous w ater quality event and an indication of at least a portion of the first location associated with the anomalous w ater quality event based on the output.

[0028] According to some embodiments, anon-transitory computer readable media comprising or consisting essentially of computer-executable instructions, w herein execution of the computer-executable instructions by a first computing system causes the first computing system to: obtain an input indicating a first location, in response to obtaining the input, identify historical water data associated with one or more second locations and timing data, wherein the one or more second locations are associated with the first location, wherein the historical w ater data is associated w ith a plurality' of first data formats, normalize the historical w ater data to obtain normalized water data, wherein the normalized water data is associated with a second data format, provide the normalized water data to a model, wherein the model is configured to predict a likelihood of occurrence of one or more anomalous water qualify events, detect an anomalous water quality' event associated with w ater located at the first location based on an output of the model, and provide, to a second computing system, an indication of a likelihood of occurrence of the anomalous water qualify event and an indication of at least a portion of the first location associated with the anomalous water quality event based on the output.

[0029] According to some embodiments, a computer-implemented method comprising or consisting essentially of obtaining, from one or more data sources, raw water data, providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events, receiving an output from the model, wherein the output is indicative of an adverse water quality event, and instructing actuation of a valve based on the output.

[0030] According to some embodiments, a computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data, providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events, receiving an output from the model, wherein the output is indicative of an adverse water quality event, dynamically generating a user interface based on the output, and instructing display of the user interface via a user computing device.

[0031] According to some embodiments, a computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data, providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events, receiving an output from the model, wherein the output is indicative of an adverse water quality event, dynamically generating audio data based on the output, wherein the audio data indicates the adverse water quality event, and instructing output of the audio data via an audio output device.

[0032] According to some embodiments, a computer-implemented method comprising or consisting essentially of obtaining, from one or more data sources, raw water data, providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events, receiving an output from the model, wherein the output is indicative of an adverse water quality event, dynamically generating an alert based on the output, wherein the alert indicates one or more of a predicted timing, a predicted location, a predicted severity, a predicted adverse water quality event type, or a predicted effect of the adverse water quality event, and instructing output of the alert.

[0033] According to some embodiments, a computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data, providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events, receiving an output from the model, wherein the output is indicative of the one or more adverse water quality events, wherein each of the one or more adverse water quality events is associated with a respective type of adverse water quality event, generating one or more probabilistic vectors within a multidimensionalprobabilistic vector space, wherein each of the one or more probabilistic vectors corresponds to a respective adverse water quality event of the one or more adverse water quality events, and providing the one or more probabilistic vectors to a computing system.

[0034] According to some embodiments, a computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data, providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events, receiving an output from the model, wherein the output is indicative of an adverse water quality event, generating computer-executable instructions based on the output, and providing the computer-executable instructions to a computing system.

[0035] According to some embodiments, a computer-implemented method comprising or consisting essentially of: obtaining, from a user computing device, an input indicating a first location, in response to obtaining the input, identifying historical water data associated with one or more second locations, providing an input based on the historical water data to a model, wherein the model is configured to predict a likelihood of occurrence of one or more future anomalous water quality events, based on an output of the model, determining a likelihood of occurrence of a future anomalous water quality event associated with water located at the first location exceeds, matches, or is within a threshold, identifying one or more actions to reduce the likelihood of the occurrence of the future anomalous water quality event, and providing, to the user computing device, one or more identifiers of the one or more actions.

[0036] According to some embodiments, a method of assessing environmental risk to a selected location or area comprises receiving information regarding the selected location or area, receiving a first type of historical water quality data associated with the selected location or area, receiving a second type of historical water quality data associated with the selected location or area, wherein the second type of historical water qualify data is different than the first type of historical water quality data, and providing an indication of environmental risk for a selected time or time period based on, at least in part, the first type of historical water quality data and the second type of historical water qualify data, wherein environmental risk is related to risk created by at least one contaminant, and wherein the first type of historical water qualify data comprises data related to a violation of a regulatory threshold. The method further comprises providing instructions to perform at least one action, wherein the at least one action is configured to environmentally impact the selected location or area, wherein providing an indication of environmental risk comprises providing a score, and wherein the second type of historical water qualify data comprises actual measured data of at least one contaminant.BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Several embodiments of various inventive features will now be described with reference to the following drawings. Throughout the drawings, reference numbers may be re-used to indicate correspondence between referenced elements. The drawings are provided to illustrate example embodiments described herein and are not intended to limit the scope of the disclosure. Features from one figure can be combined with features of other figures.

[0038] FIG. 1 depicts an example operating environment for detection of anomalous water quality events.

[0039] FIG. 2A to FIG. 2D each depicts a schematic view of an example interface provided by a water quality anomaly detection system according to some embodiments.

[0040] FIG. 3 is a flow-chart of an example arrangement of operations for detecting anomalous water quality events.

[0041] FIG. 4 shows an example computing device that may be used to implement aspects of the present disclosure.DETAILED DESCRIPTION

[0042] Water risk modeling has been primarily focused on the topic of water availability and water usage. This may be appropriate and helpful for a wide variety of industrial users and instances (e.g., mining, bottling and beverage industries, commercial farms and agriculture more generally, other industries where the primary focus is the availability of water, etc.). On the other hand, water quality risk modeling has received far less attention and can be helpful to fill a much-needed gap. The various embodiments disclosed herein and equivalents thereof advantageously provide improved technologies related to understanding and / or addressing water quality risk and related concepts. In some embodiments, this includes, among other things, providing a water quality risk score, value, categorization or other quantitative and / or qualitative assessment.

[0043] The embodiments disclosed herein provide one or more benefits or advantages in connection with w ater monitoring (e.g., w ater quality monitoring), including, for example and without limitation, one or more of the following: determining and assigning a numeric or comparative value or other measure of water quality risk; determining risk using, at least in part, historical water quality data; predicting adverse or potentially adverse water quality events and / or conditions, currently and / or in the future; assessing (e.g., modelling,predicting, etc.) water quality risk as a function of both location (e.g., geospatial location, specific location, specific area, etc.) and time; providing information (e.g., possible solutions) related to water quality risk mitigation; and / or the like.

[0044] According to some embodiments, an output of the various systems and other technologies disclosed herein provide water quality information (e.g., one or more output(s)) that is relevant and actionable to users or other recipients of such information. In some embodiments, the information or other output(s) comprises or consists essentially of one or more of the following characteristics: it informs users about an impact on health (e.g., of humans, of an area (e.g., environmentally), etc.; it provides a spatio-temporal context (e.g., what an average measurement of a particular contaminant or plurality of contaminants (e.g., lead, PFAS, over a 10-year period, over another desired time period, etc.) for a particular location or area; it compares other locations (e.g., nearby locations) to provide an score (e.g., an equivalent water quality score); it provides a score, index, numeric value or other quantitative or comparative value that is understandable and usable by consumers; it informs consumers if a measurement is part of a larger trend (e.g.. of ongoing (e.g.. contemporaneous, at the instant or other specific time, etc.) increases or decreases in concentration of one or more contaminants; it is part of a probabilistic framework (e g., allows for probabilities of occurrence of a particular adverse event to be informed by historical occurrences of such an event; and / or the like.

[0045] According to some embodiments, the technologies disclosed herein are configured to provide to a user a cumulative risk of an occurrence of a particular adverse event (e.g., an anomalous event at a particular location and / or time or time period (e.g., present time, future time)). In some embodiments, such an event (e.g., an anomalous or adverse event or occurrence) includes, but is not limited to. the likelihood or risk related to industrial pollution, a violation of a law, rule and / or other threshold or limit (e.g., governmental, private, selfimposed, etc.), risk to a particular geographical area from one or more contaminants of interest (e.g., lead, PFAS, other forever chemicals, etc.).

[0046] By way of example, with respect to possible current or future violations, the various embodiments disclosed herein can provide a probability of occurrence of a National Pollutant Discharge Elimination System ( ‘NPDES”) permit violation for a particular area or location based on, at least in part, historical base probability of occurrence of a violation at any monitored location proximate to this location (e.g., using actual violation data, other data collected by governmental or other entity, etc.). In other configurations, the embodiments disclosed herein are configured to provide a probability of occurrence of a Safe Drinking WaterAct (“SDWA”) violation (e.g., related to a public water supply) in the vicinity of the area or location of interest based on historical occurrences of violation in this vicinity. In some embodiments, the concepts disclosed herein are configurated to provide a probability of occurrence of violations of two or more (e.g., 2, 3, 4, more than 4, etc.) rules, thresholds, standards and / or other limits.

[0047] In other embodiments, the systems and other technologies disclosed herein are configured to provide a probability, score and / or other output of having contaminated water or other fluid at a given location due to the proximity of Toxics Release Inventory (“TRI”) and / or Superfund sites in the vicinity of this location. In some embodiments, the systems and other technologies disclosed herein are configured to provide a probability, score and / or other output related to non-NPDES and / or non-SDWA wastewater or other effluent discharge events or occurrences. This can be helpful in instances where groundwater-related usage is not monitored or poorly monitored by the U.S. Environmental Protection Agency (“EP A”), other regulatory7entity and / or other governmental agency. As noted, the systems and other technologies disclosed herein are configured to provide a probability, score and / or other output related to risk posed by one or more target contaminants (e.g.. PFAS / PFOS, lead, etc.) for a particular geographic location or area.

[0048] According to some embodiments, the systems and other technologies disclosed herein are configured to incorporate two or more risk factors (e.g., based on historical data) as probabilistic vectors in a multidimensional probabilistic vector space with each vector representing the probability of occurrence of a particular type of adverse or other anomalous event. In some embodiments, the magnitudes (e.g., predicted, calculated, modelled, etc.) of the individual vectors are adjusted (e.g., normalized) so that the cumulative probability of occurrence of an adverse event has the usual interpretation on a particular scale (e.g., 0 to 1, 0 to 10, 0 to 100, A to Z, A to F, other alphanumeric scale, color scale, normalized or unnormalized, etc.).

[0049] According to some embodiments, the systems and other technologies disclosed herein are configured to provide a cumulative risk score or other output as, at least in part, a simple aggregate (e.g., mean or average, median, mode, etc.) of individual elements.

[0050] According to some embodiments, the systems and other technologies disclosed herein are configured to predict environmental quality risk. According to some embodiments, the systems and other technologies disclosed herein are configured to provide a score or other output (e.g., a cumulative risk score) is normalized to have a human readable “rating.” For instance and without restriction or limitation, for a scale of 0 to 10, “10” canindicate high risk of the occurrence of a particular adverse or anomalous event, whereas “0” can indicate no or very low risk. Thus, in some embodiments, the score or other output provided by the systems and other technologies disclosed herein provide value in a relative sense (e.g., a first area is at higher risk of PF AS contamination than a second area being considered by a user for farming, construction or other development or other use).

[0051] According to some embodiments, the predictions (e.g.. scores, other outputs, etc.) generated by the systems and other technologies disclosed herein can be indexed to a specific area or location (e.g., fixed a location), but allow for modelling over time. For example, with reference to a particular processing plant, the systems and other technologies disclosed herein can be configured to set the location to the specific geographical location of the plant (e.g.. via a latitude / longitude coordinate combination set at the areal center of the plant), and is configured to predict risk probabilities as a continuous function of time.

[0052] According to some embodiments, predictions, forecasts and / or other outputs using the systems and other technologies disclosed herein activate one or more annunciators (e.g., alarms, alerts, etc.) or other outputs (e.g., actual, virtual, etc.) that inform users of certain risk forecasts (e.g., high-risk forecasts) at a future time or in a future prediction interval. In some embodiments, by way of example, predictions, forecasts and / or other outputs using the systems and other technologies disclosed herein from PRISM actuate SCADA systems that automatically take action to mitigate the effects of a high-risk water quality event.

[0053] According to some embodiments, the systems and other technologies disclosed herein are configured to and / or generate and deliver an output (e.g., a detailed report, any other report or output, etc.). Such a report or output can comprise historical water quality information, forecasts for water quality risk and / or any other information / data. Such reports can be customized to provide data selected by and / or for a particular user, as desired or required. In some embodiments, such a report or output can be downloaded (e.g., via a website, email, REST API, another API, etc.).

[0054] According to some embodiments, the systems and other technologies disclosed herein allow users to select geographical areas of their preference dynamically via a web-application. For example, the systems can be configured to allow a user to select an area on a map or other graphical representation of a desired area. In some embodiments, this includes dropdown and / or other interactive menus that allow users to monitor risk (e.g., county level water quality risk). In other embodiments, the systems and other technologies disclosed herein comprise and / or are configured to communicate and work together with a Graphical User Interface (GUI). In some embodiments, a GUI can be configured to permit users to sketcharbitrary geographical regions of interest (e.g., circles, rectangles, polygons, irregular areas, etc.) that closely align with the location or area of interest.

[0055] According to some embodiments, the systems and other technologies disclosed herein comprise, incorporate and / or are configured to be used in connection with dynamic GUIs. This can advantageously allow users to query this information (e.g., using Artificial Intelligence chatbots).

[0056] According to some embodiments, the systems and other technologies disclosed herein are configured to enable or permit users to assess risk (e.g., obtain a risk assessment score or rating) through the simple provisioning of historical water quality risk assessments in a Data as a Service (DaaS) arrangement.

[0057] FIG. 1 is a block diagram depicting an example operating environment 100 for detection of certain water quality’ events (e.g., anomalous or adverse water quality events). The example operating environment 100 illustrated in FIG. 1 includes a water quality anomaly detection system 102, a computing device 106, a control system 110 (e.g., an automated control system), at least one component 112, a computing system 114, and a sensor system 118. The water quality anomaly detection system 102 may obtain water data (e.g., historical water data, raw water data, historical raw water data, historical violation data, etc.) and detect (e g., predict) anomalous water quality events by modelling a quality of the water based on the water data. For example, the water quality anomaly detection system 102 may predict whether a particular water quality event (e.g., an anomalous water quality event) will occur in the future and / or is currently occurring (or provide a probability’ of the occurrence of such an event in the future).

[0058] In some embodiments, the ater data may include water quality data (e.g., indicating a quality of water), contaminant data (e.g., indicating contaminants within the water), violation data (e.g.. indicating violations of one or more acts, laws, thresholds or limits, etc.), etc. The water quality data may indicate a quality of the water (e.g., a numerical score of the quality of the water), a timing of the measurement of the quality’ of the water, a location of the water, how the quality of the water is being measured (e.g., the scale), etc. By way of example, the contaminant data may indicate a level (e.g., a percentage, a likelihood, etc.) of contaminants within the water, a timing of the contaminants being located within the water, a location of the contaminants, a type of the contaminants, etc. The violation data may indicate, for example and without limitation, a timing of the violations, a location of the violations, a type of the violations, a frequency of the violations, etc.

[0059] In some embodiments, the water quality anomaly detection system (e.g., w ater quality risk detection system) 102 is configured to predict a deviation of the quality ofthe water (e.g., based on the water data) that is greater than, is less than, is within, or matches or substantially matches a threshold (e.g.. a threshold range, a threshold value, etc.) and may classify the deviation as an anomalous water qualify event (e.g., a predicted future anomalous water qualify' event). While reference may be made to herein of an anomalous water qualify event, it will be understood that the water qualify detection system 102 may be a general detection system that may detect different events (e.g., an environment quality event).

[0060] The anomalous water qualify event may be or may include a pollution event (e.g., an NPDES permit violation), a violation event (e g., a SDWA violation, a TRI violation, an SEMS violation, etc.), a discharge event (e.g., a non-NPDES and / or non-SDWA related wastewater discharge event), a chemical release event (e.g., a forever chemical such as PFAS / PFOS release event), a combination of two or more such events and / or the like.

[0061] In some embodiments, the water qualify anomaly detection system (e.g., the water qualify' risk detection system) 102, the computing device 106, the control system 110, the computing system 114, and / or the sensor system 118 is configured to communicate (e.g., sending and receiving data packets) via a network 120 (e.g., a communication network). The network 120 may be any wired network, wireless network or combination thereof (e.g.. the Internet). In addition, the network 120 may be a personal area network, local area network, wide area network, cable network, satellite network, cellular telephone network, any other network and / or any combination thereof. For example, the network 120 may be a publicly accessible network of linked networks, possibly operated by various distinct parties, such as the Internet. In some embodiments, the network 120 may be a private or semi-private network, such as a corporate or universify intranet. The network 120 may include one or more wireless networks, such as a Global System for Mobile Communications (GSM) network, a Code Division Multiple Access (CDMA) network, a Long Term Evolution (LTE) network, or some other type of wireless network. The network 120 may use protocols and components for communicating via the Internet or any of the other aforementioned types of networks.

[0062] The water qualify anomaly detection system (e.g., the water qualify risk detection system) 102 may include data processing hardware and memory hardware (e.g., the data store 104). For example, in some embodiments, the water qualify anomaly detection system 102 is configured to store alerts, water data, output of a model 116A or 1 16B and / or other data or information in the data store 104. The water qualify anomaly detection system 102 may model the water qualify (e.g., may perform water qualify risk modeling), may detect anomalous water qualify events (e.g., based on water data), and may instruct performance of one or more actions based on the detected anomalous water qualify events. For example, thewater quality anomaly detection system (e.g., the water quality risk detection system) 102 may instruct performance of the one or more actions to mitigate the effects of the detected anomalous water quality events, to reduce the probability of occurrence of the anomalous water quality events, to alert a user of the anomalous water quality events, etc.

[0063] The water quality anomaly detection system (e.g., the water quality risk detection system) 102 may include and / or implement a model 116A (e g., a probabilistic model, a mathematical model, a Bayesian model, a trained machine learning model such as a neural network or generative model, etc.). For example, the model 116A may be a large language model (“LLM”). The water quality anomaly detection system 102 may implement a large language model such as Chat Generative Pre-trained Transformer ("ChatGPT"). Pathways Language Model ( ”PaLM"). Large Language Model Meta Artificial Intelligence C’LLaM A”), etc.

[0064] In some embodiments, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may generate, train, and / or update the model 116A. For example, the water quality anomaly detection system 102 may generate, train, and / or update the model based on historical water data (e.g., historical water quality events). In some embodiments, the water quality anomaly detection system 102 may not implement the model 116A. For example, the model 116A may be trained, at least in part, by an operator (e.g., user, administrator, etc.) of the water quality anomaly detection system 102 (such as, for example, by training the model in full, or finetuning a model previously trained for more general purposes than water quality analysis), but may be implemented or executed in production by a separate server or cloud computing platform that is accessible to the water quality anomaly detection system 102 via application programming interface (“API”) calls from the water quality anomaly detection system 102.

[0065] The water quality anomaly detection system (e.g., the water quality risk detection system) 102 may utilize the model 116A to predict a likelihood of occurrence of one or more anomalous water quality events. For example, the model 116A may be configured to (e.g., trained to) predict a likelihood of occurrence of one or more anomalous water quality events.

[0066] The computing device 106 (e.g., a user computing device) may be any of a number of computing devices that are capable of communicating over a network including, but not limited to, a laptop, personal computer, personal digital assistant (PDA), hybrid PDA / mobile phone, mobile phone, smartphone, any other device or system configured to interface with a local or web-based user platform, electronic book reader, digital mediaplayer,tablet computer, gaming console or controller, smart speaker (with or without a display screen), kiosk, augmented reality device, other wireless device, set-top or other television box, and the like. The computing device 106 may include a display (e.g., a screen) and may display an interface via the display. Further, the computing device 106 includes an application 108 (e.g., a native application, a software application, etc.) through which the interface may be displayed. The computing device 106 may provide input to the water quality anomaly detection system 102 (e.g.. a request to detect anomalous water quality events, water data, etc.) and may obtain output from the water quality anomaly detection system 102 (e g., indicating an anomalous water quality event).

[0067] The control system 110 may include various hardware for performing control functions. For example, the control system 110 may be a water level control system, a water flow control system, etc. In some cases, the control system 110 may be or may include a SCADA system. The control system 110 may include one or more controllers (e.g., a programable controller). In some cases, the control system 110 may include multiple control systems (e.g., water level control system, a water flow control system, a treatment control system and / or any other control system). In some cases, all or a portion of locations may be associated with a respective control system (e.g., a first location may be associated with a first control system, a second location may be associated with a second control system, etc.).

[0068] The control system 110 may be in communication with a component 112 (e.g.. a valve, a water release component, a chemical release component, another fluid treatment or modification component, etc.). The control system 1 10 may instruct performance (and / or may cause performance) of an action (e.g., closing, actuating, opening, moving, outputting light and / or audio) by the component 112 (e.g., based on an anomalous water quality event detected by the water quality anomaly detection system 102). In some cases, the control system 110 may be in communication with (and may instruct performance of an action by) a plurality of components (e.g., a plurality' of components associated with a particular location). For example, the control system 110 can operate (e.g., open, close, modulate, etc.) one or more valves or other flow control devices, can activate a treatment protocol (e.g., a chemical addition, another type of fluid treatment, etc.) and / or the like, as desired or required.

[0069] The computing system 114 may be located separately or remote from the water quality' anomaly detection system (e.g., the water quality risk detection system) 102. In some cases, the computing system 114 may be part of the water quality anomaly detection system 102. The computing system 114 may include and / or implement a model 116B (e.g.. a probabilistic model, a mathematical model, a Bayesian model, a machine learning model, etc.).For example, the model 116B may be similar to the model 116A. In some cases, the computing system 114 may not or does not implement any model (e.g., the model 116B).

[0070] The computing system 114 may utilize the model 116B to predict a likelihood of occurrence of one or more anomalous water quality events (e.g., a violation, an exceedance of a particular concentration or other quantitative threshold, etc.). For example, the model 116B may be configured to (e.g., trained to) predict a likelihood of occurrence of one or more anomalous water quality events. In some cases, the water quality anomaly detection system 102 may train and / or update the model 116B.

[0071] The water quality anomaly detection system (e.g., the water quality risk detection system) 102 may obtain water data from a plurality of data sources (e.g., a data store, the sensor system 118. a computing system, historical data database(s), etc.). In some cases, the water data may have a plurality of data formats, which can be customizable or otherwise modifiable by a user.

[0072] The sensor system 118 may include one or more sensors or other detectors (e.g., one or more sensors that obtain water data). For example, the one or more sensors may include flow sensors, depth sensors, temperature sensors, pressure sensors. pH sensors, chemical sensors, other concentration sensors, optical sensors, bacteria sensors, etc. The one or more sensors may be distributed across (e.g., located at) a plurality of locations (e.g., a plurality of physical locations). For example, a first subset of the one or more sensors may be located at a first location and may obtain water data associated with water located at the first location and a second subset of the one or more sensors may be located at a second location and may obtain water data associated with water located at the second location. Such sensors can be located at any desired locations within a particular area. For example, sensors can be positioned (e.g., spatially distributed) throughout a designated area of interest. In some embodiments, sensors can be located before and after (e.g., upstream and downstream) of one or more control devices or systems (e.g., flow control devices, treatment devices, etc.) that may be included in and / or configured to cooperate with the water quality risk detection system or other water quality anomaly detection system.

[0073] In some cases, the operating environment 100 may not include the sensor system 118. For example, the water quality anomaly detection system 102 may not obtain the water data from the sensor system 118.

[0074] In some cases, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may obtain the water data from a data store, a computing system, an internal or external database, etc. For example, the water quality anomaly detectionsystem 102 may access a data store, obtain data stored by the data store, and filter (e.g., parse) the data to identify filtered data associated with a particular location. In another example, the water quality anomaly detection system 102 may request water data associated with a particular location from a computing system.

[0075] In some cases, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may perform a web crawl, a web scrape, etc. and may obtain the water data associated with the particular location. For example, the water quality anomaly detection system 102 may perform a web crawl, a web scrape, etc. on one or more websites (e.g., government websites, regulatory websites, etc.). The water quality anomaly detection system 102 may include one or more components for performing the web crawl, the web scrape, etc. (e.g.. a web scraper, a web crawler, a web scraping extension, etc.).

[0076] The water quality anomaly detection system (e.g., the water quality risk detection system) 102 may identify one or more websites (e.g., trusted websites), may monitor the one or more websites, and may obtain (e.g., periodically or aperiodically) water data from the one or more websites or data sources (e.g., governmental or other databases). In some cases, the water quality anomaly detection system 102 may obtain the water data in response to a request (e.g., a request to monitor for and / or predict anomalous water quality' events).

[0077] As discussed herein, the water quality' anomaly detection system (e.g., the water quality risk detection system) 102 may obtain (e.g., receive) water data (e.g., historical data relating, directly or indirectly, to water quality). In some cases, the water quality anomaly detection system 102 may obtain the water data in response to an input from the computing device 106. The input may include a request to perform anomalous water quality event detection. In some cases, the input may include an identifier of a location for performance of the anomalous water quality event detection (e.g., a particular body of water, coordinates, a zip code, a neighborhood, a city, a state, a designated geographical area (e.g., as selected by a user), etc.). In some cases, the input may include an identifier of a timing for performance of the anomalous water quality event detection (e.g., predict whether an anomalous water quality event is to occur in the next week, continuously obtain and monitor water data and perform the anomalous water quality event detection for the next six hours, etc.). In some cases, the input may include an identifier of an operation (e g., a construction operation), a scope of the operation (e.g., a timing of the operation, a location of the operation, etc.).

[0078] In some embodiments, the water data may be time indexed and / or location indexed water data. For example, the water data may be associated with a time index and / or a location index. In some cases, the water quality’ anomaly detection system (e.g., the waterquality risk detection system) 102 may generate (or may be configured to generate) an index (e.g., a time index and / or a location index) for the water data. The water quality anomaly detection system 102 may utilize the index (e.g., the indexes) to filter the water data based on the input.

[0079] In some embodiments, the input may indicate a location and the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may filter and / or otherwise manipulate the water data based on the location using a location index to identify a filtered subset of water data associated with the location (e g., for performance of the anomalous water quality event detection). In some cases, the input may indicate a first location and the water quality anomaly detection system 102 may filter the water data based on the first location using a location index to identify’ a filtered subset of water data associated with the one or more second locations (e.g., for performance of the anomalous water quality event detection). For example, the one or more second locations may include or may not include the first location. In another example, the water quality anomaly detection system may determine the one or more second locations for filtering the water data based on one or more characteristics of the one or more second locations and the first location (e.g., the characteristics may indicate the one or more second locations and the first location are subsets of a larger location, share access to a body of water, a size, location, a historical water quality, etc. of the one or more second locations is within a threshold of a size, location, a historical water quality, etc. of the first location, etc.).

[0080] The water quality anomaly detection system (e.g., the water quality risk detection system) 102 may provide the water data (e.g., the filtered water data) to the model (e.g., the model 116A and / or the model 116B). The model may be trained and / or otherwise configured to output a likelihood (e.g., a probability) of occurrence of an anomalous water quality event (e.g., 75% likelihood of occurrence, another percentage or indication of occurrence, etc.). In some embodiments, the output may further include a timing of the anomalous water quality event (e.g., within the next five hours or other selected time period, 75% (or other percentage or indication of occurrence) likelihood of occurring within the next 12 hours or other selected time period, etc.), a severity of the anomalous water quality event, an effect (e.g., an impact) of the anomalous water quality event, and / or an adverse water quality event type of the anomalous water quality event (e.g., a contaminant violation), for example.

[0081] In some embodiments, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may (e.g.. may be configured to) process (e.g., perform prompt engineering on) the water data prior to providing the processed water data to the modelas an input. For example, the water data may be associated with multiple data formats and the water quality’ anomaly detection system 102 may normalize the water data such that the normalized water data is associated with one data format.

[0082] In some embodiments, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may not provide the water data directly to the model. Instead, for example, the water quality anomaly detection system 102 may process the water data to generate one or more tokens (e.g., token stnngs). The water quality anomaly detection system 102 may generate a file (e.g., a JSON file that includes the generated one or more tokens) and may provide the file to the model as an input. In some cases, the water quality anomaly detection system 102 may generate a file and may train the model using the file.

[0083] The water quality anomaly detection system (e.g., the water quality risk detection system) 102 may obtain an output of the model. The output may indicate a likelihood of occurrence of an anomalous water quality event, a timing of the anomalous water quality event, a severity7of the anomalous water quality7event, an effect of the anomalous water quality event, an adverse water quality event type of the anomalous water quality event and / or the like, as desired or required.

[0084] The water quality anomaly detection system (e.g., the water quality risk detection system) 102 may be configured to detect an anomalous water quality7event based on comparing the output of the model to a threshold or another value or range (e.g., to determine whether the likelihood of an anomalous water quality event exceeds or matches a threshold or other value). For example, based on comparing the output of the model to the threshold, the water quality7anomaly detection system 102 may determine the likelihood (e.g., a likelihood of occurrence, risk associated with such likelihood, etc.) of a first ty pe of an anomalous water quality event occurring with a first timing (e.g., in a first day. a first week, another designated time period, etc.) exceeds or matches a first threshold while the likelihood of a second type of an anomalous water quality7event occurring with a second timing (e.g., in a day) does not exceed or match a second threshold. In some cases, the threshold may be provided by (and / or defined by) the computing device 106.

[0085] In some embodiments, the water quality anomaly detection system (e.g.. the water quality7risk detection system) 102 may assign a value (e.g., a numerical value, a spatiotemporal value, etc.) to an anomalous water quality7event based on the output of the model. For example, the value may indicate a likelihood of occurrence of the anomalous water quality event (e.g.. a violation of a regulatory requirement, an exceedance of a particular concentration or other quantitative value of a specific contaminant or other property, etc.). In some cases,the value may be a function of time (e.g., a time series prediction of a likelihood of occurrence of the anomalous water quality’ event based on the water data, the location, etc.).

[0086] The water quality anomaly detection system (e.g., the water quality risk detection system) 102 may generate the value for assignment by aggregating a plurality of outputs of the model. For example, the water quality anomaly detection system 102 may obtain an output from a model that includes one or more vectors (e.g., one or more probabilistic vectors) within a vector space (e.g.. a multidimensional probabilistic vector space). All or a portion of the one or more vectors may represent a likelihood of occurrence of a respective anomalous water quality' event at a respective location with a respective timing, a respective severity' of the event, and / or a respective effect of the event. To generate the value to assign to a particular anomalous water quality event at a particular location with a particular timing, a particular severity, and / or a particular effect, the water quality anomaly detection system 102 may process (e.g., adjust aggregate) all or a portion of the one or more vectors. For example, the water quality anomaly detection system 102 may adjust all or a portion of the one or more vectors such that the cumulative likelihood of the all or a portion of the one or more vectors is adjusted to a scale of 0 to 1 or any other alphanumeric and / or graphical scale (e.g., 0 to 10. 0 to 100, A to Z, A to F, Green to Red or other color scale / shading scale, etc.) and may aggregate the all or a portion of the one or more vectors to obtain the value.

[0087] The value may be specific to the input (obtained from the computing device 106). For example, a first value may be assigned for a first input indicating a first timing (e.g., the next 24 hours) and a second value may be assigned for a second input indicating a second timing (e.g., the next 48 hours).

[0088] In some embodiments, the water quality' anomaly detection system (e.g., the water quality risk detection system) 102 may and / or is configured to normalize the value. For example, the water quality anomaly detection system 102 may normalize the value to obtain a human readable score (e.g., a score on a scale of 0 to 10), a human readable rating, etc.

[0089] In some embodiments, the water quality' anomaly detection system (e.g., the water quality risk detection system) 102 may generate one or more trends (e.g., water quality trends) based on the output of the model. For example, the water quality anomaly detection system 102 may generate one or more temporal trends and / or spatial trends indicating how the output of the model (e.g., predicting a likelihood of occurrence of an anomalous water quality event) tracks or deviates from historical anomalous water quality events associated with the same location and / or anomalous water quality events predicted to have the same or a similar timing.

[0090] In some cases, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may cause display of an interface (e.g.. via the computing device 106, another display or output, etc.) based on the output of the model. For example, the water quality anomaly detection system 102 may dynamically generate a graphical user interface that indicates the assigned value, the anomalous water quality event, the one or more trends, etc. and may cause display of the dynamically generated graphical user interface. As discussed herein, the interface may be interactive such that a user can enter and / or receive designed information (e.g., obtain and / or access the assigned value, the anomalous water quality7event, the one or more trends, etc., make one or more selections or requests, etc.).

[0091] In some embodiments, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may cause display of an interface that indicates the water data. For example, the water quality anomaly detection system 102 may provide the water data via a data as a service model.

[0092] According to several embodiments, the water quality7anomaly detection system 102 may provide the water data, the assigned value, the anomalous water quality7event, the one or more trends, etc. via a web application, a website, a representational state transfer application programming interface, a message (e.g., an email message, another electronic or virtual message, etc.).

[0093] According to some embodiments, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may and / or is configured to identify (e.g., generate) one or more actions (e g., risk mitigation actions, functions, operations, etc.) based on the output of the model. For example, the one or more actions may include closing a valve, adding chemicals to a water supply, performing testing, conducting treatment of a desired fluid source, requesting the assistance of an agency and / or other company, entity7or individual, turning on a back-up system (e.g., for monitoring, diverting, treating and / or otherw ise taking action in connection with a fluid source), etc.

[0094] Based on identifying the one or more actions, the water quality anomaly detection system (e.g., the water quality risk detection system) 102 may instruct performance (and may cause performance) of the one or more actions. To instruct performance of the one or more actions, the water quality anomaly detection system 102 may route instructions (e.g., computer-executable instructions) directly or indirectly to the control system 110. The instructions may identify an action and a component for performance of the action. In the example of FIG. 1, the instructions may identify an action for performance using the component 112 and, in response to receiving the instructions, the control system 110 maycontrol (e.g., actuate) the component 112. In some cases, the water quality anomaly detection system 102 may instruct automatic control (e.g., automatic actuation, treatment, flow control, etc.) of the component 112 by the control system 110 such that the control system automatically controls the component 112.

[0095] In some embodiments, the water quality' anomaly detection system (e.g., the water quality risk detection system) 102 may generate an alert based on the output of the model. For example, the alert may be an audio alert (e.g., audio content that is prerecorded or dynamically generated using text-to-speech or other techniques), a visual alert, a haptic alert, an actual alert, a virtual alert, etc. The alert may indicate detection of an anomalous water quality event, a likelihood of occurrence of the anomalous water quality' event, a timing of the anomalous water quality event, a severity of the anomalous water quality event, an effect of the anomalous water quality event, an adverse water quality event t pe of the anomalous water quality event, etc. In some embodiments, the alert may indicate performance (e.g., scheduled performance, pending performance, planned performance, etc.) of one or more actions (e.g., closing a valve, performing a treatment protocol, etc.).

[0096] The water quality anomaly detection system (e.g.. the water quality risk detection system) 102 may cause output of the alert via an audio output device, a display, etc. For example, the water quality anomaly detection system 102 may cause periodic or aperiodic output of the alert via a microphone or other audible system. In some embodiments, alerts are virtual (e.g., delivered electronically to a compute, smartphone, other computing devices).

[0097] FIG. 2A depicts a schematic view 200A of an example interface 202 (e.g., a graphical user interface) provided by a yvater quality anomaly detection system (e.g., a yvater quality risk detection system) as discussed herein. For example, the water quality anomaly detection system discussed herein may cause display of the interface 202 via a computing device as discussed herein for receiving an input indicating a particular location for performing detection of the anomalous yvater quality' events. In some embodiments, the interface 202 may reflect a map (e.g., a physical map, a site map, an environment map, a topographic map, a nautical map, an aquatic map, a yvatershed map, a hydrogeological map, a water resource map, a geo-spatial rending, a satellite view of an environment, a plot map. a road map, etc.). In some embodiments, the interface 202 is configured such that the map is interactive (e.g., via user manipulation and other input).

[0098] The water quality anomaly detection system (e.g., the water quality risk detection system) may obtain map data. For example, the water quality anomaly detection system may obtain the map data from a map data store and / or a mapping computing system.In some cases, the water quality anomaly detection system may determine a location of the computing device (e.g.. for display of the interface 202) and the water quality anomaly detection system may identify a portion of the map data associated with the location. In some cases, the water quality anomaly detection system may obtain an input from the computing device (e.g., indicating a location, a geographical area, etc.) and the water quality anomaly detection system may identify a portion of the map data associated with the location.

[0099] The water quality anomaly detection system (e.g.. the water quality risk detection system) may generate the interface 202 using the map data. For example, the water quality7anomaly detection system may generate the interface 202 using the map data such that the interface indicates a portion of the map associated with a particular location (e.g., associated with the computing device).

[0100] The interface 202 may be interactive such that a user can interact with the interface 202 (e.g., via the computing device, a touchscreen, any other input or output device, etc.) to dynamically define and dynamically select a portion of the map. For example, the interface 202 may include one or more interactive components (e.g., an interactive search bar, an interactive selector, etc.) to enable a user to interact with the interface 202 (e.g.. to define a portion of a map for detection of anomalous water quality' events).

[0101] In some embodiments, the interface 202 includes an interactive search bar 204. A user may interact with the interactive search bar 204 (e.g., using the computing device) by providing a search term (e.g.. a location identifier). For example, the user may provide the search term via a numerical and / or textual input, a selection from a menu, any other input method, etc. Based on input received via the interactive search bar, the water quality' anomaly detection system may update the interface 202 (e.g., to reflect the location associated with the search term).

[0102] According to some embodiments, the interface 202 includes an interactive selector 206 to enable a user to interact with the interface 202 to define a portion of a map represented by the interface 202 (e.g., define and save a boundary around the portion of the map). The interactive selector 206 may enable a user to select a portion of a map via a free hand selection, via a placement of a shape, inclusion of a zip code, a set of geospatial coordinates, any other geo-positioning data, etc. For example, the interactive selector 206 may enable a user to define a polygon or any other shape (e.g., oval, circle, irregular shape, etc.) relative to the map. In the example of FIG. 2A, the interface 202 indicates a selection of a portion of a map including Springfield but excluding Carthage.

[0103] The water quality anomaly detection system (e.g., the water quality risk detection system) may perform the detection of anomalous water quality events based on inputs received via the one or more interactive components. In some cases, based on inputs received via the one or more interface components, the water quality anomaly detection system may continuously or non-continuously (e.g., periodically, according to a predetermined frequency, etc.) monitor water data associated with the location and detect anomalous water quality events.

[0104] FIG. 2B depicts a schematic view 200B of an example interface 212 (e.g.. a graphical user interface) provided by a water quality anomaly detection system (e.g., a water quality risk detection system) as discussed herein. The water quality anomaly detection system may cause display of the interface 212 in response to an input received indicating a selection of a portion of a map (e.g., via the interface 202 as discussed herein). For example, the input may include selection of a portion of a map that includes Springfield and excludes Carthage.

[0105] In some embodiments, the interface 212 reflects an output of the water quality anomaly detection system (e.g., that is based on the output of the model as discussed herein). The interface 212 may reflect the output of the water quality anomaly detection system as divided into different subsets of the output (e.g.. a snapshot subset, a value profile subset, and a deep dive subset). The interface 212 includes a first component 211 to select a particular subset of the output for display (e.g., the snapshot subset, the value profile subset, or the deep dive subset). The interface 212 further includes a second component 214 and a third component 216 for display of the subset of the output of the water quality anomaly detection system and / or data associated with the selection of the particular location (e.g., as received via the interface 202).

[0106] In the example embodiment illustrated in FIG. 2B. the first component 211 indicates a selection of the snapshot subset for display. The second component 214 indicates that the snapshot includes the area of selection is “5706.87 square miles,” the location centroid is “[“-93.2002”, “37.1589”],” the county is “Greene County,” the state is “Missouri,” and the zip code is “65809.” The third component 216 indicates “3 Total NPDES Violations,” “1 Known PFAS / PFOS Contamination,” “9 Total SDWA Violations.” “1 Toxic Waste Release Site,” “1.3 Miles From Nearest NPDES Violation,” “20 Miles From Nearest PFAS Contamination,” “5 Miles From Nearest SDWA Violation,” and “2.7 Miles From Nearest Toxic Waste Site.” The numeric values shown are for illustrative purposes in order to indicate the types of values that may be displayed, as w ell as a sample user interface configuration, but the particular displayed values are not necessarily the actual values that may be generated in any particular embodiment for the location data shown.

[0107] FIG. 2C depicts a schematic view 200C of an example interface 222 (e.g., a graphical user interface) provided by a water quality anomaly detection system (e.g., a water quality risk detection system) as discussed herein. The water quality anomaly detection system may cause display of the interface 222 in response to an input received indicating a selection of a portion of a map (e.g., via the interface 202 as discussed herein).

[0108] In some embodiments, the interface 222 reflects an output of the water quality anomaly detection system (e.g., that is based on. at least in part, the output of the model as discussed herein). The interface 222 includes a first component 221 to select a particular subset of the output for display (e.g., the snapshot subset, the value profile subset, or the deep dive subset). The interface 222 further includes a second component 224 and a third component 226 for display of the subset of the output of the water quality anomaly detection system and / or data associated with the selection of the particular location (e.g., as received via the interface 202).

[0109] In the example of FIG. 2C, the first component 221 indicates a selection of the value profile subset for display. The second component 224 indicates the value profile includes one or more values (e.g., scores, risks, risk values, etc.) assigned to the particular location by the water quality anomaly detection system. In the example of FIG. 2C, the second component 224 indicates the NPDES Violations Value is “7.4 / 10,” the SDWA Violations Value is “8.0 / 10,” the PFAS / PFOS Value is “6.4 / 10,” the Toxic Chemicals Value is “6.0 / 10,” the Public Waters Value is “6.0 / 10,” and the Cumulative Anomalous Water Quality Event Value is “6.9 / 10.” The third component 226 indicates one or more graphical components (e.g., with respect to time). In the example of FIG. 2C, the third component 226 indicates the NPDES Violations have changed from 3 in 2020 to 4 in 2021 to 1 in 2022 to 5 in 2023, the SDWA Violations have changed from 2 in 2020 to 1 in 2021 to 3 in 2022 to 2 in 2023, and the distance to PFAS / PFOS has changed from 3 miles in 2020 to 4 miles in 2021 to 2 miles in 2022 to 1 mile in 2023.

[0110] FIG. 2D depicts a schematic view 200D of an example interface 232 (e.g., a graphical user interface) provided by a water quality anomaly detection system (e.g., a water quality risk detection system) as discussed herein. The water quality anomaly detection system may cause display of the interface 232 in response to an input received indicating a selection of a portion of a map (e.g., via the interface 202 as discussed herein).[OHl] In some embodiments, the interface 232 reflects an output of the water quality risk detection system or other water quality anomaly detection system (e.g., that is based on the output of the model as discussed herein). The interface 232 includes a first component231 to select a particular subset of the output for display (e.g., the snapshot subset, the value profile subset, or the deep dive subset). The interface 232 further includes a second component 234 and a third component 236.

[0112] In the example of FIG. 2D, the first component 231 indicates a selection of the deep dive subset for display. The second component 234 and the third component 236 includes providing inputs to and receiving output from a computing system (e.g., implementing artificial intelligence, such as using an LLM or other type of generative machine learning model). In the example of FIG. 2D, the second component 234 indicates an initial output of the computing system “Ask Questions About Water Quality in Natural Language. Receive Customized Insights!” and the third component 236 is an input component to provide a prompt (e.g., a textual prompt) to the computing system.

[0113] FIG. 3 is a flowchart of an example arrangement of operations for detecting anomalous water quality events using or by a water quality anomaly detection system (e.g., a water quality risk detection system) as discussed herein, according to some examples of the disclosed technologies.

[0114] In the illustrated embodiment, at block 302. the water quality anomaly detection system (e g., the water quality risk detection system) obtains an input indicating a first location. For example, the first location may be a geographical region of interest. The input may include a text input, a free hand drawing with respect to a map, a modifiable shape with respect to a map, a selection from a list, a selection from a drop down menu, etc. to indicate the first location.

[0115] At block 304, the water quality anomaly detection system (e.g., the water quality risk detection system) identifies water data (e.g., historical water data, raw water data, etc.) associated with one or more second locations. The water quality anomaly detection system may identity the water data in response to obtaining the input. The one or more second locations may be associated with the first location. For example, the one or more second locations may include the first location, may be within a threshold distance of the first location, may have one or more characteristics that are the same or similar to characteristics of the first location, etc. Further, the water quality anomaly detection system may identify the one or more second locations (and may identify the water data) based on, at least in part, a determination that the one or more second locations are wi thin a threshold of the first location.

[0116] The water data may include contaminant data, water quality data, and / or violation data, among others. For example, the water data may include water quality assessment data, algae bloom data, water load data, impaired water data, enforcement data,compliance data, grant data, protected areas data, safe drinking water data, sensor data, watershed data, river data, data obtained from one or more databases (e.g., regulatory or government databases, private databases, etc.) and / or the like.

[0117] The water data may indicate an assessment of one or more anomalous water quality events (e.g., one or more historical water quality events, one or more adverse water quality events, a likelihood of such events, etc ). For example, the water data may indicate one or more historical anomalous water quality events associated with the one or more second locations and / or the first location. The water data may further include timing data indicating a timing of all or a portion of the one or more anomalous water quality events. The water data may further include severity data indicating a severity of all or a portion of the one or more anomalous water quality events. The water data may further include impact data indicating an impact of all or a portion of the one or more anomalous water quality events. The water data may further include type data indicating a type of all or a portion of the one or more anomalous water quality events.

[0118] The water quality anomaly detection system (e.g., the water quality risk detection system) may obtain the water data from a plurality of data sources (e.g.. a plurality of sensors, a plurality of websites, a plurality of public and / or government databases or data sources, etc.). As the water quality anomaly detection system may obtain the water data from a plurality of data sources, the water data may be associated with (e.g., may have) a plurality of data formats (e.g., based on the data formats of data provided by the plurality of data sources). In some embodiments, the water quality anomaly detection system is configured to receive and / or interact with various data formats during use.

[0119] In some embodiments, the water quality anomaly detection system (e.g., the water quality risk detection system) may perform a web crawl and / or a web scrape and may obtain the water data via the w eb craw l and / or the web scrape.

[0120] According to some embodiments, as illustrated in FIG. 3, at block 306, the water quality anomaly detection system (e.g., the water quality risk detection system) normalizes the water data. The water quality’ anomaly detection system may normalize the w ater data by normalizing a data format of the w ater data to produce normalized water data that is associated with (e.g., has) the same data format.

[0121] At block 308, the water quality anomaly detection system (e.g., the water quality risk detection system) provides the normalized water data to a model. The water quality anomaly detection system and / or a separate system may implement the model. The water quality anomaly detection system or a separate system may generate, train, and / or update themodel. For example, in some embodiments, the water quality anomaly detection sy stem trains and / or is configured to train the model to predict a likelihood of occurrence of one or more anomalous water quality events (e.g., one or more future anomalous water quality events). In some cases, the water quality anomaly detection system may parse the water data and / or generate the model based on parsing the water data.

[0122] The model may be a machine learning model, a statistical model (e.g., a Bayesian model), a mathematical model, etc. For example, the model may be a large language model.

[0123] With continued reference to the embodiment illustrated in FIG. 3, at block 310, the water quality anomaly detection system (e.g., the water quality risk detection system) detects or is configured to detect an anomalous water quality event (e.g.. a predicted future anomalous water quality event) associated with water or other monitored fluid located (e.g., physically) at the first location based on an output of the model. For example, the anomalous water quality event may be a discharge event, a chemical release event, a pollution event, a violation event, etc. The water quality anomaly detection system may determine (e.g., continuously, substantially continuously, intermittently, etc.) the likelihood (e.g., a numerical likelihood, a relative or comparative indication, other assessment) of occurrence of the anomalous water quality event based on the output. The water quality anomaly detection system may compare the likelihood to a threshold and determine whether the likelihood is greater than, is less than, or matches a threshold. Based on determining the likelihood is greater than, is within, or matches the threshold, the water quality anomaly detection system may detect the anomalous water quality event. It wi 11 be appreciated that in some embodiments, the model may be fully trained or otherwise configured well in advance of the model being provided with a request to determine a likelihood of an anomalous event for a given location at block 308 (e.g., training may occur days or weeks prior to the model being requested to provide a prediction or score for a particular location based on that location’s current input data). In other embodiments, the model may be finetuned, retrained or updated for the given location(s) (e.g., based on up-to-date training data) just prior to the model being utilized to generate an output for the given location(s).

[0124] In some embodiments, the water quality anomaly detection system (e g., the water quality risk detection system) may determine a respective likelihood (e.g., a respective numerical likelihood) of occurrence of each anomalous water quality event of the one or more anomalous water quality events based on the output. The model may additionally or alternatively generate a cumulative likelihood of any adverse event in a given categoryoccurring (e.g.. determination of a cumulative likelihood of a water quality violation event by combining the likelihood of a water quality violation event belonging to one of multiple relevant categories, such individual likelihoods of an NPDES violation, an SDWA violation, an adverse measurement of a corresponding contaminant, and / or an adverse hazardous waste event). In other embodiments, the model can generate a likelihood or other output based on fewer, more and / or different considerations (e.g., selected area of interest, temporal input(s), future impact of treatment and other mitigation of pollutants, etc.), as desired or required.

[0125] At block 312, the water quality anomaly detection system (e.g., the water quality7risk detection system) provides (e.g., to a computing system) an indication of a likelihood of occurrence of the anomalous water quality7event (e.g., based on the output). In some cases, the water quality anomaly detection system may provide (e.g., to the computing system) an indication of at least a portion of the first location associated with the anomalous water quality7event (e.g., based on the output). In some cases, the water quality7anomaly detection system may provide (e.g., to the computing system) an indication of a timing (e.g., a predicted timing) associated with the anomalous water quality event (e.g., based on the output). In some cases, the water quality anomaly detection system may provide (e.g.. to the computing system) an indication of atype (e.g., apredicted anomalous water quality7event type) associated with the anomalous water quality7event (e.g., based on the output).

[0126] The output may be indicative of one or more anomalous water quality events. All or a portion of the one or more anomalous water quality events may be associated with a respective event type (e g., a contamination event, a violation event, etc.). The water quality anomaly detection system (e.g., the water quality7risk detection system) may generate one or more probabilistic vectors within a multidimensional probabilistic vector space. All or a portion of the one or more probabilistic vectors may correspond to a respective anomalous water quality7event of the one or more anomalous water quality events. The water quality anomaly detection system may provide the one or more probabilistic vectors to a computing system.

[0127] In some embodiments, based on, at least in part, detecting the anomalous water quality event, the water quality anomaly detection system (e.g.. the water quality risk detection system) may instruct performance of one or more actions (e.g., risk mitigation actions, automated actions, etc.) by a computing system. The water quality7anomaly detection system may determine the one or more actions based on the output. The one or more actions may be one or more actions to reduce the likelihood of occurrence of the anomalous water quality event. For example, to instruct performance of the one or more actions, the waterquality anomaly detection system may identify a valve or a set of valves or other flow control devices or features (e.g.. valves, conduits, reservoirs, overflows, drainage basins, etc.) based on the output and instruct actuation of the value based on the output. In another example, to instruct performance of the one or more actions, the water qualify anomaly detection system may instruct a control system (e.g., a SCADA system) to control a component based on the output and may route instructions to the control system.

[0128] In some embodiments, the action the system recommends or otherwise determines should be taken includes a treatment. Such treatment can include, without limitation or restrictions, one or more of the following: addition of chemicals (e.g., for pH adjustment, disinfection, solids control, etc.), diversion for further testing, storage and / or treatment, chemical and / or biological treatment (e.g.. settling or other solids removal, coagulation, disinfection, etc.). In some embodiments, data obtained from testing systems (e.g., one or more autonomous fluid testing devices) can be used to determine whether treatment is necessary7or desired. Further, data obtained from such testing systems can be used to provide additional insight (e.g., determine the efficacy of any treatment that was performed, determine if additional treatment protocols are necessary, etc.).

[0129] In some embodiments, the water qualify anomaly detection system (e.g., the water qualify risk detection system) may generate (e.g., dynamically) image data and / or audio data based on the output. The image data and / or the audio data may indicate the anomalous water quality event. The water quality anomaly detection system may output or instruct output of the image data and / or the audio data using a display and / or an audio or other output device.

[0130] In some embodiments, the water qualify anomaly detection system (e.g., the water qualify risk detection system) may generate (e.g., dynamically) an alert based on the output. The alert may indicate one or more of a predicted timing, a predicted location, a predicted severity, a predicted event ty pe, or a predicted effect of the anomalous water qualify event. The water quality7anomaly detection system may instruct output of the alert (e.g., via a computing device). In some embodiments, alerts generated in connection with various embodiments described herein can be virtual. For example, an alert can be provided to a user via an electronic communication (e.g., electronic email, an electronic message or alert on a smartphone, tablet and / or other computing device, etc.). In some embodiments, alerts are provided via a user portal or other interface and / or using any other communication method or technology, as desired or required.

[0131] The water qualify anomaly detection system (e.g., the water qualify risk detection system) may generate instructions (e.g., computer-executable instructions) based on,at least in part, the output to instruct performance of an action and may provide the instructions to a computing system (e.g., a control system). In some embodiments, the water quality anomaly detection system may provide an identifier of the one or more actions (e.g., cause display of the identifier) by a computing device (e.g., a user computing device).

[0132] According to some embodiments, the water quality anomaly detection system (e.g., the water quality risk detection system) may identity’ one or more trends (e.g., water quality trends) based on the water data and / or the output. The water quality anomaly detection system may provide an indication of the one or more trends to the computing system. Such indication can be provided according to a particular time frequency, based on the occurrence of a particular event or set of events and / or the like.

[0133] In some embodiments, the water quality anomaly detection system (e.g.. the water quality risk detection system) may determine user data associated with a user (e.g., a user associated with the input). For example, the user data may indicate an operation, a location, a status, etc. of the user. The water quality’ anomaly detection system may determine an impact of the anomalous water quality event on the user based on the user data and the output. The water quality anomaly detection system may provide, to the computing system, an indication of the impact of the anomalous water quality event on the user.

[0134] In some embodiments, the water quality’ anomaly detection system (e.g., the water quality risk detection system) may detect occurrence or non-occurrence of the anomalous water quality event. For example, the water quality anomaly detection system may confirm whether a predicted future anomalous water quality event physically occurred. The water quality anomaly detection system may train the model based on the output and detecting the occurrence or the non-occurrence of the anomalous water quality event. In some cases, the water quality anomaly detection system may obtain one or more updates (e.g., model updates) and may update the model using the one or more updates.

[0135] In some embodiments, the water quality anomaly detection system (e.g., the water quality risk detection system) may generate (e.g., dynamically) an interface based on the output and may instruct display of the interface via a computing device. In some cases, the water quality anomaly detection system may generate a water quality score and / or a recommendation (e.g., a human readable water quality score and / or a human readable recommendation). For example, the water quality’ anomaly detection system may normalize the output to obtain the water quality score and / or the recommendation. The water quality anomaly detection system may generate the interface such that the interface includes or indicates the water quality score and / or the recommendation.

[0136] FIG. 4 illustrates an example computing system 400 configured to execute the processes and implement the features described above. In some embodiments, the computing system 400 may include: a processing unit 402 (e.g., computer processor), such as aphysical central processing unit (“CPU”); a network interface 404, such as a network interface card (“NICs”); a computer readable medium drive 406, such as a high density disk (“HDD”), a solid state drive (“SDD”), a flash drive, and / or other persistent non-transitory computer- readable media; an input / output device interface 408. such as an input / output (“IO”) interface in communication with one or more microphones; and a memory 410 (e.g., a computer readable memoty), such as random access memory (“RAM”) and / or other volatile non-transitory computer-readable media.

[0137] The network interface 404 can provide connectivity to one or more networks or computing systems. The processing unit 402 can receive information and instructions from other computing systems or services via the network interface 404. The network interface 404 can also store data directly to the memory 410. The processing unit 402 can communicate to and from the memory 410, execute instructions and process data in the memory 410. etc.

[0138] The memory 410 may include computer program instructions that the processing unit 402 executes in order to implement one or more embodiments. The memory 410 can store an operating system 412 that provides computer program instructions for use by the processing unit 402 in the general administration and operation of the computing system 400. The memory 410 can further include computer program instructions and other information for implementing aspects of the present disclosure. For example, in one embodiment, the memory 410 may include interface rendering instructions 414 (e.g., for rendering the interface 202, the interface 212, the interface 222, and / or the interface 232 as referred to herein). As another example, the memory 410 may include a data store 416. In another example, the memory 410 may include a model 418 (e.g., the model 116 referred to herein).

[0139] Depending on the embodiment, certain acts, events, or functions of any of the processes or algorithms described herein can be performed in a different sequence, can be added, merged, or left out altogether (e g., not all described operations or events are necessary for the practice of the algorithm). Moreover, in certain embodiments, operations or events can be performed concurrently, e.g., through multi-threaded processing, interrupt processing, or multiple processors or processor cores or on other parallel architectures, rather than sequentially.

[0140] The various illustrative logical blocks, modules, routines, and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, or as a combination of electronic hardware and executable software. To clearly illustrate this interchangeability, various illustrative components, blocks, modules, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware, or as software that runs on hardware, depends upon the particular application and design constraints imposed on the overall system. The described functionality can be implemented in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosure.

[0141] Moreover, the various illustrative logical blocks and modules described in connection with the embodiments disclosed herein can be implemented or performed by a machine, such as a processor device, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A processor device can be a microprocessor, but in the alternative, the processor device can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor device can include electrical circuitry configured to process computer-executable instructions. In another embodiment, a processor device includes an FPGA or other programmable device that performs logic operations without processing computer-executable instructions. A processor device can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor device may also include primarily analog components. For example, some or all of the signal processing algorithms described herein may be implemented in analog circuitry or mixed analog and digital circuitry'. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a mainframe computer, a digital signal processor, a portable computing device, a device controller, or a computational engine within an appliance, to name a few.

[0142] The elements of a method (e.g., a computer-implemented method), process, routine, or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor device, or in acombination of the two. A software module can reside in RAM memory, flash memory', readonly memory ("ROM"), erasable programmable read-only memory (“EPROM”), electronically erasable programmable read-only memory (“EEPROM”), registers, hard disk, a removable disk, a compact disc read-only memory (“CD-ROM”), or any other form of a non- transitory computer-readable storage medium. An example storage medium can be coupled to the processor device such that the processor device can read information from, and write information to. the storage medium. In the alternative, the storage medium can be integral to the processor device. The processor device and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor device and the storage medium can reside as discrete components in a user terminal.

[0143] Conditional language used herein, such as, among others, "can," "could," "might," "may," “e.g.,” and the like, unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements and / or steps. Thus, such conditional language is not generally intended to imply that features, elements and / or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without other input or prompting, whether these features, elements and / or steps are included or are to be performed in any particular embodiment. The terms “comprising,” “including,” “having,” and the like are synonymous and are used inclusively, in an open-ended fashion, and do not exclude additional elements, features, acts, operations, and so forth. Also, the term “or” is used in its inclusive sense (and not in its exclusive sense) so that when used, for example, to connect a list of elements, the term “or” means one, some, or all of the elements in the list.

[0144] Disjunctive language such as the phrase “at least one of X, Y. Z,” unless specifically stated otherwise, is otherwise understood with the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (e.g., X, Y, and / or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one of Y, or at least one of Z to each be present.

[0145] Unless otherwise explicitly stated, articles such as “a” or “an” should generally be interpreted to include one or more described items. Accordingly, phrases such as “a device configured to” are intended to include one or more recited devices. Such one or more recited devices can also be collectively configured to carry out the stated recitations. For example, “a processor configured to carry out recitations A, B and C” can include a firstprocessor configured to carry' out recitation A working in conjunction with a second processor configured to carry out recitations B and C.

[0146] While the above detailed description has shown, described, and pointed out novel features as applied to various embodiments, it can be understood that various omissions, substitutions, and changes in the form and details of the devices or algorithms illustrated can be made without departing from the spirit of the disclosure. As can be recognized, certain embodiments described herein can be embodied within a form that does not provide all of the features and benefits set forth herein, as some features can be used or practiced separately from others. The scope of certain embodiments disclosed herein is indicated by the appended claims in addition to or rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.

Claims

WHAT IS CLAIMED IS:

1. A method of assessing environmental risk to a selected location or area, the method comprising: receiving information regarding the selected location or area; receiving a first type of historical water quality data associated with the selected location or area; receiving a second type of historical water quality' data associated with the selected location or area, wherein the second type of historical water quality data is different than the first type of historical water quality' data; and providing an indication of environmental risk for a selected time or time period based on. at least in part, the first type of historical water quality data and the second type of historical water quality data; wherein environmental risk is related to risk created by at least one contaminant; and wherein the first type of historical water quality data comprises data related to a violation of a regulatory threshold.

2. The method of Claim 1, further comprising providing instructions to perform at least one action; wherein the at least one action is configured to environmentally impact the selected location or area; wherein providing an indication of environmental risk comprises providing a score; and wherein the second ty pe of historical water quality' data comprises actual measured data of at least one contaminant.

3. The method of Claim 2, wherein the at least one action comprises manipulating at least one flow control device.

4. The method of Claim 2, wherein the at least one action comprises performing at least partial treatment on a fluid.

5. The method of Claim 2, wherein the second type of historical water quality data comprises actual measured data of at least one contaminant.

6. The method of Claim 1, wherein the first ty pe of historical water quality data comprises at least one of: data related to violations of a law. a rule a permit or another instrument that includes a threshold or limit, concentration data for at least one contaminant ofinterest; and data related to water quality relevant to the selected location or area; data related to water flow and distribution relevant to the selected location or area.

7. The method of Claim 1, wherein each of the first type and the second type of historical water quality data comprises at least one of: data related to violations of a law, a rule a permit or another instrument that includes a threshold or limit, concentration data for at least one contaminant of interest; and data related to water quality relevant to the selected location or area; data related to water flow and distribution relevant to the selected location or area.

8. The method of Claim 1, wherein receiving information regarding the selected location or area comprises receiving geographical data related to an area of a map.

9. The method of Claim 8, wherein the map comprises an interactive map selected by a user.

10. A method according to any one of preceding claims, wherein the first type of historical water quality’ data or the second type of historical water quality’ data comprises data stored on a database.

11. The method of Claim 10, wherein the database comprises a government database.

12. The method of Claim 1, wherein at least one of the first type and the second type of historical water quality’ data comprises data obtained and / or maintained by a governmental agency or body.

13. The method of Claim 1, wherein environmental risk is related to risk created by at least one contaminant.

14. The method of Claim 13, wherein the at least one contaminant comprises a forever chemical.

15. The method of Claim 14. wherein the forever chemical comprises PF AS.

16. The method of Claim 13, wherein the at least one contaminant comprises lead.

17. The method of Claim 1, wherein providing an indication of environmental risk comprises providing a score.

18. The method of Claim 17, wherein the score comprises an alphanumeric score.

19. The method of Claim 18. wherein the alphanumeric score ranges from 0 to 1. 0 to 10 or 0 to 100.

20. The method of Claim 1, wherein providing an indication of environmental risk comprises providing a graphical output.

21. The method of Claim 20. wherein the graphical output comprises a darkness or color scale.

22. A method according to any one of the preceding claims, wherein the selected time or time period is a present time.

23. A method according to any one of Claims 1 to 21, wherein the selected time or time period is a future time or a future time period.

24. The method Claim 1, further comprising providing instructions to perform at least one action.

25. The method of Claim 24, wherein the at least one action comprises manipulating at least one flow control device.

26. The method of Claim 24, wherein the at least one action comprises performing at least partial treatment on a fluid.

27. A method of Claim 24. wherein the at least one action comprises providing an alert.

28. A computer-implemented method comprising: obtaining an input indicating a first location; in response to obtaining the input, identifying historical water data associated with one or more second locations and timing data, wherein the one or more second locations are associated with the first location, wherein the historical water data is associated with a plurality of first data formats; normalizing the historical water data to obtain normalized water data, wherein the normalized water data is associated with a second data format; providing the normalized water data to a model, wherein the model is configured to predict a likelihood of occurrence of one or more anomalous water qualify events; detecting an anomalous water qualify event associated with water located at the first location based at least in part on an output of the model; determining a likelihood of occurrence of the anomalous water qualify event based at least in part on the output; providing, to a computing system, an indication of the likelihood of occurrence of the anomalous water qualify event and an indication of at least a portion of the first location associated with the anomalous water qualify event based at least in part on the output; determining one or more risk mitigation actions based on the likelihood of the occurrence of the anomalous water qualify event; andsending instructions to at least one device for performance of the one or more risk mitigation actions based on the likelihood of the occurrence of the anomalous water quality event.

29. The computer-implemented method of Claim 28, further comprising generating the instructions for execution by the at least one device, wherein execution of the instructions causes the at least one device to take one or more actions.

30. The computer-implemented method of Claim 29, wherein the one or more actions comprises altering one aspect of fluid flow to an area impacted by the historical water data.

31. The computer-implemented method of Claim 29, wherein the one or more actions comprises modifying a valve or other device configured to alter a flow of fluids.

32. The computer-implemented method of Claim 29, wherein the one or more actions comprises activating a water treatment device or system33. The computer-implemented method of Claim 32, wherein the water treatment device or system comprises a chemical addition station.

34. The computer-implemented method of Claim 32. wherein water treatment device or system comprises a filter.

35. The computer-implemented method of Claim 32, wherein water treatment device or system comprises at least one of a chemical treatment system and a biological treatment system.

36. The computer-implemented method of Claim 32, wherein water treatment device or system comprises at least temporarily storing a volume of fluid.

37. A computer-implemented method according to any one of Claims 28 to 36, wherein the historical water data comprises government-obtained or government-maintained data.

38. The computer-implemented method of Claim 37, wherein the government- obtained or government-maintained data comprises data about violations of laws, permits or other thresholds.

39. A computer-implemented method according to any one of Claims 28 to 38. wherein the historical water data does not comprise actual concentration data.

40. A computer-implemented method comprising: obtaining an input indicating a first location; in response to obtaining the input, identifying historical water data associated with one or more second locations and timing data, wherein the one or more secondlocations are associated with the first location, wherein the historical water data is associated with a plurality of first data formats; normalizing the historical water data to obtain normalized water data, wherein the normalized water data is associated with a second data format; providing the normalized water data to a model, wherein the model is configured to predict a likelihood of occurrence of one or more anomalous water quality events; detecting an anomalous water quality event associated with water located at the first location based on an output of the model; and providing, to a computing system, an indication of a likelihood of occurrence of the anomalous water quality event and an indication of at least a portion of the first location associated with the anomalous water quality event based on the output.

41. The computer-implemented method of Claim 40, further comprising: determining the likelihood of the occurrence of the anomalous water quality event based on the output.

42. The computer-implemented method of Claim 40 or 41, further comprising: determining a respective numerical likelihood of occurrence of each anomalous water quality event of the one or more anomalous water quality events based on the output.

43. A computer-implemented method according to any one of Claims 40 to 42, further comprising: determining a respective numerical likelihood of occurrence of each anomalous water quality event of the one or more anomalous water quality events based on the output; and determining that a numerical likelihood of occurrence of the anomalous water quality event exceeds, matches, or is within a threshold, wherein detecting the anomalous water quality event is based on determining that the numerical likelihood of the occurrence of the anomalous water quality event exceeds, matches, or is within the threshold.

44. A computer-implemented method according to any one of Claims 40 to 43, further comprising: determining a respective numerical likelihood of occurrence of each anomalous water quality event of the one or more anomalous water quality events based on the output; redetermining that a numerical likelihood of occurrence of the anomalous water quality event exceeds, matches, or is within a threshold, wherein detecting the anomalous water quality event is based on determining that the numerical likelihood of the occurrence of the anomalous water quality event exceeds, matches, or is within the threshold; and instructing performance of one or more risk mitigation actions.

45. A computer-implemented method according to any one of Claims 40 to 44. wherein the anomalous water quality' event comprises a predicted future anomalous water quality event.

46. A computer-implemented method according to any one of Claims 40 to 45, wherein detecting the anomalous water quality event comprises: determining that the likelihood of occurrence of the anomalous water quality event exceeds, matches, or is within a threshold.

47. A computer-implemented method according to any one of Claims 40 to 46, further comprising: continuously determining a respective likelihood of occurrence of the one or more anomalous water quality events with respect to the water located at the first location.

48. A computer-implemented method according to any one of Claims 40 to 47, wherein the input comprises: a text input; a free hand drawing with respect to a map; a modifiable shape with respect to a map; a selection from a list; or a selection from a drop down menu.

49. A computer-implemented method according to any one of Claims 40 to 48, wherein the first location is a geographical region of interest.

50. A computer-implemented method according to any one of Claims 40 to 49, further comprising: providing, to the computing system, an indication of a timing of the anomalous water quality event.

51. A computer-implemented method according to any one of Claims 40 to 50, further comprising:providing, to the computing system, an indication of an anomalous water quality event type of the anomalous water quality event.

52. A computer-implemented method according to any one of Claims 40 to 51, further comprising: identifying one or more water qualify trends based on the historical water data and the output.

53. A computer-implemented method according to any one of Claims 40 to 52. further comprising: identify ing one or more water qualify trends based on the historical water data and the output; and providing, to the computing system, an indication of the one or more water qualify trends.

54. A computer-implemented method according to any one of Claims 40 to 53, further comprising: determining user data associated with a user; determining an impact of the anomalous water qualify- event on the user based on the user data and the output; and providing, to the computing system, an indication of the impact of the anomalous water quality event on the user.

55. A computer-implemented method according to any one of Claims 40 to 54, wherein the historical water data indicates one or more historical anomalous water qualify events.

56. A computer-implemented method according to any one of Claims 40 to 55, wherein the historical water data indicates one or more historical anomalous water qualify events associated with the first location.

57. A computer-implemented method according to any one of Claims 40 to 56, further comprising: determining an automated action for performance based on the output; and instructing performance of the automated action.

58. A computer-implemented method according to any one of Claims 40 to 57, further comprising: routing instructions to a Supervisory- Control and Data Acquisition system based on the output.

59. A computer-implemented method according to any one of Claims 40 to 58, further comprising: instructing actuation of a valve based on the output.

60. A computer-implemented method according to any one of Claims 40 to 59, further comprising: identifying a valve based on the output; and instructing actuation of the valve.

61. A computer-implemented method according to any one of Claims 40 to 60, further comprising: detecting occurrence or non-occurrence of the anomalous water qualify event; and training the model based on the output and detecting the occurrence or the nonoccurrence of the anomalous water qualify event.

62. A computer-implemented method according to any one of Claims 40 to 61, further comprising: obtaining one or more updates; and updating the model using the one or more updates.

63. A computer-implemented method according to any one of Claims 40 to 62, further comprising: dynamically generating an interface based on the output; and instructing display of the interface via a computing device.

64. A computer-implemented method according to any one of Claims 40 to 63, further comprising: dynamically generating an interface based on the output, wherein the interface indicates one or more of a human readable water qualify score or a human readable recommendation based on the output; and instructing display of the interface via a computing device.

65. A computer-implemented method according to any one of Claims 40 to 64, further comprising: normalizing the output to obtain one or more of a human readable water quality score or a human readable recommendation; dynamically generating an interface based on the output, wherein the interface indicates the one or more of a human readable water quality score or a human readable recommendation based on the output; andinstructing display of the interface via a computing device.

66. A computer-implemented method according to any one of Claims 40 to 65, further comprising: determining a water quality’ score based on the output; dynamically generating an interface based on the output, wherein the interface indicates the water quality score: and instructing display of the interface via a computing device.

67. A computer-implemented method according to any one of Claims 40 to 66, wherein the one or more second locations comprise the first location.

68. A computer-implemented method according to any one of Claims 40 to 67, further comprising: determining that the one or more second locations are within a threshold of the first location, w h erein identifying the historical water data is based on determining that the one or more second locations are w ithin the threshold of the first location.

69. A computer-implemented method according to any one of Claims 40 to 68, wherein the historical water data comprises at least one of: contaminant data, water quality data, or violation data.

70. A computer-implemented method according to any one of Claims 40 to 69, wherein the historical water data comprises at least one of: w ater quality assessment data, algae bloom data. water load data, impaired water data, enforcement data, compliance history data indicating compliance or noncompliance with permitted pollutant levels, not reporting scheduled permit reports, or serious or non- serious offender designation associated with one of more frequencies, wherein the frequencies comprise one or more of quarterly, monthly, or yearly), grant data, protected areas data, safe drinking water data, sensor data,watershed data, facility data comprising one or more reports or datasets regarding one or more facilities and at least a portion of their historical w ater data, or river data.

71. A computer-implemented method according to any one of Claims 40 to 70, further comprising at least one of: performing a web crawl to obtain at least a portion of the historical water data; or obtaining at least a portion of the historical water data from one or more data sources, wherein the method further includes: cleaning the obtained at least a portion of the historical water data to remove duplicates and inconsistencies; formatting the cleaned data into a tabular format that includes two or more tables; joining tables in the formatted and cleaned data; and providing at least one user with network-accessible access to the formatted and cleaned data via a Data as a Service (DaaS) interface.

72. A computer-implemented method according to any one of Claims 40 to 71, wherein the anomalous water quality event comprises at least one of: a discharge event, a chemical release event, a pollution event, or a violation event.

73. A computer-implemented method according to any one of Claims 40 to 72, wherein the model comprises a Bayesian model.

74. A computer-implemented method according to any one of Claims 40 to 73, wherein the model comprises a machine learning model.

75. A computer-implemented method according to any one of Claims 40 to 74, further comprising: implementing the model.

76. A computer-implemented method according to any one of Claims 40 to 75, further comprising: parsing the historical water data; and generating the model based on parsing the historical w ater data.

77. A water quality anomaly detection system comprising or consisting essentially of: data processing hardware: and memory in communication with the data processing hardware, the memory storing instructions, wherein execution of the instructions by the data processing hardware causes the data processing hardware to: obtain an input indicating a first location; in response to obtaining the input, identify historical water data associated with one or more second locations and timing data, wherein the one or more second locations are associated with the first location, wherein the historical water data is associated with a plurality of first data formats; normalize the historical water data to obtain normalized water data, wherein the normalized water data is associated with a second data format; provide the normalized water data to a model, wherein the model is configured to predict a likelihood of occurrence of one or more anomalous water quality events; detect an anomalous water quality event associated with water located at the first location based on an output of the model; and provide, to a computing system, an indication of a likelihood of occurrence of the anomalous water quality event and an indication of at least a portion of the first location associated with the anomalous water quality event based on the output.

78. Non-transitory computer readable media comprising or consisting essentially of computer-executable instructions, wherein execution of the computer-executable instructions by a first computing system causes the first computing system to: obtain an input indicating a first location; in response to obtaining the input, identify historical water data associated with one or more second locations and timing data, wherein the one or more second locations are associated with the first location, wherein the historical water data is associated with a plurality of first data formats; normalize the historical water data to obtain normalized water data, wherein the normalized water data is associated with a second data format; provide the normalized water data to a model, wherein the model is configured to predict a likelihood of occurrence of one or more anomalous water quality events;detect an anomalous water quality event associated with water located at the first location based on an output of the model; and provide, to a second computing system, an indication of a likelihood of occurrence of the anomalous water uality event and an indication of at least a portion of the first location associated with the anomalous water uality event based on the output.

79. A computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data; providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events; receiving an output from the model, wherein the output is indicative of an adverse water quality event; and instructing actuation of a valve based on the output.

80. A computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data; providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events; receiving an output from the model, wherein the output is indicative of an adverse water quality event; dynamically generating a user interface based on the output: and instructing display of the user interface via a user computing device.

81. A computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data; providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events; receiving an output from the model, wherein the output is indicative of an adverse w ater quality event; dynamically generating audio data based on the output, wherein the audio data indicates the adverse water quality event; and instructing output of the audio data via an audio output device.

82. A computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data; providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events;receiving an output from the model, wherein the output is indicative of an adverse water quality event; dynamically generating an alert based on the output, wherein the alert indicates one or more of a predicted timing, a predicted location, a predicted severity, a predicted adverse water quality event type, or a predicted effect of the adverse water qualityevent; and instructing output of the alert.

83. A computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data; providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events; receiving an output from the model, wherein the output is indicative of the one or more adverse water quality events, wherein each of the one or more adverse water quality events is associated with a respective ty pe of adverse water quality- event; generating one or more probabilistic vectors within a multidimensional probabilistic vector space, wherein each of the one or more probabilistic vectors corresponds to a respective adverse water quality event of the one or more adverse water quality- events; and providing the one or more probabilistic vectors to a computing system.

84. A computer-implemented method comprising or consisting essentially of: obtaining, from one or more data sources, raw water data; providing the raw water data as input to a model, wherein the model is trained to predict one or more adverse water quality events; receiving an output from the model, wherein the output is indicative of an adverse water quality7event; generating computer-executable instructions based on the output; and providing the computer-executable instructions to a computing system.

85. A computer-implemented method comprising or consisting essentially of: obtaining, from a user computing device, an input indicating a first location; in response to obtaining the input, identifying historical water data associated with one or more second locations; providing an input based on the historical water data to a model, wherein the model is configured to predict a likelihood of occurrence of one or more future anomalous water quality events;based on an output of the model, determining a likelihood of occurrence of a future anomalous water quality- event associated with water located at the first location exceeds, matches, or is within a threshold; identifying one or more actions to reduce the likelihood of the occurrence of the future anomalous water qualify event; and providing, to the user computing device, one or more identifiers of the one or more actions.

Images