CN106990216A - A kind of shallow lake wawter bloom risk analysis early warning system and its analysis and early warning method - Google Patents

Abstract

The invention discloses a risk analysis and early warning system for algal blooms in shallow lakes and an analysis and early warning method thereof. The system includes a data collection module and a data mining module. The data collection module is composed of a conventional monitoring device, an acoustic telemetry device and an auxiliary device. The data mining module is driven by It is composed of factor identification unit, driving factor suitable interval determination unit, driving factor independent effect quantification unit, environmental driving mode analysis and comparison unit and algal bloom risk assessment unit. The invention realizes the high spatial resolution monitoring of environmental factors while avoiding the large-scale deployment of hardware facilities faced by traditional environmental monitoring methods, and has a low rate of misreporting and false alarms for early warning of algae blooms, and can serve the prevention and control of algal bloom disasters in shallow lakes and water resources. manage.

Description

A shallow lake algae bloom risk analysis and early warning system and its analysis and early warning method

technical field

The invention belongs to the technical field of water environment protection, and in particular relates to a risk analysis and early warning system for algal blooms in shallow water lakes and an analysis and early warning method thereof.

Background technique

Globally, more and more shallow lakes are characterized by varying degrees of eutrophication, and the frequency and scale of algal blooms are increasing. This problem has become a focus of social attention. Eutrophication of water body refers to the state where the concentration of nutrients contained in water body is too high, which is easy to lead to rapid growth of aquatic plants, deterioration of water quality and destruction of water ecosystem balance. Phytoplankton in eutrophic lakes grow rapidly and accumulate under suitable conditions, which is called algal bloom. The negative impacts of algal blooms are well known to academics and the general public, including water supply crisis, discoloration of water bodies, lack of oxygen in water bodies, and fish kills.

The growth of phytoplankton involves the satisfaction of physical and chemical factors to their physiological needs, and is also affected by the capture of zooplankton. Therefore, algae blooms can be regarded as the response of algae to the water environment. However, the continuous change of the lake's environment has an increasingly strong impact on its ecosystem. On the one hand, human activities have significantly increased the flux of limiting nutrients (such as nitrogen, phosphorus, etc.) into lake water, thus paving the way for the elimination of the rapid growth of algae. On the other hand, under high temperature conditions, the growth rate of cyanobacteria is increased, the vertical turbulence of the water body is suppressed, and the viscosity of the water body is weakened, which creates favorable conditions for the growth of cyanobacteria. Therefore, global warming is likely to aggravate the eutrophication of lakes. cyanobacteria bloom hazard.

At present, the basic environmental information required for early warning of algae blooms in shallow lakes is often obtained through two main methods, one is satellite image inversion, and the other is fixed-point monitoring of water bodies. The former is more convenient from the perspective of data acquisition, but the existing inversion methods have great uncertainty, and the obtained environmental factor data is accompanied by large errors; at the same time, satellite image inversion is also affected by bad weather (cloudy, rainy, Great constraints from snow, etc.). Generally speaking, fixed-point monitoring of water bodies has high precision, and data acquisition is relatively more reliable and stable. However, this method will inevitably lead to the problem of intensive infrastructure deployment under the condition of meeting the needs of high spatial resolution monitoring of environmental factors, which will bring high costs. environmental monitoring costs. In the selection of early warning methods, the process-based water quality model has a good theoretical basis, but it faces many problems such as long calculation time and difficult parameter estimation in practical application; data-driven methods can realize efficient early warning of algae bloom risk, but in extreme cases The results obtained below are often not ideal. In view of this, it is particularly urgent to provide a new shallow lake algal bloom risk analysis and early warning system and method for the increasingly prominent problems of eutrophication and algal bloom in shallow lakes.

Contents of the invention

Purpose of the invention: The purpose of the present invention is to solve the deficiencies in the prior art and provide a risk analysis and early warning system for algal blooms in shallow lakes and an analysis and early warning method thereof.

Technical solution: The present invention discloses a risk analysis and early warning system for algal blooms in shallow lakes, including a data collection module (1) and a data mining module (2); the data collection module (1) includes a conventional monitoring device (11), an acoustic Telemetry device (12) and auxiliary device (13); Described conventional monitoring device (11) comprises floating station and pile station, monitors the physical and biochemical information relevant to algae growth and algae bloom outbreak in shallow water lake jointly in real time, and this floating The floating station is a buoy arranged at the vertices of a square array at equal intervals in a shallow lake, and the bottom of the buoy is equipped with a multi-parameter water quality sensor. The pile body, the underwater part of the pile body is equipped with a multi-parameter water quality sensor and a chlorophyll a sensor, and the above-water part of the pile body is equipped with a multi-parameter meteorological sensor and a photosynthetically active radiation sensor; the acoustic telemetry device (12) includes an acoustic label carrier, Acoustic tags and hydrophones, real-time monitoring of multiple water body physical parameters related to algae growth and algae blooms at the location of the acoustic tag, the carrier of the acoustic tag is fish in shallow water lakes, the acoustic tag is equipped with a water quality sensor and fixed to the acoustic tag with a nylon cable tie On the dorsal fin of the tag carrier, the acoustic tag uses ultrasonic waves to periodically send its identification information and physical parameter monitoring results to the surrounding water body. The hydrophones are arranged at each floating station and pile station. When submerged in water, the information sent by the surrounding acoustic tags is received in real time; the auxiliary device (13) includes storage equipment and communication equipment, which are all arranged at each floating site and pile site, and the storage equipment preserves the conventional monitoring device (11) and The monitoring data of the acoustic telemetry device (12), the communication equipment realizes the data transmission between the storage device and the data mining module (2); the data mining module (2) is determined by the driving factor identification unit (21) and the suitable interval of the driving factor unit (22), driving factor independent effect quantification unit (23), environmental driving model analysis and comparison unit (24) and algal bloom risk assessment unit (25); Monitor the data, screen out some environmental factors that are significantly associated with the concentration of chlorophyll a and drive the occurrence of water blooms as the driving factor; the suitable range determination unit for the driving factor (22) determines the range of variation of each driving factor that promotes the concentration of chlorophyll a to be at a high level as respective suitable intervals; the driving factor independent effect quantification unit (23) quantifies the independent response of chlorophyll a concentration to the change of each driving factor when other driving factors are limited to their respective suitable intervals; the environmental driving mode analysis and comparison Unit (24) respectively determines the three types of water bloom risk models and their respective critical risk values considering the cumulative impact, multiplicative impact or comprehensive impact of driving factors on the occurrence of water blooms. After comparison, the optimal water bloom risk model and its Critical risk value; described water bloom risk assessment unit (25) draws water bloom occurrence risk distribution under the current environment of shallow lakes in combination with driving factor real-time monitoring information and optimal water bloom risk model, When the risk of algae bloom is greater than the critical risk value, an early warning of algae bloom will be issued to the lake management department.

Further, the acoustic tag in the acoustic telemetry device (12) adopts a long baseline positioning method for positioning.

The present invention also discloses an analysis and early warning method of a risk analysis and early warning system for algal blooms in shallow lakes, which specifically includes the following steps:

(1) The conventional monitoring device (11) of the data collection module (1) monitors the physical and biochemical information related to algae growth and algae blooms in shallow lakes in real time; the acoustic telemetry device (12) detects the acoustic tags scattered in shallow lakes Perform real-time positioning and receive multiple physical parameters of water bodies monitored by acoustic tags; the auxiliary device (13) saves the monitoring information acquired by the conventional monitoring device (11) and the acoustic telemetry device (12) to a storage device;

(2) Data Mining Module (2) Call the historical monitoring data of environmental factors of shallow lakes to build a risk model of algal blooms in shallow lakes, and perform the following steps:

(a) The driving factor identification unit (21) retrieves the historical monitoring data of all environmental factors at the pile site from the storage device of the data collection module (1);

(b) The driving factor identification unit (21) uses the partial mutual information method to screen out some environmental factors that are significantly associated with the concentration of chlorophyll a and drive the occurrence of the driving factor based on historical monitoring data;

(c) The determination unit for the appropriate interval of driving factors (22) uses the orthogonal table design to select the driving factor level combination that meets the orthogonality from the historical monitoring data, and uses the range analysis to obtain the change law of the concentration of chlorophyll a with the level of a single driving factor , and then determine the variation range of each driving factor that promotes the chlorophyll a concentration to be at a high level as their respective appropriate intervals;

(d) the driving factor independent effect quantification unit (23) quantifies the independent response of the chlorophyll a concentration to the change of each driving factor when other driving factors are limited to their respective appropriate intervals;

(e) On the basis of the information provided by the driving factor independent effect quantification unit (23), the environmental driving mode analysis and comparison unit (24) adopts an evolutionary algorithm to determine the accumulative, multiplicative, or comprehensive effects of the driving factors on algae blooms. The three affected algae bloom risk models and their respective critical risk values ensure that each model has the highest prediction accuracy for the occurrence/non-occurrence of algae blooms. After comparing the respective accuracy rates, the optimal algal bloom risk model and its corresponding critical risk value;

(3) The algae bloom risk assessment unit (25) of the data mining module (2) invokes the real-time monitoring information of the driving factors of shallow water lakes to carry out risk analysis and early warning of algae blooms in shallow lakes, and executes according to the following steps:

(A) retrieve the driving factor real-time monitoring data of the conventional monitoring device (11) and the acoustic telemetry device (12) from the storage device of the data collection module (1);

(B) Using Kriging interpolation algorithm to spatially interpolate real-time monitoring data to the entire lake range;

(C) Based on the optimal algal bloom risk model given by the environment-driven model analysis and comparison unit (24), the algal bloom risk distribution in the current environment of shallow lakes is obtained, and the areas where the algal algal bloom risk is greater than the critical risk value are identified as predicted algal blooms. Hua area;

(D) If there are areas where water blooms are currently predicted to occur, release the risk distribution of water blooms in shallow lakes and the areas where water blooms are predicted to occur to the lake management department.

Beneficial effects: the present invention realizes the monitoring of environmental factors with high spatial resolution while avoiding the large-scale deployment of hardware facilities faced by traditional environmental monitoring methods, and the rate of misreporting and false alarms of algae bloom early warning is low, and it can serve for the prevention and control of algae bloom disasters in shallow lakes and water resource management.

Description of drawings

Fig. 1 is a schematic diagram of the system structure of the present invention;

Fig. 2 is a schematic flow chart of the method of the present invention;

Fig. 3 is the arrangement mode of conventional monitoring device and acoustic telemetering device in embodiment 1;

Fig. 4 is the risk distribution of water bloom occurrence in embodiment 2 and the area where water bloom is predicted to occur.

detailed description

The technical solution of the present invention will be described in detail below, but the protection scope of the present invention is not limited to the described embodiments.

Example 1:

A risk analysis and early warning system for algal blooms in shallow lakes in this embodiment, as shown in FIG. 1 , consists of a data collection module 1 and a data mining module 2 .

Part 1: Data Collection Module 1

Data collection module 1 is made up of conventional monitoring device 11, acoustic telemetry device 12 and auxiliary device 13;

(1) As shown in Figure 3, the conventional monitoring device 11 includes a floating station and a pile station to monitor in real time the physical and biochemical information related to the growth of algae and the outbreak of algal blooms in shallow water lakes; wherein, the floating station is arranged in a shallow lake The buoys at the vertices of the square array at equal intervals are equipped with GDYS-201M multi-parameter water quality analyzers at the bottom of the buoys; the pile stations are arranged at the vertices of the square array at twice the spacing of the floating stations in shallow lakes and fixed on the lake bed. The underwater part of the pile body is equipped with GDYS-201M multi-parameter water quality analyzer and AP-700-SDI chlorophyll a sensor, and the above-water part is equipped with IIES-1128 multi-parameter meteorological sensor and HAD-WHY photosynthetically active radiation sensor.

(2) The acoustic telemetry device 12 includes an acoustic tag carrier, an acoustic tag and a hydrophone, and monitors in real time multiple water body physical parameters related to the growth of algae and the outbreak of algal blooms at the position of the acoustic tag; wherein the acoustic tag carrier is fish in shallow water lakes ; Acoustic tags (products of Vemco Company) are equipped with water quality sensors and fixed on the dorsal fin of the acoustic tag carrier with nylon cable ties; hydrophones (products of Vemco Company) are installed on each floating site and pile site, facing downward and submerged in the water.

(3) The auxiliary device 13 includes storage equipment and communication equipment, which are arranged at each floating site and pile site. The storage equipment saves the monitoring data of the conventional monitoring device 11 and the acoustic telemetry device 12, and the communication equipment realizes the storage equipment and data mining module 2 data transmission between.

Part II: Data Mining Module 2

The data mining module 2 is composed of a driving factor identification unit 21, a driving factor suitable interval determination unit 22, a driving factor independent effect quantification unit 23, an environmental driving mode analysis and comparison unit 24, and an algal bloom risk assessment unit 25, and is used to construct shallow water lake algal blooms risk model.

(1) The driving factor identification unit 21 screens out some environmental factors that are significantly related to the concentration of chlorophyll a and drive the occurrence of algae blooms as driving factors according to the historical monitoring data of environmental factors in shallow water lakes;

(2) The driving factor suitable interval determination unit 22 determines the variation range of each driving factor that promotes the chlorophyll a concentration to be at a high level as the respective suitable interval

(3) The driving factor independent effect quantification unit 33 quantifies the response of the chlorophyll a concentration to the change of each driving factor when other driving factors are in their respective appropriate intervals;

(4) The analysis and comparison unit 24 of the environment driving mode determines three kinds of algal bloom risk models and their respective critical risk values considering the accumulative impact, multiplicative impact or comprehensive impact of the driving factors on algae blooms, and gives the optimal one after comparison Algal bloom risk model and its critical risk value.

(5) The water bloom risk assessment unit 25 combines the real-time monitoring information of driving factors and the optimal water bloom risk model to obtain the risk distribution of water blooms in the current environment of shallow lakes, and report to the lake management department when the risk of water blooms is greater than the critical risk value Carry out water bloom warning.

Example 2:

The risk analysis and early warning method for algal blooms in shallow lakes in this embodiment includes the following steps as shown in Figure 2:

(1) The conventional monitoring device 11 of the data collection module 1 includes a floating station and a pile station, and real-time monitoring of physical and biochemical information related to algae growth and blooms in shallow lakes; the floating stations are arranged at equal intervals in shallow lakes The buoys at the vertices of the square array, the pile sites are arranged at the vertices of the square array at twice the spacing of the floating sites in the shallow lake and fixed on the piles on the lake bed; the acoustic telemetry device 12 performs real-time monitoring of each acoustic tag scattered in the shallow lake Locate and receive multiple water body physical parameters monitored by the acoustic tag; the auxiliary device 13 saves the monitoring information acquired by the conventional monitoring device 11 and the acoustic telemetry device 12 to a storage device;

(2) The data mining module 2 calls the historical monitoring data of environmental factors of shallow lakes to construct a risk model of algal blooms in shallow lakes, and executes according to the following steps:

From the storage device of the data collection module 1, the historical monitoring data of all environmental factors at the pile site are retrieved;

The driving factor identification unit 21 uses the partial mutual information method to screen out some environmental factors that are significantly correlated with the concentration of chlorophyll a and driven to occur as driving factors based on historical monitoring data;

The determination unit 22 of the suitable interval of the driving factor uses the orthogonal table design to select the combination of driving factor levels satisfying the orthogonality from the historical monitoring data, and uses the range analysis to obtain the change rule of the concentration of chlorophyll a with the level of a single driving factor, and then determines the driving factor level that promotes chlorophyll a. a The variation range of each driving factor whose concentration is at a high level is taken as the appropriate interval;

The driving factor independent effect quantification unit 23 quantifies the independent response of the chlorophyll a concentration to the change of each driving factor when other driving factors are limited to their respective appropriate intervals;

The environmental driving mode analysis and comparison unit 24, on the basis of the information provided by the driving factor independent effect quantification unit 23, adopts an evolutionary algorithm to determine the three types of water bloom risks considering the cumulative impact, multiplicative impact or comprehensive impact of the driving factors on the water bloom. Models and their respective critical risk values to ensure that each model has the highest prediction accuracy for the occurrence/non-occurrence of water blooms. After comparing the respective accuracy rates, the optimal water bloom risk model and its corresponding critical risk values are finally given;

(3) The algae bloom risk assessment unit 25 of the data mining module 2 calls the real-time monitoring information of the driving factors of shallow lakes to carry out risk analysis and early warning of algae blooms in shallow lakes, and executes according to the following steps:

The real-time monitoring data of the driving factor of conventional monitoring device 11 and acoustic telemetry device 12 are transferred from the storage device of data collection module 1;

Use kriging interpolation algorithm to spatially interpolate real-time monitoring data to the entire lake range;

Based on the optimal algal bloom risk model given by the environment-driven model analysis and comparison unit 24, the algal bloom risk distribution under the current environment of shallow lakes is obtained, and the areas where the algal bloom occurrence risk is greater than the critical risk value are identified as areas where algal blooms are predicted to occur (as shown in Fig. 4);

If there are areas where water blooms are currently predicted to occur, the risk distribution of water blooms in shallow lakes and the areas where water blooms are predicted to occur will be released to the lake management department.

Claims (3)

1. a kind of shallow lake wawter bloom risk analysis early warning system, it is characterised in that dug including data collection module (1) and data Dig module (2)；

The data collection module (1) includes routine monitoring device (11), acoustic telemetry device (12) and servicing unit (13)；Institute Stating routine monitoring device (11) includes floating website and stake formula website, common monitoring shallow lake in real time and algal grown and wawter bloom Break out relevant physics and Biochemical Information, the floating website is be arranged in equidistant square array summit in shallow lake floating Mark, buoy bottom is equipped with multi-parameter water quality sensor, and stake formula website refers to be arranged in 2 times of floating website spacing in shallow lake Square array summit and the pile body that is fixed on lakebed, pile body underwater portion is equipped with multi-parameter water quality sensor and chlorophyll A sensors, and pile body above water is equipped with multi-parameter meteorological sensor, light together valid radiation sensor；The acoustic telemetry Device (12) includes acoustics label carrier, acoustics label and hydrophone, in real time monitoring acoustics label present position and algal grown The multiple water body physical parameter relevant with breakout of water bloom, acoustics label carrier is shallow lake fish, acoustics label allocated water quality Sensor is simultaneously fixed on the dorsal fin of acoustics label carrier using nylon cable tie, and acoustics label is using ultrasonic wave periodically by it Identity identification information and physical parameter monitoring result are sent to surrounding water, and hydrophone is arranged in each floating website and stake formula station At point, hydrophone is directed downward and submerged in water, real-time reception ambient acoustic label transmitted information；The servicing unit (13) include storage device and communication apparatus, be arranged at each floating website and stake formula website, storage device preserves conventional The Monitoring Data of monitoring device (11) and acoustics telemetering equipment (12), communication apparatus realizes storage device and data-mining module (2) data transfer between；

The data-mining module (2) is by driven factor recognition unit (21), the suitable interval determination unit of driven factor (22), drive The sub- independence effect quantifying unit (23) of reason, environment drive pattern com-parison and analysis unit (24) and wawter bloom risk assessment unit (25) Composition；The driven factor recognition unit (21) filters out and chlorophyll a according to shallow lake envirment factor Historical Monitoring data The component environment factor that concentration significantly association and driving wawter bloom occur is used as driven factor；The suitable interval determination of the driven factor Unit (22) determines the excursion for promoting chlorophyll-a concentration to be in each high-order driven factor as respective Suitable Area Between；The driven factor independence effect quantifying unit (23) other driven factors be defined in it is each suitable interval in the case of, Quantify the separate responses that chlorophyll-a concentration changes to each driven factor；The environment drive pattern com-parison and analysis unit (24) Respectively calibration consider driven factor wawter bloom occurs cumulative influence, tired three kinds of wawter bloom risk models for multiplying influence or combined influence and Respective critical risk value, provides optimal wawter bloom risk model and its critical risk value more afterwards；The wawter bloom risk assessment Unit (25) combines the real-time monitoring information of driven factor and optimal wawter bloom risk model draws wawter bloom under shallow lake current environment Occurrence risk is distributed, and bloom prealarming is carried out to lake management department in the case where wawter bloom occurrence risk is more than critical risk value.

2. shallow lake wawter bloom risk analysis early warning system according to claim 1, it is characterised in that：The acoustic telemetry Acoustics label in device (12) is positioned using Long baselines location method.

3. a kind of analysis of the shallow lake wawter bloom risk analysis early warning system based on described in claim 1 to 2 any one is pre- Alarm method, it is characterised in that：Specifically include following steps：

(1) the routine monitoring device (11) of data collection module (1) monitors shallow lake and algal grown and breakout of water bloom in real time Relevant physics and Biochemical Information；Acoustic telemetry device (12) carries out real-time to each acoustics label being dispersed in shallow lake Position and receive the multiple water body physical parameter of acoustics label monitoring；Servicing unit (13) is by routine monitoring device (11) and acoustics Monitoring information acquired in telemetering equipment (12) is preserved to storage device；

(2) data-mining module (2) calls shallow lake envirment factor Historical Monitoring data, builds shallow lake wawter bloom risk Model, is performed as follows：

(a) driven factor recognition unit (21) is transferred at a formula website from the storage device of data collection module (1) environment Factor Historical Monitoring data；

(b) driven factor recognition unit (21) is based on Historical Monitoring data, is filtered out using inclined mutual information method dense with chlorophyll a The component environment factor that degree significantly association and driving occur is used as driven factor；

(c) the suitable interval determination unit of driven factor (22) is picking out satisfaction just using orthogonal trial from Historical Monitoring data The driven factor horizontal combination for the property handed over, chlorophyll-a concentration is drawn with the horizontal changing rule of single driven factor using range analysis, And then determine the excursion for promoting chlorophyll-a concentration to be in each high-order driven factor as respective suitable interval；

(d) driven factor independence effect quantifying unit (23) other driven factors be defined in it is each suitable interval in the case of, Quantify the separate responses that chlorophyll-a concentration changes to each driven factor；

(e) environment drive pattern com-parison and analysis unit (24) provides information in driven factor independence effect quantifying unit (23) On the basis of, using evolution algorithm, calibration considers that driven factor cumulative influence occurs on wawter bloom, tired multiplies influence or combined influence respectively Three kinds of wawter bloom risk models and respective critical risk value to ensure that the prediction that wawter bloom occur/does not occur each model is accurate Optimal wawter bloom risk model and its corresponding critical risk value are finally provided after rate highest, relatively more respective accuracy rate；

(3) the wawter bloom risk assessment unit (25) of data-mining module (2) calls shallow lake driven factor to monitor letter in real time Breath, carries out the risk analysis of shallow lake wawter bloom and early warning, performs as follows：

(A) drive of routine monitoring device (11) and acoustics telemetering equipment (12) is transferred from the storage device of data collection module (1) The sub- Real-time Monitoring Data of reason；

(B) Kriging regression algorithm is used by Real-time Monitoring Data space interpolation to whole lake scope；

(C) calculated based on the optimal wawter bloom risk model that environment drive pattern com-parison and analysis unit (24) is provided and obtain shallow lake Wawter bloom occurrence risk is distributed under current environment, and it is prediction generation water that wawter bloom occurrence risk is more than into the region recognition of critical risk value Magnificent region；

(D) if current predictive occurs wawter bloom region and existed, shallow lake wawter bloom occurrence risk is distributed and predicted generation wawter bloom Issued to lake management department in region.