CN116665802A - Assessment method and system for regional ozone concentration - Google Patents
Assessment method and system for regional ozone concentration Download PDFInfo
- Publication number
- CN116665802A CN116665802A CN202310589347.3A CN202310589347A CN116665802A CN 116665802 A CN116665802 A CN 116665802A CN 202310589347 A CN202310589347 A CN 202310589347A CN 116665802 A CN116665802 A CN 116665802A
- Authority
- CN
- China
- Prior art keywords
- concentration
- factor
- ozone concentration
- factors
- ozone
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- CBENFWSGALASAD-UHFFFAOYSA-N Ozone Chemical compound [O-][O+]=O CBENFWSGALASAD-UHFFFAOYSA-N 0.000 title claims abstract description 43
- 239000003344 environmental pollutant Substances 0.000 claims abstract description 22
- 231100000719 pollutant Toxicity 0.000 claims abstract description 22
- 238000012216 screening Methods 0.000 claims abstract description 8
- 230000001419 dependent effect Effects 0.000 claims abstract description 7
- 230000002596 correlated effect Effects 0.000 claims abstract description 6
- 230000008569 process Effects 0.000 claims description 17
- 238000009423 ventilation Methods 0.000 claims description 13
- 239000000356 contaminant Substances 0.000 claims description 12
- 230000005855 radiation Effects 0.000 claims description 8
- 230000000875 corresponding effect Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000001556 precipitation Methods 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 4
- 238000004062 sedimentation Methods 0.000 claims description 3
- 238000013179 statistical model Methods 0.000 description 16
- 230000000694 effects Effects 0.000 description 12
- 238000012360 testing method Methods 0.000 description 8
- 238000011109 contamination Methods 0.000 description 7
- 238000009792 diffusion process Methods 0.000 description 4
- 238000006552 photochemical reaction Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000003915 air pollution Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 238000013277 forecasting method Methods 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 239000002243 precursor Substances 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 239000002341 toxic gas Substances 0.000 description 2
- 239000005436 troposphere Substances 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 101100305901 Mus musculus Rnase2b gene Proteins 0.000 description 1
- 239000011324 bead Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000007795 chemical reaction product Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000005180 public health Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/10—Analysis or design of chemical reactions, syntheses or processes
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
- Y02A50/20—Air quality improvement or preservation, e.g. vehicle emission control or emission reduction by using catalytic converters
Landscapes
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Crystallography & Structural Chemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Analytical Chemistry (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Other Investigation Or Analysis Of Materials By Electrical Means (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method and a system for evaluating regional ozone concentration, comprising the following steps: acquiring and characterizing O of ozone concentration in a region based on a pollutant concentration control equation 3 -8h concentration-dependent high altitude factors and ground factors as predictors; by taking O 3 -a first correlation coefficient of 8h concentration with the high altitude factor and a second correlation coefficient with the ground factor, and screening the forecasting factors according to the principle of the pluralism of the forecasting factors; based on AND O 3 -8h concentration of a first factor positively correlated with O 3 -8h of a second factor with negative correlation of concentration, performing MLR modeling, or constructing an ozone concentration prediction model through SVM and/or BPNN, and evaluating the ozone concentration of the target area to be measured by collecting the first factor and/or the second factor of the target area to be measured; the book is provided withThe application realizes the real-time prediction of the ozone concentration of the target area to be detected and provides data support for preventing the ozone pollution of the area.
Description
Technical Field
The application relates to the technical field of ozone concentration prediction, in particular to an assessment method and system for regional ozone concentration.
Background
In recent years, with the development and implementation of atmospheric control work and the continuous advancement of industrial structure adjustment, PM is being treated 2.5 and PM10 The treatment of the primary pollutants has been effectively controlled, but ozone (O 3 ) The treatment of secondary pollutants is not effectively improved. Based on statistical display, with O 3 The number of times of exceeding standard of the primary pollutant is 41.8%, 55.4% and 39.3% of the total number of times of exceeding standard, and O is used in most areas 3 The ratio of the number of exceeding days of the primary pollutant to the total number of exceeding days is only 34.7%, which indicates O 3 The method has become a primary pollutant affecting the atmospheric environment of an important area, the formation mechanism is more complex, and the pollution prevention and control difficulty is higher. O (O) 3 As an extremely unstable toxic gas, the concentration rise of the toxic gas can seriously influence public health and social image, and accurately forecast urban O 3 The pollution condition is helpful to establish an effective air pollution early warning mechanism and to adopt a flexible control policy to reduce air pollution.
Coastal cities in tropical northern edges at low latitude, belonging to tropical monsoon climates, have frequently suffered from O in recent years 3 Invasion of pollution, O 3 Has become a major atmospheric pollutant that restricts the continued improvement of its air quality, due to the tropical zone at the geographic location, the temperature is higher throughout the year, the sunlight time is long, the solar radiation is strong, and the meteorological conditions are more favorable for the occurrence of photochemical reaction; the area is easily affected by winter season wind, so that the area pollution is more complex; at present, the research of concentration forecasting method in the area is mainly focused on a numerical forecasting mode, and the research of concentration forecasting based on a statistical model is less, so that it is urgently required to design an assessment method and system for regional ozone concentration so as to better perform O for the area 3 And the concentration forecasting work is carried out, and the forecasting quality is improved at the same time.
Disclosure of Invention
In order to solve the problems in the prior art, the application aims to provide an evaluation technology for regional ozone concentration, which is based on pollutant concentration control and pre-processes after considering each meteorological element in a vertical boundary layerThe report factors are screened, and O is constructed by using MLR, SVM and BPNN algorithm 3 The concentration forecast model is used for checking forecast results by using the observation data to obtain O 3 The pollution control provides basis and reference.
In order to achieve the technical object, the present application provides an evaluation method for regional ozone concentration, comprising the steps of:
acquiring and characterizing O of ozone concentration in a region based on a pollutant concentration control equation 3 -8h concentration-dependent high altitude factors and ground factors as predictors;
by taking O 3 -a first correlation coefficient of 8h concentration with the high altitude factor and a second correlation coefficient with the ground factor, and screening the forecasting factors according to the principle of the pluralism of the forecasting factors;
based on AND O 3 -8h concentration of a first factor positively correlated with O 3 -8h of second factors inversely related to concentration, performing MLR modeling, or constructing an ozone concentration prediction model through SVM and/or BPNN, and evaluating the ozone concentration of the target area to be measured by collecting the first factors and/or the second factors of the target area to be measured.
Preferably, in the process of obtaining the predictor, the contaminant concentration control equation is expressed as:
wherein ,indicating local changes in contaminant concentration, +.>Advection item representing contaminant, +.>Representing turbulent flow conveying terms, S c Representing the body source item.
Preferably, in the process of acquiring the advection conveying item, the method is based on the sum O 3 -8h concentration-related 1000-850 hPa high altitude factor and ground factor, and generating advection conveying item.
Preferably, in the process of acquiring the turbulent flow conveying item, the turbulent flow conveying item is generated according to the wind speed difference, the position temperature difference and the temperature difference between two adjacent air pressure layers.
Preferably, in the process of obtaining the body source item, the body source item is generated according to the emission source, dry-wet sedimentation and chemical reaction.
Preferably, in the process of obtaining the forecasting factors, temperature, bit temperature, relative humidity, vertical speed, horizontal speed, wind speed and wind direction of all 7 levels of 1000-850 hPa are obtained, and temperature difference, bit temperature difference, relative humidity difference and wind speed difference between adjacent air pressures, and boundary layer height of the ground, ground ventilation coefficient, ground air pressure, total precipitation, total cloud quantity and ground solar radiation are used as the forecasting factors.
Preferably, in the process of acquiring the surface ventilation coefficient as the predictor, the surface ventilation coefficient is expressed as:
SVC＝PBLH×V h
wherein SVC is the surface ventilation coefficient, V h For an average wind speed of 1000hPa, PBLH represents boundary layer height, where PBLH is obtained by ERA 5.
Preferably, in the process of acquiring the bit temperature θ and the wind direction WD as predictors, the bit temperature is expressed as:
wherein i represents the pressure of the gas, P i Atmospheric pressure at a corresponding height;
the wind direction is expressed as:
wherein ,Ui and Vi Respectively is opposite toA horizontal U wind speed and a horizontal V wind speed at the barometric altitude i.
Preferably, in the process of constructing the ozone concentration prediction model, a combined detection model based on TS score, failure report rate, blank report rate and forecast deviation is constructed, and the model output result is checked, wherein the combined detection model is expressed as:
wherein NA is represented as a certain O 3 -8h of conditions of both concentration level forecast and observation; NB is denoted as a certain O 3 -8h of conditions with a forecast of concentration levels and with no observations; NC is expressed as a certain O 3 -8h concentration level forecast results were absent and observations were present.
The application discloses an evaluation system for regional ozone concentration, which comprises:
the data acquisition module is used for acquiring O used for representing the ozone concentration of the region according to a pollutant concentration control equation based on the spatial position condition of the target region to be detected 3 -8h concentration-dependent high altitude factors and ground factors as predictors;
factor screening module, by obtaining O 3 -a first correlation coefficient of 8h concentration with the high altitude factor and a second correlation coefficient with the ground factor, and screening the forecasting factors according to the principle of the pluralism of the forecasting factors;
ozone concentrationA degree evaluation module for based on the sum O 3 -8h concentration of a first factor positively correlated with O 3 -8h of second factors inversely related to concentration, performing MLR modeling, or constructing an ozone concentration prediction model through SVM and/or BPNN, and evaluating the ozone concentration of the target area to be measured by collecting the first factors and/or the second factors of the target area to be measured.
The application discloses the following technical effects:
the method realizes the real-time prediction of the ozone concentration of the target area to be detected, and provides data support for preventing the ozone pollution of the area.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of the method of the present application;
FIG. 2 is a diagram of O according to an embodiment of the present application 3 -a comparison schematic of three statistical model predictors and observations for 8h concentrations;
FIG. 3 is a diagram of O according to an embodiment of the present application 3 -correlation distribution diagram of three statistical model forecast values and observed values of 8h concentration.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present application.
As shown in fig. 1-3, example 1: the application takes the sea city as the research target to carry out O 3 -8h concentration prediction study, the specific procedure is as follows:
o for Haikou City 3 Purpose of contamination study: the Haikou city is located in the north of the south China island, and has warm and moist climate and beautiful environment. With the rapid growth of socioeconomic in recent years, O 3 Contamination events also occur at times, O 3 Has become a main atmospheric pollutant which restricts the continuous improvement of the air quality in the sea market. On the one hand, the tropical zone in the sea city has higher air temperature throughout the year, long sunlight time, strong solar radiation and better meteorological conditions, thereby being beneficial to the occurrence of photochemical reaction; on the other hand, in the area of the sea city adjacent to the triangular region of the beads, northern O is carried by winter monsoon 3 And precursors thereof further increase the Haikou city O 3 Complexity of contamination. Currently for the sea city O 3 Concentration forecasting method research mainly focuses on numerical forecasting mode, but O based on statistical model 3 Less studies are performed for concentration forecasting. To better realize the Meteorological department and the ecological environment department O in the sea city or even the whole province 3 The application uses O with maximum 8h moving average in the Haikou city of 2015-2020 to improve the forecasting quality 3 Day by day data (O) 3 -8 h) and the synchronous EAR5 re-analysis data, screening the forecasting factors after considering each meteorological element in the vertical boundary layer based on a pollutant concentration control equation, and constructing the O of the sea city by using MLR, SVM and BPNN algorithm 3 The concentration forecast model is used for testing the forecast result in 2021 by using the observation data so as to provide basis and reference for effective O3 pollution prevention and control in the sea and the city.
1. Preparation of data materials:
the application mainly adopts the data monitored by the environmental monitoring central station in the city of the sea, 2015-2021, the data resolution is 1h, and the data effective rate is over 99.9 percent.O 3 Concentration exceeding value reference standard environmental Air Quality Index (AQI) technical regulations (HJ 633-2012) and environmental air quality standard (GB 3095-2012), O 3 -8h concentration grade criteria were: 0 to 100 mu g.m -3 Is excellent; 101-160 mu g.m -3 Is good; 161-215 mu g.m -3 Is lightly polluted; 216-265 mu g.m -3 Is a moderate pollution; 266-800 mug.m -3 Heavy pollution. Wherein greater than 160. Mu.g.m -3 When it is O 3 Exceeding the standard day. In order to incorporate the meteorological elements in the vertical boundary layer as predictors into a statistical model, the application uses 5 th generation data (ERA 5) issued by ECMWF, wherein the data is derived from a Gobey climate change service center database, the time resolution is 1h, the spatial resolution is 0.25 degrees multiplied by 0.25 degrees, and the elements comprise air temperature, relative humidity, horizontal wind speed and the like of 1000-850 hPa, and the boundary layer height, ground air pressure, total precipitation, total cloud quantity and surface solar radiation near the ground.
2. Determination of the study method:
the basic idea of the application is to utilize the meteorological factors and O of the sea city selected in 2015-2020 3 MLR modeling is carried out on 8h concentration data, SVM and BPNN method training is carried out simultaneously, and finally 2021O is utilized 3 -8h concentration observations were examined for the forecast results of the three statistical models. The method comprises the following specific steps:
(1) According to the pollutant concentration control equation, firstly, weather factors of different levels of the sea city are selected as preselected factors, including the weather factors and O 3 -8h concentration-dependent 1000-850 hPa high altitude factor and ground factor;
(2) O is calculated day by day in 2015-2020 of Haikou City 3 The correlation coefficients of the concentration and each pre-selected factor are sequenced according to the absolute value of the correlation coefficient, and weather factors with larger values are selected as the forecasting factors based on the principle of the pluralism of the forecasting factors, and meanwhile, the adjacent air pressure layers of the same factors or the weather factors similar to the same height are prevented from being selected as much as possible;
(3) Based on the data selected in 2015-2020, an MLR equation is constructed and the model O of the sea opening city in 2021 3 -8h concentration for forecasting. The method for the SVM and the BPNN is that 70% of data in 2015-2020 are randomly extracted to be used as a training data set, the remaining 30% are used as a verification data set, the weight of each parameter is debugged, and finally, based on a stable SVM and BPNN model, a forecasting factor in 2021 is substituted into the model and forecasting is carried out. In addition, all meteorological factors are normalized to avoid forecast errors caused by order-of-magnitude differences among the factors.
At 2021 sea opening city O for three statistical models 3 When the concentration forecast result is evaluated for 8 hours, the method mainly selects standard error (RootMeanSquareError, RMSE), average deviation (Meanbias, MB), normalized deviation (MeanNormalizedBias, MNB) and correlation coefficient R for carrying out. RMSE can reflect the difference between the predicted value and the observed value, MB mainly represents the value of the overall predicted value of the sample larger or smaller than the observed value, MNB reflects the degree of the predicted value larger or smaller than the observed value, and the correlation coefficient R represents the degree of the correlation between the predicted value and the observed value. In addition to O 3 When the concentration grade is evaluated for 8h, the method mainly adopts TS score, point Over, PO, not Hit, NH and forecast deviation (Bias, B) for checking, and the specific formulas are as follows:
wherein NA, NB and NC have the physical meanings given in Table 1, and wherein NA is expressed as a certain O 3 -8h concentration, etcThe conditions of both the level forecast result and the observation result are present; NB is denoted as a certain O 3 -8h of conditions with a forecast of concentration levels and with no observations; NC is expressed as a certain O 3 -8h concentration level forecast results were absent and observations were present. The larger the TS score value, the more O is represented 3 The higher the accuracy of the forecast of the 8h concentration level, the interval is between 0 and 1. PO and NH represent the ratio of the number of missed reports to the number of empty reports to the number of observations, the interval of which is also distributed between 0 and 1, and the smaller the value, the model is shown to be corresponding to O 3 The higher the accuracy of the forecast of 8h concentration levels. B has a value interval of 0 to + -infinity, and when B is greater than 1, O is represented 3 -8h concentration levels appear more frequently in the forecast than they actually appear; conversely, if B is less than 1, then the model pair O is represented 3 -8h concentration level underscores.
TABLE 1
3. Results and discussion:
3.1、O 3 establishment of 8h concentration and key meteorological factors:
to give different levels of meteorological factors to O 3 -influence of 8h concentration, thereby determining critical meteorological factors, and being used for construction of a forecast equation and forecast effect inspection. The application firstly controls the equation according to the pollutant concentration:
from formula (5), it can be seen that the contaminant concentration varies locallyAdvection with contaminantsTurbulent flow delivery->Source item S c Related, wherein the body source item S c Including emissions sources, wet and dry settling, and chemical reactions. Considering that the pollutant is mainly concentrated in the boundary layer and its concentration is mainly affected by the coaction of meteorological factors in the boundary layer, so that O 3 -8h concentration preselection factor as follows:
(1) Advection terms, mainly related to wind speed magnitude and contaminant concentration gradient, due to difficulty in obtaining transient O 3 The concentration gradient, therefore, only selects each Wind Speed (WS) and Wind Direction (WD) at the height of 1000-850 hPa as one of the preselected factors;
(2) Turbulent flow conveying item, including thermodynamic turbulence and dynamic turbulence, may be used separately in the reverse temperature strengthThermodynamic stability->Wind shear->The method selects the wind speed difference between two adjacent air pressure layers, the bit temperature difference and the temperature difference as one of preselected factors;
(3) Troposphere O 3 Is a photochemical reaction product, belongs to secondary pollutants, and is not considered O in the application 3 An emissions source of (2);
(4)O 3 is a gaseous pollutant and thus does not consider the dry sedimentation pair O 3 The effect of concentration is only considered to be the photochemical decomposition effect under different water vapor conditions, so that the Relative Humidity (RH) and the relative humidity difference are introducedAs a pre-selected factor;
(5) Boundary Layer Height (PBLH) determines the atmospheric ambient volume and effective air volume for contaminant diffusion. Typically, the boundary layer is low, resulting in an increase in contaminant concentration within the boundary layer due to the suppression of vertical diffusion conditions. PBLH data from the sea city are obtained directly from ERA 5;
(6) Surface Ventilation Coefficient (SVC): the ventilation coefficient may represent the ability of the contaminants within the boundary layer to diffuse and transport horizontally, with smaller values generally being less favorable for contaminant diffusion. The magnitude of which can be expressed as the atmospheric boundary layer height multiplied by the average wind speed within the boundary layer height. The calculation formula is shown in formula (6):
SVC＝PBLH×V h (6)
wherein SVC is the surface ventilation coefficient, V h The average wind speed was 1000 hPa. In addition, the ground meteorological element of the present application also considers ground barometric pressure (SP), total Precipitation (TP), total Cloud Cover (TCC) and Surface Solar Radiation (SSRD), and the data is obtained directly from ERA 5.
The potential temperature theta is calculated by a formula (7) from the temperature T
Wherein i represents the height of the gas pressure, P i Is the atmospheric pressure of the corresponding height.
The wind direction WD is derived from U, V via equation (8):
wherein i represents the height of the air pressure, U i and Vi WD represents the wind direction for each of the horizontal U wind speed and the horizontal V wind speed at the corresponding altitude.
Table 2 shows O 3 -all pre-selected factors of 8h concentration predictors. Comprises temperature T, potential temperature theta, relative humidity RH, vertical speed W, horizontal speed U, horizontal speed V, wind speed WS and wind direction WD of 1000-850 hPa of 7 layers, and temperature difference delta T, potential temperature difference delta theta and relative humidity difference between adjacent air pressuresΔRH and wind speed difference ΔWS, and finally the boundary layer height PBLH of the ground, the ground ventilation coefficient SVC, the ground air pressure SP, the total precipitation TP, the total cloud cover TCC and the ground solar radiation SSRD, are 86 preselected factors.
Calculate the day-by-day O in 2015-2020 3 The correlation coefficient of the concentration of 8h and 86 preselected factors is comprehensively considered, the absolute value of the correlation coefficient and the principle of pluralism based on the forecasting factors are comprehensively considered, the selection of adjacent barometric layers of the same factors or the same highly similar meteorological factors is avoided, and finally 15 forecasting factors are selected totally, wherein the 15 forecasting factors are shown in tables 3-4. As apparent from the table, O in Haikou City 3 -8h concentration and T 850 、θ 850 、RH 1000 、W 1000 、U 875 、V 875 、WD 1000 、ΔRH 975-950 Has a negative correlation with TCC and delta T 975-950 、Δθ 975-950 、ΔWS 1000-975 The PBLH, SVC and SP are positively correlated, wherein the relative humidity RH of the lower layer of the boundary layer 1000 And wind direction WD 1000 Warp direction wind V of upper layer 875 With O 3 The absolute value of the correlation coefficient of the concentration of 8h exceeds 0.4, so that the method has a good indication effect. RH (relative humidity) 1000 In relation to photochemical reaction, when RH is higher, the water vapor content in the atmosphere is higher, and on one hand, solar ultraviolet radiation can be weakened; on the other hand will be with O 3 Chemical reaction occurs, thereby reducing troposphere O 3 Concentration. WD 1000 and V875 Factors related to exogenous delivery indicate proper wind direction for northern air flow to carry O 3 And its precursor delivery to the sea city, affecting O 3 Concentration variation. Temperature and head temperature of upper layer of boundary layer (T 850 and θ850 ) Temperature difference of lower layer and head temperature difference (DeltaT 975-950 and Δθ975-950 ) And ground air pressure (SP) and O 3 The absolute value of the correlation coefficient of the concentration of 8h is between 0.3 and 0.4, and the method has a good indication effect. These factors are mainly related to changes in weather conditions, O in Haikou City 3 The concentration exceeding of 8h is mostly related to the lower part of cold air, the cold high pressure control can cause the ground air pressure to rise, the air temperature to fall, and the concentration of the ground pollutants is favorably increased by matching with the appearance of upper and lower temperature reversing layers. In addition, the city of seaO 3 The 8h concentration is also related to the boundary layer lower wind speed (W 1000 ) Weft wind (U) at upper layer 875 ) Difference in relative humidity of lower layer (DeltaRH) 975-950 ) And wind speed difference (DeltaWS) 1000-975 ) The boundary layer height (PBLH), the Surface Ventilation Coefficient (SVC) and the Total Cloud Cover (TCC) are related, and the absolute value of the correlation coefficient is below 0.3, so that a certain indication effect is realized. These factors are involved in affecting O 3 Water vapor with concentration, sky condition, horizontal conveying and vertical diffusion conditions, etc., embody O in sea city 3 Complexity of contamination.
TABLE 2
TABLE 3 Table 3
| T 850 | θ 850 | RH 1000 | W 1000 | U 875 | V 875 | WD 1000 | ΔT 975-950 |
| -0.377 | -0.377 | -0.407 | -0.267 | -0.299 | -0.507 | -0.409 | 0.337 |
TABLE 4 Table 4
| Δ0 975-950 | ΔRH 975-950 | ΔWS 1000-975 | PBLH | SVC | SP | TCC |
| 0.347 | -0.202 | 0.250 | 0.267 | 0.248 | 0.333 | -0.232 |
Note that: all predictor correlation coefficients passed the 99.9% confidence test.
3.2、O 3 -8h concentration forecasting effect test:
the application constructs O in Haikou city by using 15 meteorological factors selected in the prior art 3 -8h concentration MLR forecasting model and forecasting with 2021 weather factor; by 15 meteorological factors and O in 2015-2020 3 70% of the 8h concentration data is used as a training data set, 30% of the data is used as a verification data set, the weight of each parameter of the SVM and the BPNN model is debugged, and the forecasting factors of 2021 are substituted into the two models after the models are stable, and forecasting is carried out. Finally, 2021 years O of MLR, SVM and BPNN models are obtained 3 -8h concentration forecast results and comparing analysis with observations, the results are shown in fig. 2. The figure clearly shows that the three models can basically forecast the O of the sea city 3 -8h concentration trend, i.e. 2021 years O 3 The 8h concentration shows the characteristic of change of being higher in the half winter and lower in the half summer, and is basically consistent with the observed value. In contrast, the MLR model predictive value is generally lower than the observed value, and the variation amplitude is smaller. Especially 3 and 4 months in spring, 10 and 11 months in autumn, 12 and 1 month in winter, O 3 The forecast value is obviously lower in the period when the observed value of the concentration is higher than the observed value in 8h, and is slightly higher than the observed value in 7 months and 8 months in summer. Unlike the MLR model, the SVM model forecast values are significantly higher in part time period than the observations, such as O for 1 month, 2 months, and 11 months in winter 3 -8h concentration peak period; part of the time period is significantly lower than the observed value, especially O in 5 months of spring and 6 months of summer 3 -8h concentration valley period, the SVM model predictive value exhibiting a large amplitude of variation. In contrast, the predicted value of the BPNN model is closest to the actual measured value, and the predicted value of the BPNN model is slightly lower than the predicted value of the BPNN model of 3 months and 4 months in spring and 12 months and 1 month in winter, part O 3 The concentration peak value is not predicted for 8 hours, and the rest period is basically predictedO (O) 3 -trend of 8h concentration observations. Further calculation of O 3 The correlation coefficients of the 8h concentration observation and the forecast values of the three statistical models (Table 5) are 0.591 (MLR), 0.660 (SVM) and 0.750 (BPNN), respectively, which pass the 99.9% confidence test, wherein the correlation coefficient of the BPNN model is the largest, and the forecast result is superior to the other two statistical models.
FIG. 3 further shows O for three models of 2021 3 -a correlation distribution of 8h concentration forecast values and observations. From the graph, the MLR model predictions are substantially lower than the observations, consistent with the previous analysis; discrete points of SVM model are more scattered, especially O 3 -8h concentration greater than 100. Mu.g.m -3 The period of time, the forecast value and the observed value have larger deviation; discrete points of BPNN model at O 3 -8h concentration less than 100. Mu.g.m -3 The time period being substantially distributed around the trend line and at O 3 -8h concentration greater than 100. Mu.g.m -3 The time interval distribution is also more diffuse, indicating that for O 3 The BPNN model also has obvious errors when the concentration is predicted to be larger at 8 h. Table 5 shows the standard error (RMSE), mean deviation (MB) and normalized deviation (MNB) of the three model predictions from the observations. From the RMSE values of the three models, the BPNN model is the smallest, which is 22.07. Mu.g.m -3 MLR times, 24.88. Mu.g.m -3 Maximum SVM of 27.48. Mu.g.m -3 The BPNN model forecast value is closest to the observed value, and the SVM model deviation is maximum. From the MB value, the MB values of the three models are all negative values and distributed between-3.51 and-8.93 mu g.m -3 And the three statistical model forecast values are respectively lower than the observed value. Wherein the BPNN model is most obvious with MB value of-8.93 mug.m -3 This may be related to a lower forecast in spring and winter of 2021. From MNB values, SVM and BPNN models are negative, while MLR models MNB values are positive, unlike MB values. As can be determined by combining fig. 2 and 3, the overall MLR model forecast is smaller than the observed value, so that the MB value is negative, but the MNB value is positive due to the larger individual period forecast.
TABLE 5
* Indicating that 99.9% confidence test passed.
3.3、O 3 -8h concentration level forecasting effect test:
to further examine three statistical models O in 2021 3 The application firstly counts O in Haikou city by the result of the concentration grade forecast value and the observed value of 8h 3 The number of days and percentages for the different classes of-8 h concentrations are shown in Table 6. As can be seen from the above, the air quality of the water-based air conditioner in the sea city of 2021 is better, O 3 The 8h concentration scale was only three, excellent, good and mild, respectively, with no moderate contamination and above. The majority of days are mainly superior and good, 284d and 76d respectively, accounting for 77.81% and 20.82% of the annual days. 5dO is common in the sea of 2021 3 The 8h concentration reached a light contamination level with a percentage of 1.37%. In recent years, with the expansion of population size of the sea and the enhancement of economic construction, O 3 The contamination problem also presents an increasing trend.
TS scoring is widely applied to scoring standards for quantitative rainfall forecast accuracy in meteorological departments, and the application also adopts TS scoring to O of three statistical models 3 -8h concentration grade forecast results are scored, wherein the method comprises TS scoring, report missing rate (PO), report blank rate (NH), forecast deviation (B) and the like. Table 7 shows three models O in 2021 3 -8h concentration level forecast effect. From the above, it is known that O in Haikou City 3 When the concentration grade of 8h is optimal, TS scores of the MLR, the SVM and the BPNN prediction model are all above 80%, wherein the TS score of the BPNN model is the highest and is 84.36%, and the TS scores of the MLR and the SVM are slightly higher than 81%. At this time, the PO of the SVM is significantly higher than the other two models by 10.56%, while the NH is also lower than the other two models by only 9.93%. The B of SVM is less than 100% and the MLR and BPNN are slightly higher than 100%, indicating that for the best class, SVM is slightly less reported and MLR and BPNN are slightly more reported, with the B of MLR being larger. For the good grade, the TS of the BPNN is rated 74.70%, which is obviously higher than the other two models, and the PO is the least, the NH is the most, 13.89% and 15.07% respectively, which shows that the BPNN has the least missing report and blank report at the good grade. B of SVM is less than 100%, while MLR andthe BPNN was higher than 100% indicating that the rank SVM was slightly less reported, the MLR and BPNN were slightly more reported, and the results were consistent with the superior rank. For the mild pollution grade, the TS score of the MLR and the SVM is reduced to below 70%, and the BPNN is also maintained at a high score of 73.81%, which shows that the BPNN has certain forecasting capability at the mild pollution grade, PO and NH are 15.07%, and are lower than the MLR and the SVM, and three forecasting models B are 100%. Overall, three predictive models TS scores all follow O 3 -8h concentration level increase and decrease, while PO and NH follow O 3 8h concentration level, in particular a light pollution level, mainly due to the fact that the sea city belongs to the region with better air quality, O 3 An out-of-8 h concentration belongs to a low probability event, and only 5d reaches a light pollution level in 2021, so that the forecasting difficulty is high, the hit rate is low, the TS score is low, and the PO and NH are high. In contrast, the BPNN model has better effect, and three O 3 The TS score of the 8h concentration grade is higher than that of the other two models, particularly the light pollution grade, the TS score can be maintained at 73.81%, and PO and NH are obviously lower, so that the BPNN has a certain forecasting capability in the grade.
TABLE 6
| You (percent) | Good (percentage) | Light pollution (percentage) |
| 284(77.81％) | 76(20.82％) | 5(1.37％) |
TABLE 7
4. Conclusion:
(1) Based on a pollutant concentration control equation, after comprehensively considering the absolute value of a correlation coefficient and the pluripotency principle of a forecasting factor, the method screens out the daily sea O of the sea mouth of 2015-2020 3 15 predictors of 8h concentration, O 3 -8h concentration and T 850 、θ 850 、RH 1000 、W 1000 、U 875 、V 875 、WD 1000 、ΔRH 975-950 Has a negative correlation with TCC and delta T 975-950 、Δθ 975-950 、ΔWS 1000-975 The PBLH, SVC and SP have positive correlation, and the absolute value of the correlation coefficient is mainly distributed between 0.2 and 0.507, wherein the relative humidity RH of the lower layer of the boundary layer 1000 And wind direction WD 1000 Warp direction wind V of upper layer 875 With O 3 The absolute value of the correlation coefficient of the concentration of 8h exceeds 0.4, so that the method has a good indication effect.
(2) Using normalized 15 predictors to construct MLR, SVM and BPNN statistical model and to O of 2021 Haikou city 3 The forecasting result of the concentration of 8h is checked, and three models are found to basically forecast O in the sea city 3 -8h of a change trend of higher concentration in winter and lower concentration in summer. Wherein the RMSE value of the BPNN model is minimum and is only 22.07 mu g.m -3 . The correlation coefficient between the measured value and the forecast values of the three statistical models is arranged to be 0.750 (BPNN) > 0.591 (MLR) > 0.660 (SVM) from large to small, and the forecast results of the BPNN model are superior to the other two statistical models through 99.9% confidence test.
(3) For three statistical models O in 2021 3 The result test of-8 h concentration grade forecast shows that the TS scores of the three forecast models are all along with O 3 -8h concentration level increase and decrease, while PO and NH follow O 3 -8h rise in concentration level. The BPNN model is formed by three O 3 In the forecast of the concentration grade of 8h, the TS score is higher than that of the other two models, especially the light pollution grade, the TS score can be maintained at 73.81%, and PO and NH are obviously lower, which is reflected byThe BPNN model has good forecasting performance.
While embodiments of the present application have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the application, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the application.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Claims (10)
1. A method for evaluating the concentration of ozone in a region, comprising the steps of:
acquiring and characterizing O of ozone concentration in a region based on a pollutant concentration control equation 3 -8h concentration-dependent high altitude factors and ground factors as predictors;
by taking the O 3 -a first correlation coefficient of the 8h concentration with the high altitude factor and a second correlation coefficient with the ground factor, performing predictor screening according to the multivariate principle of the predictors;
based on and the O 3 -a first factor positively correlated with the 8h concentration, and with the O 3 -8h of a second factor with negative correlation of concentration, performing MLR modeling, or constructing an ozone concentration prediction model through SVM and/or BPNN, and evaluating the ozone concentration of the target area to be tested by collecting the first factor and/or the second factor of the target area to be tested.
2. The method for evaluating regional ozone concentration of claim 1, wherein:
in the process of obtaining the predictor, the pollutant concentration control equation is expressed as:
wherein ,indicating local changes in contaminant concentration, +.>Advection item representing contaminant, +.>Representing turbulent flow conveying terms, S c Representing the body source item.
3. An assessment method for regional ozone concentration according to claim 2, characterized in that:
in the process of acquiring the advection conveying item, according to the method and the O 3 -8h concentration-dependent 1000-850 hPa high altitude factor and ground factor, generating said advection term.
4. A method for evaluating regional ozone concentration as claimed in claim 3, wherein:
in the process of acquiring the turbulent flow conveying item, the turbulent flow conveying item is generated according to the wind speed difference, the position temperature difference and the temperature difference between two adjacent air pressure layers.
5. The method for evaluating regional ozone concentration of claim 4, wherein:
in the process of obtaining the body source item, the body source item is generated according to the emission source, dry-wet sedimentation and chemical reaction.
6. The method for evaluating regional ozone concentration of claim 5, wherein:
in the process of obtaining the forecasting factors, temperature, bit temperature, relative humidity, vertical speed, horizontal speed, wind speed and wind direction of 1000-850 hPa (high-power) layers, temperature difference, bit temperature difference, relative humidity difference and wind speed difference between adjacent air pressures, and boundary layer height, surface ventilation coefficient, surface air pressure, total precipitation, total cloud quantity and surface solar radiation of the ground are obtained as the forecasting factors.
7. The method for evaluating regional ozone concentration of claim 6, wherein:
in the process of acquiring the surface ventilation coefficient as a predictor, the surface ventilation coefficient is expressed as:
SVC＝PBLH×V h wherein SVC is the surface ventilation coefficient, V h For an average wind speed of 1000hPa, PBLH represents boundary layer height, where PBLH is obtained by ERA 5.
8. The method for evaluating regional ozone concentration of claim 7, wherein:
in the process of acquiring the bit temperature theta and the wind direction WD as predictors, the bit temperature is expressed as follows:
wherein i represents the pressure of the gas, P i Atmospheric pressure at a corresponding height;
the wind direction is expressed as:
wherein ,Ui and Vi The horizontal U wind speed and the horizontal V wind speed corresponding to the air pressure height i are respectively.
9. The method for evaluating regional ozone concentration of claim 6, wherein:
in the process of constructing an ozone concentration prediction model, constructing a combined detection model based on TS score, miss report rate, empty report rate and forecast deviation, and checking a model output result, wherein the combined detection model is expressed as:
wherein NA is represented as a certain O 3 -8h of conditions of both concentration level forecast and observation; NB is denoted as a certain O 3 -8h of conditions with a forecast of concentration levels and with no observations; NC is expressed as a certain O 3 -8h concentration level forecast results were absent and observations were present.
10. An evaluation system for regional ozone concentration, comprising:
the data acquisition module is used for acquiring O used for representing the ozone concentration of the region according to a pollutant concentration control equation based on the spatial position condition of the target region to be detected 3 -8h concentration-dependent high altitude factors and ground factors as predictors;
factor screening moduleBy obtaining the O 3 -a first correlation coefficient of the 8h concentration with the high altitude factor and a second correlation coefficient with the ground factor, performing predictor screening according to the multivariate principle of the predictors;
an ozone concentration evaluation module for based on the O 3 -a first factor positively correlated with the 8h concentration, and with the O 3 -8h of a second factor with negative correlation of concentration, performing MLR modeling, or constructing an ozone concentration prediction model through SVM and/or BPNN, and evaluating the ozone concentration of the target area to be tested by collecting the first factor and/or the second factor of the target area to be tested.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310589347.3A CN116665802B (en) | 2023-05-24 | 2023-05-24 | Assessment method and system for regional ozone concentration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310589347.3A CN116665802B (en) | 2023-05-24 | 2023-05-24 | Assessment method and system for regional ozone concentration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116665802A true CN116665802A (en) | 2023-08-29 |
| CN116665802B CN116665802B (en) | 2024-01-23 |
Family
ID=87721732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310589347.3A Active CN116665802B (en) | 2023-05-24 | 2023-05-24 | Assessment method and system for regional ozone concentration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116665802B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117236528A (en) * | 2023-11-15 | 2023-12-15 | 成都信息工程大学 | Ozone concentration forecasting method and system based on combined model and factor screening |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106019409A (en) * | 2016-05-11 | 2016-10-12 | 北京市环境保护监测中心 | Partition prediction method and system for ozone concentration |
| CN110569528A (en) * | 2019-07-15 | 2019-12-13 | 北京工业大学 | Numerical simulation quantification method for PM2.5 transmission flux below atmospheric boundary layer of cross-boundary region |
| US20200278475A1 (en) * | 2017-02-27 | 2020-09-03 | Hefei Institute Of Physical Science, Chinese Academy Of Science | New method of real-time measuring vertical profiles of multiple atmospheric parameters carried by aerostat |
| CN113570163A (en) * | 2021-09-02 | 2021-10-29 | 河北科技大学 | Atmospheric ozone concentration prediction method, system and device based on mathematical model |
-
2023
- 2023-05-24 CN CN202310589347.3A patent/CN116665802B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106019409A (en) * | 2016-05-11 | 2016-10-12 | 北京市环境保护监测中心 | Partition prediction method and system for ozone concentration |
| US20200278475A1 (en) * | 2017-02-27 | 2020-09-03 | Hefei Institute Of Physical Science, Chinese Academy Of Science | New method of real-time measuring vertical profiles of multiple atmospheric parameters carried by aerostat |
| CN110569528A (en) * | 2019-07-15 | 2019-12-13 | 北京工业大学 | Numerical simulation quantification method for PM2.5 transmission flux below atmospheric boundary layer of cross-boundary region |
| CN113570163A (en) * | 2021-09-02 | 2021-10-29 | 河北科技大学 | Atmospheric ozone concentration prediction method, system and device based on mathematical model |
Non-Patent Citations (1)
| Title |
|---|
| 朱苹 等: "合肥市冬季PM_(2.5)统计预报方法初试与比较研究", 环境科学与技术, vol. 42, no. 12, pages 87 - 95 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117236528A (en) * | 2023-11-15 | 2023-12-15 | 成都信息工程大学 | Ozone concentration forecasting method and system based on combined model and factor screening |
| CN117236528B (en) * | 2023-11-15 | 2024-01-23 | 成都信息工程大学 | Ozone concentration forecasting method and system based on combined model and factor screening |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116665802B (en) | 2024-01-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ye et al. | Study on the synoptic flow patterns and boundary layer process of the severe haze events over the North China Plain in January 2013 | |
| CN106019409B (en) | Ozone concentration partition prediction method and system | |
| CN116665802B (en) | Assessment method and system for regional ozone concentration | |
| Kumar et al. | Temporal characteristics of columnar aerosol optical properties and radiative forcing (2011–2015) measured at AERONET’s Pretoria_CSIR_DPSS site in South Africa | |
| CN112180472A (en) | Atmospheric visibility integrated forecasting method based on deep learning | |
| Stojić et al. | Comprehensive analysis of PM 10 in Belgrade urban area on the basis of long-term measurements | |
| Morales et al. | Atmospheric corrosion in subtropical areas: influences of time of wetness and deficiency of the ISO 9223 norm | |
| Hart et al. | A synoptic climatology of tropospheric ozone episodes in Sydney, Australia | |
| Zeng et al. | Prediction of potentially high PM2. 5 concentrations in Chengdu, China | |
| Lewandowska et al. | Mercury bonds with carbon (OC and EC) in small aerosols (PM1) in the urbanized coastal zone of the Gulf of Gdansk (southern Baltic) | |
| Levy et al. | Coastal and synoptic recirculation affecting air pollutants dispersion: A numerical study | |
| Su et al. | Understanding the distribution and drivers of PM2. 5 concentrations in the Yangtze River Delta from 2015 to 2020 using Random Forest Regression | |
| Abhishek et al. | Long-term wind speed variations for three Midwestern US cities | |
| Hassan et al. | Interaction of PM10 concentrations with local and synoptic meteorological conditions at different temporal scales | |
| Kahya et al. | Spatio-temporal analysis of PM2. 5 over Marmara region, Turkey | |
| CN115409374A (en) | Method for early warning influence of urbanization difference on watershed water ecological service function | |
| Jia et al. | Variations of Siberian High Position under climate change: Impacts on winter pollution over north China | |
| Kulkarni et al. | Surface ozone variability and trend over urban and suburban sites in Portugal | |
| Pryor | A case study of emission changes and ozone responses | |
| Wai et al. | Extreme particulate levels at a western pacific coastal city: the influence of meteorological factors and the contribution of long-range transport | |
| Makra et al. | An objective classification system of air mass types for Szeged, Hungary, with special interest in air pollution levels | |
| CN116598016A (en) | COVID-19 analysis prediction model based on log-linear GAM | |
| Tharani et al. | Spatial distribution analysis of air pollutants and the impact of meteorological factor | |
| Napi et al. | Development of models for forecasting of seasonal ground level ozone (O3) | |
| ULAVI et al. | Regression modelling for predicting visibility in a tropical urban environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |