《2008年上半年工作总结和下半年工作计精选》由会员分享，可在线阅读，更多相关《2008年上半年工作总结和下半年工作计精选（79页珍藏版）》请在金锄头文库上搜索。
1、Section 12 Air Quality Forecasting Tools,Section 12 Air Quality Forecasting Tools,2,Background,Forecasting tools provide information to help guide the forecasting process. Forecasters use a variety of data products, information, tools, and experience to predict air quality. Forecasting tools are bui
2、lt upon an understanding of the processes that control air quality. Forecasting tools: Subjective Objective More forecasting tools = better results.,www.epa.vic.gov.au/air/AAQFS,Section 12 Air Quality Forecasting Tools,3,Background,Persistence Climatology Criteria Statistical Classification and Regr
3、ession Tree (CART) Regression Neural networks Numerical modeling Phenomenological and experience Predictor variables,Section 12 Air Quality Forecasting Tools,4,Selecting Predictor Variables (1 of 3),Many methods require predictor variables. Meteorological Air Quality Before selecting particular vari
4、ables it is important to understand the phenomena that affect pollutant concentrations in your region. The variables selected should capture the important phenomena that affect pollutant concentrations in the region.,Section 12 Air Quality Forecasting Tools,5,Selecting Predictor Variables (2 of 3),S
5、elect observed and forecasted variables. Predictor variables can consist of observed variables (e.g., yesterdays ozone or PM2.5 concentration) and forecasted variables (e.g., tomorrows maximum temperature). Make sure that predictor variables are easily obtainable from reliable source(s) and can be f
6、orecast. Consider uncertainty in measurements, particularly measurements of PM.,Section 12 Air Quality Forecasting Tools,6,Selecting Predictor Variables (3 of 3),Begin with as many as 50 to 100 predictor variables. Use statistical analysis techniques to identify the most important variables. Cluster
7、 analysis is used to partition data into similar and dissimilar subsets. Unique (i.e., dissimilar) variables should be used to avoid redundancy. Correlation analysis is used to evaluate the relationship between the predictand (i.e., pollutant levels) and various predictor variables. Step-wise regres
8、sion is an automatic procedure that allows the statistical software (SAS, Statgraphics, Systat, etc.) to select the most important variables and generate the best regression equation. Human selection is another means of selecting the most important predictor variables.,Section 12 Air Quality Forecas
9、ting Tools,7,Common Ozone Predictor Variables,Section 12 Air Quality Forecasting Tools,8,Common PM2.5 Predictor Variables,Section 12 Air Quality Forecasting Tools,9,Assembling a dataset,Determine what data to use What data types are needed and available What sites are representative What air quality
10、 monitoring network(s) to use (for example, continuous versus passive or filter) What type of meteorological data are available (surface, upper-air, satellite, etc.) How much data is available (years),Section 12 Air Quality Forecasting Tools,10,Assembling a dataset,Acquire historical data including
11、Hourly pollutant data Daily maximum pollutant metrics, such as Peak 1-hr ozone Peak 8-hr average ozone 24-hr average PM2.5 or PM10 Hourly meteorological data Radiosonde data Model data Meteorological outputs MM5/TAPM Other Surface and upper-air weather charts HYSPLIT trajectories,Section 12 Air Qual
12、ity Forecasting Tools,11,Assembling a dataset,Quality control data Check for outliers Look at the minimum and maximum values for each field; are they reasonable? Check rate of change between records at each extreme. Time stamps Has all data been properly matched by time? Time series plots can help i
13、dentify problems shifting from UTC to LST. Missing data Is the same identifier used for each field? I.e., 999. Units Are units consistent among different data sets? I.e., m/s or knots for wind speeds. Validation codes Are validation codes consistent among different data sets? Do the validation codes
14、 match the data values? I.e., are data values of 999 flagged as missing?,Section 12 Air Quality Forecasting Tools,12,Tool development is a function of Amount and quality of data (air quality and meteorological) Resources for development Human Software Computing Resources for operations Human Softwar
15、e Computing,Forecasting Tools and Methods (1 of 2),Section 12 Air Quality Forecasting Tools,13,Forecasting Tools and Methods (2 of 2),For each tool What is it? How does it work? Example How to develop it? Strengths Limitations,Ozone = WS * 10.2 +,Section 12 Air Quality Forecasting Tools,14,Persisten
16、ce (1 of 2),Persistence means to continue steadily in some state. Tomorrows pollutant concentration will be the same as Todays. Best used as a starting point and to help guide other forecasting methods. It should not be used as the only forecasting method. Modifying a persistence forecast with forecasting experience can help improve forecast accuracy.,Persistence forecast,Section 12 Air Quality Forecasting Tools,15,Persistence (2 of 2),Seven high ozone days
