Time Series Modeling - Water Quality Management: Assessment and Interpretation - page 41

Autoregressive and the first difference in moving average models represented the

chloride data have been used to study the monthly water quality data in Chung
Kang river located at the northern part of Miao-Li Country in the middle of Taiwan.
Five years of investigation were conducted and 12 monthly water quality param-
eters were studied. The result was that forecasting with seasonable data seems to
perform well when the Box-Jenkins technique is combined with non-parametric
transformation.
Jayawardena and Lai ( 1989 ) undertook a very large research program. A time
series analysis approach was applied to model the monthly COD values in the
Yuancan and Fangcan and Guangzle reach of the Pearl River in southern mean
China for a 21 year period. The basic properties of the water quality were deter-
mined, time and frequency-domain analyses were carried out, and the various
stochastic models represent the dependent stochastic component. Using the prob-
ability distribution of the independent residuals generated synthetic water quality
data, and the future water quality was forecasted.
For handling missing data and analyzing water quality data, Mahloch (1974)
demonstrated the application of multivariate statistical techniques. The results of the
study indicate that a simultaneous multiple regression technique may be used for
supplying the missing observations and that at any particular level, the entire data
matrix may be considered, thereby reducing the computational effort.
The sequential order of observations is a key concept, which is incorporated in
stochastic modeling, especially in time series models. The 1960

s witnessed a keen
interest in the probabilistic structure of the sequence dependence of observations. It
brought up the application of autocorrelation, an essential tool in the analysis of
Autoregressive Integrated Moving Average (ARIMA) models.
To establish a working language, a concise description of time series analysis
and its application in water quality studies is given in the following section.
Considering the de

'

cit of water in Iran, protection of water resources against
pollution is vital. In this regard, water quality monitoring is a tool which produces
up to date information. Having a great amount of raw data without interpretation is
not suf

cient and it is necessary to analyze data and predict the variation of water
quality in the future for any decision making on water quality management.
Recently, more researchers have become interested in the application of time series
models for the prediction of water quality.
Time series approach for analyzing water resources were first applied by Tho-

mann ( 1967 ) who studied variation of temperature by the time and dissolved
Delaware Estuary. The data was obtained by continuous recording by monitoring
stations, operated jointly by the U.S. Geological Survey Department and the city of
Philadelphia. Carlson et al. ( 1970 ) and McMichael and Hunter ( 1972 ) reported the
successful use of the Box-Jenkins method for time series analysis.
The Box-Jenkins method for the time series analysis was applied to model the
hourly water quality data recorded in the St. Clair River near Corunna, Ontario, for
chloride and dissolved oxygen levels by Huck and Farquhar ( 1974 ), the models
were physically reasonable and successful results were obtained. Autoregressive
(AR) and

first difference moving average (MA) models represented the chloride