Professional Documents
Culture Documents
2, October-December 2008
ABSTRACT
The prediction of temperature and rainfall on a seasonal time scales has been
attempted by various research groups using different techniques. The prediction of
these parameters on monthly and seasonal time scales is not only scientifically
challenging but is also important for planning and devising agricultural strategies.
This paper describes the Box-Jenkins time series seasonal ARIMA (Auto
Regression Integrated Moving Average) approach for prediction of temperature and
rainfall on monthly scales. The Box Jenkins technique is applied to predict
temperature and rainfall of next five years by analyzing last twelve years data
(1994-2006). Previous years data is used to formulate the seasonal ARIMA model
and in determination of model parameters. The performance evaluations of the
adopted models are carried out on the basis of correlation coefficient (R2) and root
mean square error (RMSE). The study conducted at Mirzapur, Uttar Pradesh (India).
The results indicate that the seasonal ARIMA model provide reliable and
satisfactory predictions for rainfall and temperature parameters on monthly scale.
Key Words : ARIMA, Meteorology, Rain fall, Hydroelectric power
506
Journal of Environmental Research And Development Vol. 3 No. 2, October-December 2008
year, the intra seasonal variability is also The selected study area Mirzapur, Uttar
crucial. Even if the mean annual rainfall is Pradesh (India) is located in the sub-tropical
normal, the delay in the monsoon onset and belt of the river Ganga plain and experiences
the unusual monsoon lulls or breaks in a dry tropical climate with marked seasonal
between may cause severe disruption of variations. Mirzapur is located between the
agricultural activities, hydroelectric power forest area of Barkachha reserve forest on the
generation and even drinking water supply, east and the Dauti Reserved forest on the
as is observed in many years in the past.
west and has a seasonally dry climate
Therefore, farmers need the location and
dominated by dry deciduous and dry
time-specific information about the monsoon
savannah grass lands. Khajuri River flowing
behavior. As a result, forecasting of monsoon
along the western end of sub water shed. The
on time scales of daily, weekly and monthly
is a major scientific issue in the field of study area has a complex alignment in
monsoon meteorology. orientation and distribution, structure, relief,
slope, estimate and vegetation. Here
The forecasting of rainfall and tempe-
rature started more than a century ago17. monsoon is divided into three seasons
After the MONEX program, with the consisting of a cool dry season of northerly
increased understanding of the monsoon winds during the months of Oct-Feb, a hot
phenomenon, more sophisticated empirical dry season from March-June and hot wet
statistical models5,7,14,15 were developed season from July to September. Monsoon
which have been used with varying degrees rain comes in the third week of June with on
of success. However, successful long-range
set of hot wet season with southeast winds
prediction on shorter than seasonal/monthly
time scale and regional space scales still (mid June – mid October). It is observed that
remains elusive. 86 % of the total annual rainfall comes in the
Data Collection
Mirzapur district’s rainfall and temperature data (1994-2006) is used for analysis and model
formulation as shown in Figs. 2-4.
observed min. Temperature
35.00
Min.Temperature (C)
30.00
25.00
20.00
15.00
10.00
5.00
0.00
Aug- Jan- May- Sep- Feb- Jun- Nov- Mar- Aug- Dec- Apr-
93 95 96 97 99 00 01 03 04 05 07
Months
Fig. 2 : Observed minimum temperature in Mirzapur district (Jan. 1994-Dec. 06)
available. These models help in knowing the of models have been developed to generate
probable weekly, monthly or annually rainfall and runoff.
rainfall. Over the past decade or so, a number
Min. temp
MIN.TEMP MIN.TEMP
Min. temp
1.0 1.0
.5 .5
Partial ACF
ACF
0.0 0.0
-.5 -.5
-1.0 -1.0
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69
510
Journal of Environmental Research And Development Vol. 3 No. 2, October-December 2008
Max. temp
MAX.TEMP Max. temp
MAX.TEMP
1.0 1.0
.5 .5
Partial ACF
ACF
0.0 0.0
-.5 -.5
-1.0 -1.0
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69
Rainfall
RAINFALL Rainfall
RAINFALL
1.0 1.0
.5 .5
Partial ACF
ACF
0.0 0.0
-.5 -.5
-1.0 -1.0
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69
Table 1
Seasonal ARIMA model
Parameters R2 RMSE
(pdq, psdsqs)
Minimum Temperature (504,202) 0.94 6.572
Maximum Temperature (303,202) 0.87 5.81
Rainfall (302,201) 0.72 62.40
Observed and predicted values of next fives years are determined and plotted as shown in
Figs. 11-13
511
Journal of Environmental Research And Development Vol. 3 No. 2, October-December 2008
50 Observed predicted
45
Max. Temperature (C)
40
35
30
25
20
15
10
5
0
Mar-94 Feb-96 Jan-98 Dec-99 Nov-01 Oct-03 Sep-05 Aug-07 Jul-09 Jun-11
Months
Fig. 11 : Comparison of observed and predicted values of maximum temperatures
In the analysis of observed data there are good accuracy as statistics of models
many missing values which results into less indicate. The accuracy of predictions made
accuracy of predicted model in case of
rainfall. Higher values of R2 of models for for rainfall by seasonal ARIMA model is less
temperatures and rainfall show better because data is abrupt and contains many
accuracy of prediction and can be used in missing values, which increases white noise
forecasting temperature and rainfall a few in the system; because no value is assigned
months ahead.
to missing values and are assumed zero.
CONCLUSION Accuracy of these predictions can be
Seasonal ARIMA models can predict increased in future by using these predicted
minimum and maximum temperature with values for missing values.
25.00
20.00
15.00
10.00
5.00
0.00
Aug-93 Sep-95 Sep-97 Oct-99 Nov-01 Nov-03 Dec-05 Jan-08 Jan-10 Feb-12
M onths
Fig. 12 : Comparison of observed and predicted values of minimum temperatures
512
Journal of Environmental Research And Development Vol. 3 No. 2, October-December 2008
Rainfall(cm) 500
400
300
200
100
0
Aug-93 Sep-95 Sep-97 Oct-99 Nov-01 Nov-03 Dec-05 Jan-08 Jan-10 Feb-12
Months
Fig. 13 : Comparison of observed and predicted rainfall
Box-Jenkins ARIMA technique can be 4. Basu S. and Andharia H.I. The chaotic
used to predict the seasonal and monthly time-series of Indian monsoon rainfall
mean rainfall and temperature, using rainfall and its prediction. Proc Indian Acad Sci
and temperature time series as inputs.
(Earth Planet Sci), 101, 27-34, (1992)
Various verification statistics have shown
that prediction skill is good. Previous 12 5. Gowariker V., Thapliyal V., Sarker
years mean monthly rainfall and temperature R.P., Mandal G.S. and Sikka D.R.
values are used to predict next year’s values Parametric and power regression
and can be predicted with sufficient lead- models-new approach to long range
time and good accuracy. This indicates that it forecasting. Mausam 40, 115-122,
may be possible to develop a suitably (1989)
configured seasonal ARIMA model for
predicting monthly rainfall and temperature 6. Tang Z., Almeida C. and de Fishwick
of any region. P.A., Time series forecasting using
neural networks versus Box-Jenkins
REFERENCES methodology. Simulations 57, 303-310,
1. Ashan M. O. and Connor K.M., A (1991)
reappraisal of the kalman filtering 7. Shukla J. and Paolino D.A., The
technique, as applied in river flow Southern Oscillation and long-range
forecasting. Journal of Hydrology 161, forecasting of the summer monsoon
197–226, (1994) rainfall over India, Mon Weather Rev
2. Haltiner J. and Salas J.D., Short term 111, 1830-1837, (1983)
forecasting of snowmelt runoff using 8. Sarle W.S., Neural networks and
armax models, Water Resources statistical models. Proc 19th Ann SAS
Bulletin 24, 1083–1089, (1988) Users Group International Conference,
3. Shamseldin A.Y., Application of a 1538-1550, (1994)
neural network technique to rainfall- 9. Satyan V., Is there an attractor for Indian
runoff modelling. Journal of Hydrology summer monsoon? Proc Ind Acad Sci
199, 272–294, (1997) (Earth Planet Sci), 97, 49-52, (1988)
513
Journal of Environmental Research And Development Vol. 3 No. 2, October-December 2008
10. Navone H.D. and Ceccatto H.A., Peninsular India. India Meteorol Dept
Predicting Indian monsoon rainfall: a Monogr Climatology, 12/81, (1981)
neural network approach., Clim Dyn, 15. Thapaliyal V., Stochastic dynamical
10, 305-312, (1994) model for long-range prediction of
11. Normand C., Monsoon seasonal monsoon rainfall in peninsular India.
forecasting QJR Meteorol Soc, 79, 463- Mausam 33, 399-404, (1982)
473, (1953) 16. Vautard R., Yiou P. and Ghill M.,
12. Perrone T.J. and Miller R.G., Singular spectrum analysis: a toolkit for
Generalised experimental markov and short, noisy and chaotic series. Physica D,
model output statistics: a comparative 58, 95-126, (1992)
verification, Mon Weather Rev, 113, 17. Walker G.T., Seasonal weather and its
1524-1541, (1985) prediction. Brit Assoc Adv Sci, 103, 25-
13. Pathasarathy B., Munot A.A. and 44, (1933)
Kothawale D.R. All India monthly and 18. Webster P.J., Magana V.O., Palmer
seasonal rainfall series : 1871±1993. T.N., Shukla J., Thomas R.A., Yanai M.
Theor ApplClimatol, 49, 217±224, and Yasunari T., Monsoons: Processes,
(1994) predict-ability, and prospects for
14. Thapaliyal V., ARIMA model for long- prediction. J Geo Phys Res., 13, 14451-
range prediction of monsoon rainfall in 14510, (1998)
514