300
P. Verma et al.
rainfall is extremely important. Some literatures compare the method of regression,
neural network, and clustering of data to get a better result in the prediction [3] of
seasonal event like rainfall. Some literature [4] compares the approaches based on
auto-regressive integrated moving average (ARIMA), the fuzzy time series (FST)
model, and the non-parametric method (Theil’s regression). In some work [5, 6],
conventional regression model is modified for the prediction of rainfall of different
nature by iterating the past values and thereby adding some percentage of error to
the input values [7] or taking multiple input like cloud liquid water content, wind
gust, humidity, and temperature as a cause of rainfall.
As the rainfall prediction is very important considering agriculture of the country,
considering risk [8] due to landslide as a consequent effect of rainfall, this work
investigates the best modification of conventional statistical prediction models to
predict the rainfall over high rainfall hill region, and the developed model has been
validated by the remote sensing data of different time duration.
2 Study Area
See Fig. 1.
Fig. 1 Study area
P. Verma et al.
rainfall is extremely important. Some literatures compare the method of regression,
neural network, and clustering of data to get a better result in the prediction [3] of
seasonal event like rainfall. Some literature [4] compares the approaches based on
auto-regressive integrated moving average (ARIMA), the fuzzy time series (FST)
model, and the non-parametric method (Theil’s regression). In some work [5, 6],
conventional regression model is modified for the prediction of rainfall of different
nature by iterating the past values and thereby adding some percentage of error to
the input values [7] or taking multiple input like cloud liquid water content, wind
gust, humidity, and temperature as a cause of rainfall.
As the rainfall prediction is very important considering agriculture of the country,
considering risk [8] due to landslide as a consequent effect of rainfall, this work
investigates the best modification of conventional statistical prediction models to
predict the rainfall over high rainfall hill region, and the developed model has been
validated by the remote sensing data of different time duration.
2 Study Area
See Fig. 1.
Fig. 1 Study area