Predicting Meander Migration of the Barak River by Empirical …
125
Fig. 5 Predicted radius
versus measured radius for
the time sequence maps
R² = 0.6418
0
200
400
600
800
1000
1200
1400
1600
0
200
400
600
800 1000 1200 1400
Predicted radius (m)
Measured radius (m)
and method was analyzed by considering two active parameters radius of curvature
and width of river even if this method gave a moderate correlation with R
2
= 0.50, the
scatter is moderately significant on the basis of data alone. So this method appears
to be a reasonably safe method, and it was found from the scatter plot that bend
curvature (r c /b) alone may be the main influencing parameter in the process river
migration.
Based on the previous year records of bends, a prediction tool ‘channel migration
toolbox’ was used for the prediction of bend radius and centroid for the year 2012 and
2017. Also the scatter between measured and predicted radius is significant (Fig. 5).
Indeed, as the GIS approach is effective on the decadal scale, the predictions for
2012 and 2017 were based on the bend records of approximately 10 years. The
comparisons are presented in Table 4. As can be seen the time sequence maps gives a
reasonably satisfactory prediction of the radius of the meander. From 1984 to 2017,
migration rates of the Barak River ranged from 1.4 to 58.3 m/year, with a mean and
standard deviation value of 13.01 m/year and 12.46 m/year, respectively, (Fig. 6).
5 Conclusion
In this study, three empirical equations Hooke [18], Brice [19], Nanson and Hickin
[7] and time sequence extrapolation are used to predict the meander migration rates.
12 meandering reaches of the Barak River are considered using multiperiod Landsat
remote sensing images to evaluate the accuracy of these methods by comparing
predicted and measured migration. Comparisons made from the empirical methods
indicate that Hooke and Brice’s method are reasonably traditional. On the basis of this
data alone, Nanson and Hickin method provides significant scatters, so this method
is reasonably safe. For the radius of meander, satisfactory predictions were given by
time sequence extrapolation method. This method is superior to empirical methods
because it gives a complete position of the meander. However, time sequence extrapolation method is more operator dependent than the empirical methods. The author
believes that this study will help in conducting the better planning of flood protection
125
Fig. 5 Predicted radius
versus measured radius for
the time sequence maps
R² = 0.6418
0
200
400
600
800
1000
1200
1400
1600
0
200
400
600
800 1000 1200 1400
Predicted radius (m)
Measured radius (m)
and method was analyzed by considering two active parameters radius of curvature
and width of river even if this method gave a moderate correlation with R
2
= 0.50, the
scatter is moderately significant on the basis of data alone. So this method appears
to be a reasonably safe method, and it was found from the scatter plot that bend
curvature (r c /b) alone may be the main influencing parameter in the process river
migration.
Based on the previous year records of bends, a prediction tool ‘channel migration
toolbox’ was used for the prediction of bend radius and centroid for the year 2012 and
2017. Also the scatter between measured and predicted radius is significant (Fig. 5).
Indeed, as the GIS approach is effective on the decadal scale, the predictions for
2012 and 2017 were based on the bend records of approximately 10 years. The
comparisons are presented in Table 4. As can be seen the time sequence maps gives a
reasonably satisfactory prediction of the radius of the meander. From 1984 to 2017,
migration rates of the Barak River ranged from 1.4 to 58.3 m/year, with a mean and
standard deviation value of 13.01 m/year and 12.46 m/year, respectively, (Fig. 6).
5 Conclusion
In this study, three empirical equations Hooke [18], Brice [19], Nanson and Hickin
[7] and time sequence extrapolation are used to predict the meander migration rates.
12 meandering reaches of the Barak River are considered using multiperiod Landsat
remote sensing images to evaluate the accuracy of these methods by comparing
predicted and measured migration. Comparisons made from the empirical methods
indicate that Hooke and Brice’s method are reasonably traditional. On the basis of this
data alone, Nanson and Hickin method provides significant scatters, so this method
is reasonably safe. For the radius of meander, satisfactory predictions were given by
time sequence extrapolation method. This method is superior to empirical methods
because it gives a complete position of the meander. However, time sequence extrapolation method is more operator dependent than the empirical methods. The author
believes that this study will help in conducting the better planning of flood protection