*4.3.4 Results of the statistical test*

The t-test results for each profile line is summarised in **Table 7**. Line H-H<sup>0</sup> shows a p-value that is less than 0.05, meaning that statistically, there is a significant difference between the depth values of the 2008 and 2014 data. It implies that some significant changes took place on the lagoon bed either through accretion or erosion around the profile section H-H<sup>0</sup> .

The fact that the p-values of the other three profile sections (G-G<sup>0</sup> , I-I<sup>0</sup> and J-J<sup>0</sup> ) are greater than 0.05 does not mean there is no change experienced between the gap year of the repeated data. The result of the ANOVA test shows F-ratio as 9.18 (**Table 8**), and this is the ratio of the between-group estimate to the within-group estimate. The p-value of the F-test is less than 0.05; this implies that there is a statistical significance between the means of the 8 variables at 95% confidence level. A multiple range test was carried out on the eight profiles, considering each profile

*Morphodynamics in a Tropical Shallow Lagoon: Observation and Inferences of Change DOI: http://dx.doi.org/10.5772/intechopen.90189*


**Table 8.**

*ANOVA table.*


## **Table 9.**

*Multiple range test.*

as a variable so as to determine which of the profile depth mean (average) is significantly different from the other (**Table 9**).

From the table of results on multiple range tests, six contrasts show a result that is significantly different, which implies significant variations in the depth of the 2008 and 2014 data sets. Further confirmation of the change is graphically displayed in **Figure 15**. The difference in the mean of the dataset on line H-H<sup>0</sup> that was overlaid on each other shows a wide variation. The variations in the mean values of the two datasets on the same profiles are very visible on profiles H-H<sup>0</sup> , I-I<sup>0</sup> and J-J<sup>0</sup> . It could be inferred from **Figure 15** that a mean depth of 3.1 m in 2014 against the mean depth of 2.5 m in 2008 shows erosion (whether by dredging or naturally) around and along the profile section H-H<sup>0</sup> . On the contrary, accretion (that is sediment gain) was shown from the region of profile H<sup>0</sup> to profile J.

*4.3.4 Results of the statistical test*

**Figure 13.**

**Figure 14.**

**98**

*Analysis of mean for all changes in depths from profile A-A*<sup>0</sup> *to J-J*<sup>0</sup>

*Lagoon Environments Around the World - A Scientific Perspective*

around the profile section H-H<sup>0</sup>

The t-test results for each profile line is summarised in **Table 7**. Line H-H<sup>0</sup> shows

*.*

are greater than 0.05 does not mean there is no change experienced between the gap year of the repeated data. The result of the ANOVA test shows F-ratio as 9.18 (**Table 8**), and this is the ratio of the between-group estimate to the within-group estimate. The p-value of the F-test is less than 0.05; this implies that there is a statistical significance between the means of the 8 variables at 95% confidence level. A multiple range test was carried out on the eight profiles, considering each profile

, I-I<sup>0</sup> and J-J<sup>0</sup>

)

a p-value that is less than 0.05, meaning that statistically, there is a significant difference between the depth values of the 2008 and 2014 data. It implies that some significant changes took place on the lagoon bed either through accretion or erosion

*Profile section G-G*<sup>0</sup> *showing degree of variation in the depths of the lagoon repeated bathymetric data.*

The fact that the p-values of the other three profile sections (G-G<sup>0</sup>

.

**Figure 16.** *Profile section I-I*<sup>0</sup> *showing degree of variation in the depths of the lagoon repeated bathymetric data.*

However, the ArcGIS model result in **Figure 16** confirms the region of accretion and erosion on the lagoon bed within that interval of 6 years. To put it differently, in a graphical representation, the changes on the lagoon bed moving in the direction west to east are depicted in **Figures 11–14.**

**4.5 Volume analysis**

**Figure 17.**

Volume estimates were calculated using CUTFILL tool in ArcGIS's 3D Analyst. The uncertainty inherent in the volume estimation using CUTFILL tool is computed in terms of percentage deviation (�5%). The depth values of the two repeated bathymetric datasets from 2008 and 2014 were used to determine how much sediment has been accumulated or eroded on any part of the lagoon water bed. The two dataset (2008 and 2014 bathymetric data) were plotted on ArcGIS and then converted to shapefiles, and next was the conversion of the shapefile to vectorbased digital geographic data using triangular irregular network (TIN) in order to make it a surface morphology. A TIN is a vector data structure that stores and displays surface models; it partitions geographic space using a set of irregularly spaced data points; each of which has x, y and z values. These points are connected by edges that form contiguous, non-overlapping triangles and create a continuous surface that represents the terrain. The CUTFILL tool in the ArcGIS environment was used to identify the areas where dredging/erosion and deposition/accretion

*Profile section J-J*<sup>0</sup> *showing degree of variation in the depths of the lagoon repeated bathymetric data.*

*Morphodynamics in a Tropical Shallow Lagoon: Observation and Inferences of Change*

*DOI: http://dx.doi.org/10.5772/intechopen.90189*

The single beam hydrographic data of 2008 and 2014 were used to determine the degree of changes that took place over a period of 6 years. Hence, the amount of sediment eroded or gained was calculated using the depth range from the datasets created on triangular irregular network (TIN). The TIN morphological surface was converted to raster data and was used in the CUTFILL tool to determine the volume gain or loss. To analyse the change in the sediment volume between 2008 and 2014, a statistical summary from the ArcGIS model was used. A summary of the gain/loss analysis is depicted in **Table 11**. The amount of accretion was found to be higher than that of erosion/dredging on the lagoon water bed despite all the local sand

It can be inferred from **Table 11** that 858,932 m2 on the lagoon gained 137,429 m<sup>3</sup> volume of sediment between 2008 and 2014. Hence, the depth of

have taken place in the study area on the lagoon (**Figure 16**).

*4.5.1 Calculation of volume gained*

**101**

extraction going on consistently in the lagoon.

accreted sediment over the area was computed as:

## **4.4 Overall test on the lagoon spatial depth characterisation**

Further to the statistical test carried out on west-east and south-north directional profiles, the differences in the depths of 2008 and 2014 data were extracted and arranged profile by profile. An ANOVA test with a posteriori comparison (**Table 10**) was carried out on the depth differences. The ANOVA decomposes the variances of all the datasets into two components: a between-group and a withingroup component. A high value of F-ratio (5.00) with p-value 0.00, therefore, is evidence against the null hypothesis that was originally set as equality of all the profile data set population means. Hence, there is a statistically significant difference in the lagoon bed between 2008 and 2014 derived from the repeated bathymetric surveys. The analysis of means plot with 95% decision limits revealing a high level of significant difference in profiles H-H<sup>0</sup> and I-I<sup>0</sup> . These were the two profiles that exceeded decision limits (**Figure 17**) that were set as 95% decision limits at both upper and lower limit of the mean.

It can be inferred from the results of the test that around the region of profiles H-H<sup>0</sup> and I-I<sup>0</sup> significant changes took place on the lagoon bed. Correlating the region between profile H-H<sup>0</sup> and profile I-I<sup>0</sup> with the erosion/accretion result in **Figure 16**, a high level of erosion or loss of sediment has taken place in the area, which is shown as a net loss in **Figure 16**.


**Table 10.** *ANOVA with a posterior test.* *Morphodynamics in a Tropical Shallow Lagoon: Observation and Inferences of Change DOI: http://dx.doi.org/10.5772/intechopen.90189*

**Figure 17.** *Profile section J-J*<sup>0</sup> *showing degree of variation in the depths of the lagoon repeated bathymetric data.*

## **4.5 Volume analysis**

However, the ArcGIS model result in **Figure 16** confirms the region of accretion and erosion on the lagoon bed within that interval of 6 years. To put it differently, in a graphical representation, the changes on the lagoon bed moving in the direction

*Profile section I-I*<sup>0</sup> *showing degree of variation in the depths of the lagoon repeated bathymetric data.*

Further to the statistical test carried out on west-east and south-north directional profiles, the differences in the depths of 2008 and 2014 data were extracted and arranged profile by profile. An ANOVA test with a posteriori comparison

. These were the two profiles

(**Table 10**) was carried out on the depth differences. The ANOVA decomposes the variances of all the datasets into two components: a between-group and a withingroup component. A high value of F-ratio (5.00) with p-value 0.00, therefore, is evidence against the null hypothesis that was originally set as equality of all the profile data set population means. Hence, there is a statistically significant difference in the lagoon bed between 2008 and 2014 derived from the repeated bathymetric surveys. The analysis of means plot with 95% decision limits revealing a high

that exceeded decision limits (**Figure 17**) that were set as 95% decision limits at

It can be inferred from the results of the test that around the region of profiles H-H<sup>0</sup> and I-I<sup>0</sup> significant changes took place on the lagoon bed. Correlating the region between profile H-H<sup>0</sup> and profile I-I<sup>0</sup> with the erosion/accretion result in **Figure 16**, a high level of erosion or loss of sediment has taken place in the area,

**Source Sum of squares Df Mean square F-Ratio P-Value** Between groups 70.1009 9 7.78898 **5.00 0.0000**

**4.4 Overall test on the lagoon spatial depth characterisation**

*Lagoon Environments Around the World - A Scientific Perspective*

level of significant difference in profiles H-H<sup>0</sup> and I-I<sup>0</sup>

Within groups 711.974 457 1.55793

both upper and lower limit of the mean.

which is shown as a net loss in **Figure 16**.

Total (corr.) 782.075 466

**Table 10.**

**100**

*ANOVA with a posterior test.*

west to east are depicted in **Figures 11–14.**

**Figure 16.**

Volume estimates were calculated using CUTFILL tool in ArcGIS's 3D Analyst. The uncertainty inherent in the volume estimation using CUTFILL tool is computed in terms of percentage deviation (�5%). The depth values of the two repeated bathymetric datasets from 2008 and 2014 were used to determine how much sediment has been accumulated or eroded on any part of the lagoon water bed. The two dataset (2008 and 2014 bathymetric data) were plotted on ArcGIS and then converted to shapefiles, and next was the conversion of the shapefile to vectorbased digital geographic data using triangular irregular network (TIN) in order to make it a surface morphology. A TIN is a vector data structure that stores and displays surface models; it partitions geographic space using a set of irregularly spaced data points; each of which has x, y and z values. These points are connected by edges that form contiguous, non-overlapping triangles and create a continuous surface that represents the terrain. The CUTFILL tool in the ArcGIS environment was used to identify the areas where dredging/erosion and deposition/accretion have taken place in the study area on the lagoon (**Figure 16**).
