Our survey found out whether fishing debris impacted coral reefs or not. The results revealed that fishing debris significantly impacted coral health in all the surveyed sites in the Androth reefs of Lakshadweep. A total of 4 turtles were observed entangled in the fishing debris at two sites, and 34 fish were also found entangled during the survey period. Moreover, coral reefs entangled in the fishing ghost nets were calculated based on the coral genera. The site-wise assessment of coral health entangled in fishing ghost nets was calculated in percentage terms (Fig. 2)
The results of Kruskal-Walli’s test showed that there was a statistically significant difference between the medians of coral health for the different depths (p < 0.0001). Further pairwise comparisons showed that the mean rank of coral health was significantly lower for fish net-entangled corals at each depth (p < 0.0001). This meant that the coral health was significantly worse in the coral entangled with fishing debris (Table 1). Chi-Square = 49.67044, DF = 5, p < 0.0001.
Table 1
The Kruskal Wallis test results data from fishing ghost net debris with coral health groups
|
Mean Rank Difference
|
Z
|
Prob
|
Sig
|
"Depth" "DC"
|
-23
|
-2.94501
|
0.04844
|
p < 0.001
|
"Depth" "R"
|
-38.8
|
-4.96811
|
< 0.0001
|
p < 0.001
|
"Depth" "FNT"
|
-50
|
-6.4022
|
< 0.0001
|
p < 0.001
|
"HC" "FNT"
|
-30.3
|
-3.87973
|
0.00157
|
p < 0.001
|
"DC" "FNT"
|
-27
|
-3.45719
|
0.00819
|
p < 0.001
|
"FNT" "AA"
|
31.5
|
4.03339
|
0.000824683
|
p < 0.001
|
*DC- Dead coral, R- Rubbles, FNT-Fishing net, HC- Healthy coral
Pearson's correlation analysis revealed that the fish net debris correlated significantly with the coral's health (HC, DC, R, AA) at all the depths sampled (Table 2). Therefore, the fishing ghost net debris significantly impacted the coral reefs' health
Table 2
Pearson's Correlations analysis results between the Fishing ghost net debris and coral health
characteristic
|
characteristic
|
Pearson's r
|
p
|
Lower 95% CI
|
Upper 95% CI
|
Depth
|
HC
|
-0.863
|
< .001
|
-0.967
|
-0.512
|
Depth
|
DC
|
0.904
|
< .001
|
0.637
|
0.977
|
Depth
|
R
|
0.349
|
0.323
|
-0.36
|
0.802
|
Depth
|
FNT
|
0.907
|
< .001
|
0.645
|
0.978
|
Depth
|
AA
|
0.092
|
0.801
|
-0.571
|
0.682
|
HC
|
DC
|
-0.978
|
< .001
|
-0.995
|
-0.908
|
HC
|
FNT
|
-0.934
|
< .001
|
-0.985
|
-0.74
|
HC
|
AA
|
-0.32
|
0.368
|
-0.79
|
0.388
|
DC
|
R
|
0.53
|
0.115
|
-0.149
|
0.87
|
DC
|
FNT
|
0.969
|
< .001
|
0.872
|
0.993
|
DC
|
AA
|
0.358
|
0.31
|
-0.351
|
0.806
|
R
|
FNT
|
0.622
|
< .001
|
-0.013
|
0.899
|
FNT
|
AA
|
0.292
|
0.414
|
-0.414
|
0.778
|
*HC- Healthy coral, DC- dead coral, R- Rubbles, AA- algae assemblage, FNT – Fishing ghost net size
Table 3 shows the mean zooxanthellae densities in each coral genera from all the sites with and without entangled fishing ghost nets sampled during the survey periods. The same is presented in central tendency box plots (Fig. 3).
Table 3
shows the zooxanthellae mean density difference between the fish net entangled and not entangled corals in genus-wise overall sites.
Zooxanthellae density
|
Genus
|
N
|
Mean
|
SD
|
Coefficient of variation
|
Fish net-entangled corals
|
Acropora
|
10
|
5.784
|
3.195
|
0.552
|
|
Dipsastraea
|
10
|
5.077
|
3.108
|
0.612
|
|
Favites
|
10
|
5.149
|
3.833
|
0.744
|
|
Goniopora
|
10
|
4.627
|
2.498
|
0.54
|
|
Lobophyllia
|
10
|
5.649
|
3.212
|
0.569
|
|
Montipora
|
10
|
5.278
|
3.192
|
0.605
|
|
Platygyra
|
10
|
5.456
|
3.57
|
0.654
|
|
pocillopora
|
10
|
5.436
|
3.573
|
0.657
|
|
Porites
|
10
|
5.352
|
3.31
|
0.618
|
|
Turbinaria
|
10
|
5.52
|
3.447
|
0.624
|
Fish net-Not entangled corals
|
Acropora
|
10
|
6.022
|
3.292
|
0.547
|
|
Dipsastraea
|
10
|
5.381
|
3.187
|
0.592
|
|
Favites
|
10
|
5.474
|
3.819
|
0.698
|
|
Goniopora
|
10
|
4.948
|
2.521
|
0.509
|
|
Lobophyllia
|
10
|
6.049
|
3.076
|
0.509
|
|
Montipora
|
10
|
5.485
|
3.192
|
0.582
|
|
Platygyra
|
10
|
5.608
|
3.565
|
0.636
|
|
pocillopora
|
10
|
5.749
|
3.486
|
0.606
|
|
Porites
|
10
|
5.657
|
3.275
|
0.579
|
|
Turbinaria
|
10
|
5.768
|
3.487
|
0.605
|
Table 4
shows two ANOVA results of zooxanthellae density from all sites with and without entangled fishing ghost nets among genus-wise coral colonies analyzed using two-way ANOVA (F = 64.342, p < 0.001) whereas site-wise (F = 54.038, p < 0.001)
Within Subjects Effects
|
|
|
|
|
|
Cases
|
Sum of Squares
|
df
|
Mean Square
|
F
|
p
|
Zooxanthellae density (Entangles and Not entangled corals)
|
3.956
|
1
|
3.956
|
64.342
|
< .001
|
Zooxanthellae density ✻ Sites
|
29.906
|
9
|
3.323
|
54.038
|
< .001
|
Residuals
|
5.534
|
90
|
0.061
|
|
|
Between Subjects Effects
|
|
|
|
|
|
Cases
|
Sum of Squares
|
df
|
Mean Square
|
F
|
p
|
Sites
|
1274.115
|
9
|
141.568
|
18.73
|
< .001
|
Residuals
|
680.245
|
90
|
7.558
|
|
|
Further, a multiple regression model was used to identify how fishing debris impacted coral health. (Total entanglements observed at sites, n = 10, Mean incident catch rates (95% CI) for healthy corals, dead corals, algae assemblages, rubbles). The multiple regression model also supported the significance p value p < 0.001(Table 5).
Table 5
ANOVA of Multiple linear regression model analysis fishing ghost net entangled corals health assessments data.
Model
|
|
Sum of Squares
|
df
|
Mean Square
|
F
|
p
|
H₁
|
Regression
|
10010.857
|
4
|
2502.714
|
34.447
|
< .001
|
|
Residual
|
363.268
|
5
|
72.654
|
|
|
|
Total
|
10374.125
|
9
|
|
|
|
Table 6
The results of the Multiple linear regression model analysis of fishing ghost net entangled coral health assessments data show a significant association between fishing ghost net debris with entangled coral health.
Coefficients
|
|
|
|
|
|
|
95% CI
|
95% CI
|
Model
|
|
Unstandardized
|
Standard Error
|
Standardized
|
t
|
p
|
Lower
|
Upper
|
H₀
|
(Intercept)
|
206.25
|
10.736
|
|
19.211
|
< .001
|
181.963
|
230.537
|
H₁
|
(Intercept)
|
31.504
|
70.728
|
|
0.445
|
0.675
|
150.308
|
213.316
|
|
DC
|
3.108
|
0.993
|
1.303
|
3.128
|
0.026
|
0.554
|
5.661
|
|
R
|
0.17
|
0.113
|
0.149
|
1.508
|
0.192
|
-0.12
|
0.461
|
|
AA
|
-0.099
|
0.13
|
-0.069
|
-0.758
|
0.483
|
-0.433
|
0.236
|
|
HC
|
0.71
|
0.732
|
0.396
|
0.97
|
0.376
|
-1.172
|
2.593
|
Table 7
Fishing ghost net with other plastic debris found in each site with depths during the survey periods.
location
|
Depth
|
F.net 6mm
|
F.net 5mm
|
Rope mm
|
Buoy
|
Line
|
SW1
|
7
|
2
|
1
|
3
|
2
|
4
|
SW2
|
7.3
|
3
|
3
|
4
|
2
|
2
|
SW3
|
7.5
|
3
|
1
|
4
|
5
|
1
|
SW4
|
8
|
2
|
2
|
3
|
2
|
0
|
S 5
|
8
|
3
|
1
|
2
|
5
|
0
|
S 6
|
8.3
|
4
|
0
|
4
|
0
|
2
|
NW 7
|
9
|
3
|
2
|
4
|
2
|
1
|
NW 8
|
9
|
3
|
0
|
3
|
1
|
0
|
W 9
|
10
|
2
|
0
|
2
|
0
|
1
|
W 10
|
11
|
2
|
1
|
2
|
0
|
0
|
In the overall sites during the survey periods, we found a percentage of abandoned fishing gear of 6mm fishing ghost nets (28%) 5mm fishing ghost nets (11%) rope (31%), buoy (19%), and line (11%). (Fig. 5)
The raw data collected consisted of the abundance of coral colonies in healthy condition (HC), dead condition (DC), fishing ghost net size (FNT), and R (Rubbles) colonies with algal assemblage (AA) associated with the reef habitat. The abundance data collected in this study used line intersect transect (300 m2).
Principal component analysis (Fig. 6) showed that there was strong variability (PC1: 68.3% variance) between the groups of sites (Number of sites − 10). In this, the contribution of coral colonies affected by fishing ghost net debris (FNT) was observed to be higher in site no. S 5, S 6, NW 8 followed by NW 7 than in the other sites and the corresponding variability was calculated as 68.3%.
Table 8
Abandoned fishing ghost net was removed from entangled corals in three sites.
SL
|
location
|
Depth
|
Fishing ghost net weight in Kg
|
rope weight/kg
|
1
|
Sw2
|
8
|
12.5
|
2
|
2
|
SW3
|
9
|
10.25
|
1
|
3
|
NIOT
|
11
|
48
|
4
|