Ecosystem health assessment of three inland water bodies in South-west, Nigeria based on sh diversity, pollution status, ecological and health risk indices

7 This study was conducted to determine the health status of three water bodies (Badagry Creek, 8 Ologe Lagoon and River Owo) because of the large amount of effluent they receive from industries 9 around Lagos as well as the services they provide to sustain the large human population in an 10 emerging mega city like Lagos. Water, sediment and fish samples were collected monthly from 11 the three water bodies between January and December, 2018. Standard methods were used for the 12 analysis of physico-chemical parameters, heavy metals, length-weight relationship, condition 13 factor, fish diversity indices, sediment pollution indices, ecotoxicology of heavy metals in 14 sediment and potential ecological risks as well as health risk assessment of heavy metals. The 15 geoaccumulation index (Igeo) of heavy metals in sediments of the sampling sites ranged from - 16 12.14 to -0.38. The mean quotients using the probable effect level (m-PEL-Q) are 3.91 x10-4, 4.77 17 x10-4 and 7.87 x10-4 for Ologe Lagoon, Badagry Creek and River Owo respectively. The trend 18 was the same with mean quotients using effect range-median (m-ERM-Q). The estimated daily 19 intake (EDI) ranged from 0.00 mgkg-1day-1 in Pb from River Owo to 1.15 x10-3 mgkg-1day-1 in 20 Fe still from River Owo. The range of values of the target hazard quotient (THQ) of the metals in 21 Badagry Creek, River Owo and Ologe Lagoon are 1.23x10-4 - 1.65x10-2, 0.00 - 1.64x10-2 and 22 5.76x10-5 - 1.65x10-2 respectively. The study showed that the three aquatic ecosystems are 23 healthy but require regular monitoring to promptly detect sudden changes in their health status. Findings : The three aquatic ecosystems are healthy but they require regular monitoring to 26 promptly detect sudden changes in their health status.


INTRODUCTION
the dynamic nature of aquatic ecosystems. However, there is a consensus opinion on what a healthy 78 ecosystem should be (Palmer and Febria, 2012). Ecosystem health assessment methods have also 79 varied and in most cases depended on research objectives, available resources and the discipline 80 of the authors, with the last factor being the most influential of the three. Ecosystem health can be immediately after collection in a cooler to ensure that the physical properties of the water samples 118 were maintained. Temperature, pH, conductivity, salinity, turbidity and dissolved oxygen of 119 samples were measured in-situ using a mercury-in-glass thermometer, digital pH meter (model: 120 Hanna HI98107), HACH-HQ40D portable multi-meter, turbidity meter (HACH 2100Q) and 121 portable dissolved oxygen meter (HI9146) respectively. Ammonia, total suspended solids (TSS), 122 total dissolved solid (TDS), total solid (TS), total alkalinity, biochemical oxygen demand (BOD), 123 carbon dioxide, and chlorophyll were determined in the laboratory using methods described by 124 American Public Health Association (APHA, 1985).  (Rickter, 1973) (1) 162 Where W = weight (g), L = length (cm), a (y-intercept) = the initial growth coefficient, and b 163 (slope) = the growth coefficient.

164
The values of constants a and b were estimated after logarithmic transformation of Eq. (1) using 165 least square linear regression (Zar, 1984) to give: Before the regression analysis of log W on log L, log-log plots of length and weight values were 170 performed for visual inspection of outliers (Froese, 2006 Where SE is the standard error of b.

178
In order to confirm whether b values obtained in the linear regressions were significantly different 179 from the isometric value of ±95% CI of b at α = 0.05, t-test was applied as expressed by the 180 equation according to Sokal and Rohlf (1987): where ts is the t-test value, b = slope and SE = the standard error of the slope (b).

186
All the statistical analyses were considered at significance level of 5% (p<0.05 to remove plant materials, stones, and other unwanted particles (Varol, 2011). In order to obtain  The powdered fish sample (0.5 g) was added to a mixture of 6 ml HNO3 (65% Suprapur, Merck,  procedure had good precision, which was calculated as the relative standard deviation (RSD) and 257 the values obtained ranged from 6 and 9%. The analysis of standard solution had precision value 258 that is better than 5%. Each of the analyses was repeated twice, and the results obtained were 259 reported as the average. Metal contents were expressed as mg/kg dry weight.  where ni is the number of observations from the sample in the i th of k (non-empty) categories 284 and n is the sample size.

286
Fish species richness in the sites was evaluated using two indices; menhinick's and margalef's 287 indices.

293
Where S is the total number of species and N is the total number of individuals.

295
Fisher's alpha is a diversity index, defined implicitly by the formula: 298 299 where S is number of species, n is number of individuals and a is the Fisher's alpha.

301
The Berger-Parker index equals the maximum pi value in the dataset or sampling station, i.e. the 302 proportional abundance of the most abundant species.

304
Where pi = ni/N as has been earlier expressed. where Cn is the measured concentration of the metals in the sediment samples and Bn is the where CF is the contamination factor, n is the number of metals studied. of a given study site: The following classification of modified degree of contamination in sediments has been proposed: 348 very low contamination (mCd < 1.5); low degree of contamination (1.5 ≤ mCd < 2); moderate 349 degree of contamination (2 ≤ mCd < 4); high degree of contamination (4 ≤ mCd < 8); very high 350 degree of contamination (8 ≤ mCd< 16); extremely high degree of contamination (16 ≤ mCd < 32) 351 and ultra high degree of contamination (mCd ≥ 32). The potential ecological risk posed by the metals was assessed by the method described by 372 Håkanson (1980). The formulae are: 377 378 where R1 is the sum of all risk factors for all metals in sediments, E i r is the monomial potential 381 ecological risk factor, T i r is the toxic-response factor for a given metal/substance, C i r is the 382 contamination factor, C i 0 is the concentration of metals in the sediment samples, and C i n is a 383 reference/background value for metals.  where Cmetal is the concentration (mg kg -1 ) of the heavy metals in the muscle tissue of C. zillii, DNI 396 is the daily nutritional intake in (g day -1 ), and C f is the factor for conversion of fresh fish tissues 397 to dry constant weight.

Length-weight relationship and condition factor 475
A total of 1,367 individuals belonging to 21 genera were analysed in this study. In Badagry Creek, 476 13 genera were encountered while River Owo and Ologe Lagoon recorded 8 and 11 genera 477 respectively. Three species were common to the 3 sampling stations; Chrysichthys nigrodigitatus, 478 Coptodon zillii and Sarotherodon melanotheron. The species, number of specimens, length-weight 479 relationship parameters a and b, condition factor (K), 95% confidence interval for b, correlation 480 coefficient (r), and growth type (allometric or isometric) are presented in Table 2. The sample size

Sediment pollution indices 521
The enrichment factors of the heavy metals in sediment of the sampling stations vary from 4.05x10 -522 4 in Cu from River Owo to 8.65x10 -1 in Cd from Ologe Lagoon (Table 4). The geoaccumulation 523 index (Igeo) of heavy metals in sediments of the sampling sites ranged from -12.14 to -0.38 (Table   524 4). Contamination factor (CF), pollution load index (PLI) and modified degree of contamination 525 (mCd) of metals in sediments of the 3 sampling stations are also shown in     The health risk associated with the consumption of the muscle of Coptodon zillii was evaluated by 575 calculating the estimated dietary intake (EDI) and target hazard quotients (THQ) ( Table 6). The

576
EDI ranged from 0.00 mgkg -1 day -1 in Pb from River Owo to 1.15 x10 -3 mgkg -1 day -1 in Fe still from 577 River Owo. The EDI for Zn and Cu followed the same pattern as does Pb and Cd but the EDI for 578 As was the same value (8.41 x10 -7 mgkg -1 day -1 ) in all the sampling stations. Table 6   One of the indices used to assess aquatic ecosystem health is the physico-chemistry of the all the fish species from the three sampling stations were within the expected range of 2.5 to 3.5 636 (Froese, 2006   analyse their results, the absence of a good number of these species (more than 50%) indicates that 670 the environment may not be as conducive as it used to be or the fish stock has been over-exploited.

671
Oral evidence from fishers in the sampling stations states that their catches have progressively 672 dwindled over time forcing some of them to take up alternative or additional sources of livelihood.

673
They attributed this to intense sand mining activities in the sampling stations. In addition, the 674 presence of Agbara Industrial Estate in Ogun State, which empties its effluent into Ologe Lagoon 675 and Badagry Creek could also be responsible for the loss in fish species and reduction in catch by  the studied sites because their potential ecological risk indices (E 1 r) (5.75x10 -4 -6.90) are <40.

731
However, E 1 r for Cd was considerably higher than the values obtained for other metals. Yi et al 732 (2011) opined that the high ecological risk of Cd in aquatic ecosystems is due to its high toxic-733 response factor. The potential ecological risk indices for individual metals stressors (E 1 r) indicated 734 that intensity of pollution of the five metals decreased in the following order: Cd>Pb>As>Cu>Zn.