Effectiveness of Macroinvertebrate Species to Discern Pollution Levels in Aquatic Environment

The World is transforming more rapidly than ever before as a result of urbanization and industrialization. Such unrelenting destruction of nature has surpassed the capacity of mother Earth to support the aquatic ecosystem. Apart from freshwater macroinvertebrate species, there is no single measure of declining freshwater ecosystem that can capture either the short and long-term changes or the trend of overall freshwater ecosystem health. In that regard, the macroinvertebrates and physico-chemical variables were used as surrogates to determine levels of impairment within and between Pangani and Wami-Ruvu rivers’ basins in Tanzania. Spatial distribution of macroinvertebrate communities in the basins is significantly influenced by varying levels of environmental variables as a result of geomorphology and improper land uses. Principal Components Analysis (PCA) ordination showed two distinct patterns of biometrics that clearly discriminate reference sites from monitoring sites at each basin and consequently demonstrate the differences in water quality and physical habitat between the site categories. Similarly, distinctive macroinvertebrate species were observed and varied considerably among the site categories in the studied rivers as a function of tolerance levels. Impacted sites are characterized by either absence of any sensitive taxa or presence of few if any; greater dominance of only a few taxa that are tolerant to pollution. Therefore, the more diverse orders with a wider range of occurrences and tolerance to pollution (Ephemeroptera (E), Diptera (D), Odonata (O) and Trichoptera (T)) can be considered as potential bio-indicators in developing biomonitoring index for Tropical African Rivers as they showed a significant discriminating power that separated reference from monitoring sites.


Introduction
Tropical African rivers are subject to the most pressing requirements for improved attention to sustainable use due to the rapidly increasing anthropogenic pressure that threatens their ecological and socio-economic values [1] [2] [3] [4]. Freshwater macroinvertebrate species are at a high risk of extinction due to habitat degradation following overwhelming human activities (i.e. invasive industrialization, agriculture, and urban development) near rivers [4]- [9]. It is unlikely that there is a substantial number of freshwater bodies remaining that have not been irreversibly altered from their original state as a result of anthropogenic activities [10].
Improper land uses near rivers have caused significant changes in the flow regimes of rivers while altering negative impacts on the environment and loss of ecosystem functioning [8] [11]. The presence of human induced stressors (such as pollution, habitat destruction, and hydrological alterations) can directly impact freshwater habitat by significantly changing the biotic integrity and functional ability of a vast number of riverine ecosystems [12] [13]. In Tanzania, for example, Themi (near Arusha town), Karanga, Njoro and Rau (near Moshi town in Kilimanjaro) and Mzinga, Msimbazi, Yombo, and Kizinga (in Dar es Salaam) rivers were all found to be polluted by urban-based industrial and domestic wastes [6] [14] [15] [16]. Similarly, human induced activities (such as pollution, habitat transformation of landscape and hydrological alterations) have direct impacts on freshwater habitat as they significantly change biotic integrity and functional ability of many river ecosystems worldwide, particularly in urban and agricultural areas [8] [12] [13]. This is due to the fact that macro-invertebrate communities play the role of transforming organic inputs and as indicators of the quality, structural and functioning of aquatic environment. Apart from macroinvertebrates, there is no single measure of declining freshwater ecosystem that can capture either the short and long term changes or the trend of overall freshwater ecosystem health. In that regard, this study was designed to examine the correlation of macroinvertebrates with environmental variables and reveal their ability to discern reference sites from monitoring sites. Moreover, legislations should set ecological standards and quality objectives and make bio-monitoring programmes of aquatic ecosystems mandatory.

Description of Study Areas
Eighty-five (85) sampling sites of varying degradation levels along Pangani and Wami-Ruvu river basins were selected for sampling to ensure the characterization of macro-invertebrates and determination of physico-chemical parameters.
The basins provide a wide range of riverine systems, climate, geology, topography and human disturbance within different hydrological patterns. Pangani river basin is located in the north-east of Tanzania mainland, 36˚23'E -39˚13'E and

Sampling Design
The two river basins were divided into two site categories in which 39 reference (least impacted) and 46 monitoring (highly impacted) sites were established. Triplicate water and macro-invertebrate assemblage samples were collected at each site near the end of dry and wet seasons to capture the effect of respective seasons. The sampling sites were selected based on their ease of accessibility, presence and or absence of sustained anthropogenic disturbances, pools, riffles and runs, and degree of water physico-chemical and habitat degradation.

Macro-Invertebrate Samples
Macro-invertebrate samples were collected at the end of wet and dry seasons to capture the effect of respective seasons. Samples were collected throughout each sampling reach of 100 m by the same operator using a 30 × 30 cm kick-net with a 250 μm mesh size according to the Barbour et al. [20] method. To avoid bias due to spatial variations or patchiness, samples were collected in triplicates and at random locations within 200 m reach, making nine samples per reach or sampling site. The nine individual samples were pooled together as one composite sample was sorted grossly in the field at order level before preservation in 10% formaldehyde solution prior to transportation to the laboratory for identification [21]. In the laboratory, macro-invertebrate specimens were identified to the lowest possible taxonomic level) under the help of dissecting microscope (100× magnification) according to the method of Merritt and Cummins [22] and Thorp and Covich [23] in relation to the local conditions, followed by listing and counting of individuals.
Metrics were standardized into three score ranges based on the degree of impairment. Maximum score of 5 was assigned to little impaired sites, 3 for moderately impaired sites, and 1 for severely impaired sites. These scores are simply arbitrary standards according to Karr and Chu [1]. Standardized metric scores were then added to produce the B-IBI score on a 70-point scale (involving 14 characters each with a maximal score of 5) and 14-point score (involving 14 characters with minimal value of 1). The B-IBI values were then standardized to a 100-point scale: giving 68 to 100 (little impaired), 46 to 67 (moderately impaired), 20 to 45 (highly impaired) and 0 to 19 (severely impaired) as shown in Table 1 based on actual rating criteria prescribed by Pond et al. [25] and Arslan et al.  [26]. For the purpose of this study, streams/rivers B-IBI values below a score of 68 would be impaired (i.e., fair, poor and very poor).   PO -P − detected at Wami-Ruvu reference sites (Table 3).

Sites versus Biometric Data
B-IBI score was also weighted toward reference sites (on right side) in ordination since by definition all reference sites have good water and/or habitat quality.
Percentages of B-IBI score, EPT, Ephemeroptera, Trichoptera, Baetidae, and Plecoptera were weighted toward isolated reference sites on the right site of PCA ordination (Figure 1

Significant Tests on Differences between the Basins and Site Categories
Seven of the 30 biometric and environmental variables tested were significantly differentiating the basins and site categories, with p values < 0.05 (Table 4).

Water Quality Status of the Basins
The B-IBI score was used for this study as it combines several distinctive, stressinfluenced community characteristics into a single aggregate value that can be used to compare the level of stress evidenced by communities from different rivers localities. The B-IBI scores calculated from 14 biometric data resulted in categorization of sites based on their impairment levels with reference sites out-scoring monitoring sites at each basin. The scores indicated that 43.53% of the sampling sites  analyzed in the catchment basin presented very good water quality, 56.47% fair to good water quality (Table 5).

Discussion
Macro-invertebrate communities have become somewhat out of balance among site categories and the basins, both taxonomically and ecologically. Composition of macro-invertebrates and environmental variables were different not only between site categories and the basins but also presented seasonal variation. When comparing the taxonomic list of these two basins, macro-invertebrate organisms in Pangani seem to be more diverse and abundant. It was also possible to observe greater representativeness of some sensitive taxa like Coleoptera, and Ephemeroptera, and Trichoptera in all site categories of Pangani and Wami-Ruvu, respectively, which were not affiliated with high impacts. The results are in line with similar findings by Rosenberg and Resh [32], Bryne and Dates [33], Moog et al. [34], Compin and Céréghino [35], Morse et al. [36], Song et al. [37] and Foto Menbohan [38] who associated presence of coleopterans, ephemeropterans and trichopterans with good water quality and habitat suitability. The occurrence of these sensitive taxa in all site categories further suggests a possible improvement in the environmental conditions related to the decrease in concentrations of nutrients and changes in some of the physico-chemical parameters (e.g., dissolved oxygen, pH, temperature and electrical conductivity), and hence, reflecting in the greater richness of macro-invertebrate assemblages. Maximum values of water conductivity that reaches (7854.20 µS/cm), TDS (5838.70 mg/L), COD (928.01 mg/L), BOD (500.80 mg/L), 2 4 SO − (444.51 mg/L) and turbidity (104.44 NTU) and depletion of DO from 16.97 to 0.58 mg/L were registered at Wami-Ruvu monitoring sites and could have been responsible for the absence of Nepidae, Notonectidae and Lumnichidae families. Indeed, the absence of these families in Wami-Ruvu basin is undoubtedly due to differences in hydrological patterns between Pangani and Wami-Ruvu basins and levels of impairment caused by the uncontrolled discharge of domestic sewages, agrochemical inputs and industrial wastes in the rivers [39] [40]. These results are consistent with those of Compin and Céréghino [35], Song et al. [37] and Foto Men-Open Journal of Ecology bohan [38] who showed that a decrease in Coleoptera richness (i.e., Lumnichidae family) in human impacted rivers is clearly related to changes in water quality and habitat suitability. It can therefore be hypothesized that these taxa would have historically been present at Wami-Ruvu before human disturbances, as most of the environmental variables i.e., TDS and DO were found with values above recommended limits of 500 mg/L and 5 mg/L, respectively [41] [42]. Moreover, the results also agree with studies of Fuji [43] who reported the effect of environmental variables on the occurrence and distribution of macro-invertebrate organisms in freshwater ecosystem.
Statistical tests revealed a correlation of the metrics related to impacts with TDS, turbidity, COD, BOD, DO, temperature conductivity, potassium, sulphate, nitrogen, and phosphorus contents associated with improper land uses near basins [44] [45]. The two site categories of Pangani and Wami-Ruvu basins were clearly separated on PCA plot ordination, with two distinct patterns of biometrics that representing least and more pollution tolerant macro-invertebrate communities (Figure 1(a) & Figure 1(b)). Although, sensitivities of macro-invertebrates to pollution do vary [46], increase or decrease of physico-chemical variables beyond required limits is considered harmful to least tolerant living biota [23] [46] [47] [48]. For instance, the increased in nutrients beyond required limits are likely to cause the reduced occurrences of intolerant taxa (trichopterans and plecopterans) and favour the tolerant taxa (dipterans i.e., Chironomidae), which can strive better in low oxygenated conditions [23] [49] [50]. However, the dominance does not always reflect better environment, as mild disturbance may favor some tolerant taxa with subsequent reduction in sensitive taxa.
Reference sites of the two basins were prominently located in riffles and un-  [65], as Perlidae sp was rarely encountered and totally absent in severely degraded sites. In another sense, the dominance of certain taxa (i.e., Chironomidae and Naididae), and absence of the other (i.e., Plecoptera), at some sites can also be associated with habitat modification [8] [55] [66]. In summary, this provides further evidence to support the fact that, presence of human induced activities discharging various forms of pollutants especially nutrients into watersheds, can predict macro-invertebrates structure and function.
Therefore, more diverse orders with a wider range of occurrences and tolerance to pollution (Ephemeroptera (E), Diptera (D), Odonata (O) and Trichoptera (T)) can be considered potential bio-indicators in developing biomonitoring index for Tanzanian rivers as they showed a significant discriminating power that separated reference from monitoring sites.

Conclusion
With the aid of measured physico-chemical variables, all identified orders were useful in detecting disparities between site categories and basins at a family level.
However, increasing their taxonomic resolution to genus or species levels might improve or enhance the ability to detect differences among site categories and the two basins with respect to their macro-invertebrate assemblages and environmental variables. Updated ecological inventory and the taxonomic list (including distribution records and descriptions of new taxa) generated from this study will contribute to new effort of documenting existing macro-invertebrate species and development of regional identification guides and cost-effective biomonitoring index. It was recognized that more diverse orders with wider range