Mangroves in tropical and subtropical regions have been well documented in terms of the advantages they provide and their role in structuring ichthyofaunal assemblages, but little is known about their warm temperate counterparts. The study aimed to investigate the importance of warm temperate mangroves by comparing the abundance, diversity and distribution of small fishes in mangrove and non-mangrove estuaries in warm temperate South Africa. A 50×2 m (12-mm mesh) seine net was used over three summer seasons to sample small fishes in the Gonubie, Qora, Nahoon and Xhora estuaries (the latter two being mangrove estuaries). Fish abundance and diversity showed little variation among estuaries, despite the presence of mangroves. Estuaries in warm temperate areas are not only at the edge of mangrove distribution, but also offer alternative habitats which lend similar advantages to fish survival. It appears that warm temperate ichthyofauna have not yet evolved a dependence on mangrove systems in terms of the food, refuge and other ecological services they provide. Understanding the function of habitats and their value in enhancing fish survival in estuarine nursery areas is essential for fish conservation.
El papel de los manglares en la estructuración de las comunidades de peces en regiones tropicales y subtropicales está bien documentado, sin embargo, es poco conocido en las zonas templadas. El estudio investiga la importancia de los manglares de zonas templadas comparando la abundancia, diversidad y distribución de juveniles de peces en estuarios con y sin manglares de Sudáfrica. Se utilizó una red de cerco de 50 m (12 mm de malla) durante tres veranos para muestrear juveniles de peces en cuatro estuarios: Gonubie, Qora, Nahoon y Xhora (los dos últimos con manglares). La abundancia y diversidad de peces mostró una escasa variación entre estuarios, a pesar de la presencia o ausencia de manglares. Los estuarios de las zonas templadas están en los límites de distribución de los manglares y pueden, además, favorecer la supervivencia de los peces. Estos resultados implicarían que la ictiofauna no ha evolucionado todavía en sistemas de manglares en función de las ventajas (ej. alimento, refugio) que pueden proporcionar estos ecosistemas. El estudio de estos hábitats y su valor como refugio de los juveniles de peces es esencial para la conservación de estas especies.
Mangroves are distributed throughout the tropical regions of the globe, where their distribution is restricted to the 20°C winter seawater isotherm in both the southern and northern hemispheres (
Mangrove forests are considered one of the most productive of all marine and coastal ecosystems (
Mangroves form a vital component in the life history of many fish species in both tropical and subtropical regions (
Despite their economic and ecological importance, mangroves are under threat globally. Approximately 90% of mangroves occur in developing countries, where they are critically endangered and on the brink of local extinction in 26 known countries (
In South Africa, mangroves are restricted to the eastern coastline and can be found in 37 estuaries covering almost 1700 ha (
The role of mangroves in warm temperate regions remains relatively unstudied in terms of the advantages they provide (including a refuge/nursery habitat for larval- and juvenile-stage fishes and feeding opportunities). It is therefore important to investigate the role of warm temperate mangroves for fishes utilizing estuaries as nursery areas, especially since these vegetation types are under threat. This knowledge of ecosystem value will help to enable the proper conservation of habitats for fishes. The aim of the study was to investigate catches of juvenile and small adult fishes during the peak summer recruitment period in mangrove and non-mangrove estuaries to determine whether differences in catches exist, and whether mangrove presence lends an advantage to fish survival in warm temperate South Africa. It was hypothesized that mangrove estuaries would have a greater abundance and diversity of young fishes than non-mangrove estuaries.
Four estuaries (Nahoon, Gonubie, Qora and Xhora), all of which drain into the Indian Ocean, were selected within the warm temperate region of the Eastern Cape, South Africa (
Field sampling took place over a three-year period from 2015 to 2017, with data collected over the first-quarter moon phase in January of each year. Prior to sampling, five fixed sites were chosen remotely along the length of each estuary and marked using a GPS. Sites were spaced at one-kilometre intervals, with the first site being situated approximately 500 m from each estuary mouth. Physico-chemical measurements including temperature (°C), turbidity (NTU), salinity (PSU), conductivity (S m–1), pH, dissolved oxygen (mg L–1), and total dissolved solids were measured at each site using a YSI-6600 multimeter. Habitat type was also recorded at each site based on sediment and vegetation type. Six different habitat types were identified, including mud, mud and mangrove, mud and rock, mud and
Small fishes were sampled at each site using a 50x2 m seine net with a 12-mm stretched mesh. The seine net was deployed from a boat and pulled ashore, covering an estimated area of 400 m2. A consistent deployment of the seine net was maintained at each site, while a heavy sinker line allowed the net to be dragged through eelgrass (
All fishes were identified in situ to the species level, measured (in millimetres) and quantified prior to being released back into the estuary. Individuals which could not be identified in the field were placed in sample jars containing a 10% formalin solution for further identification in the laboratory. Fishes were identified in the laboratory by doing lateral line scale counts, as well as teeth counts for Mugilidae species following
Prior to statistical analyses, all factors were tested for normality and homogeneity of variance using the Shapiro-Wilk test and Levene test, respectively. Of the environmental variables, temperature and dissolved oxygen met the assumptions of normality and homogeneity of variance, and so did turbidity after square-root transformation. Therefore, parametric tests (one-way ANOVA and Tukey test) were used for these variables. The remaining environmental variables did not meet parametric assumptions, even after transformation, so non-parametric tests were used. The Kruskal-Wallis test was used to compare non-parametric environmental variables among years, estuaries, sites and habitats. The Mann-Whitney test was then used to further explore the data if the Kruskal-Wallis test returned a significant
Catch data were separated into marine and estuarine usage guilds, which were analysed independently to avoid any confounding effects of different estuary use by fishes that are resident versus immigrant. Fish communities were explored using diversity indices (Margalef species richness and Shannon-Wiener diversity) in PRIMER v.6 (
Generalized additive models were used to explore trends of species richness, abundance and diversity in relation to physico-chemical variables, as well as the influence that mangrove presence or absence and habitat type had on the same catch parameters. Catch data of fishes from each guild and of dominant species were added to the generalized additive models using a negative binomial distribution with log link. Rare species were removed from the model. The physico-chemical variables for each GAM were determined using forward stepwise variable selection. When habitat type was added to the model, dispersion was tested with ‘dispersiontest’ by
Within estuaries, horizontal physico-chemical variables exhibited a relatively uniform gradient (
Among the estuaries, the variables temperature, salinity, pH and dissolved oxygen varied significantly, though all four estuaries shared a similar climate regime due to their geographical location. Mean temperatures were significantly warmer in the more northern (Qora and Xhora) estuaries than in the Gonubie Estuary in the south of the study area (
Over the three-year study period, a total of 11625 fishes were caught among all four estuaries sampled, including fishes from both the marine and estuarine guilds. Catches consisted of 52 taxa represented by 26 families, with only 17 taxa making up 97% of the total catch (
Nahoon | Gonubie | Qora | Xhora | |||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Guild | Family | Species | Mean CPUE (range) | Mean length (mm) (range) | N | Total catch (%) | Mean CPUE (range) | Mean length (mm) (range) | N | Total catch (%) | Mean CPUE (range) | Mean length (mm) (range) | N | Total catch (%) | Mean CPUE (range) | Mean length (mm) (range) | N | Total catch (%) |
Estuarine | Ambassidae | 0 | 58.9 (47-73) | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-1) | 51.0 (51-51) | 1 | 0 | 0.3 (0-25) | 54.3 (42-67) | 25 | 5 | |
8.0 (0-601) | - | 601 | 56 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | |||
Atherinidae | 2.3 (0-140) | 66.4 (26-82) | 173 | 16 | 1.1 (0-43) | 70.6 (50-85) | 82 | 23 | 0 | - | 0 | 0 | 0.4 (0-22) | 67.5 (58-84) | 27 | 5 | ||
Clupeidae | 2.9 (0-74) | 62.0 (29-80) | 217 | 20 | 0.4 (0-7) | 54.4 (20-78) | 28 | 8 | 21.1 (0-943) | 60.9 (21-90) | 1580 | 83 | 4.3 (0-100) | 62.6 (50-90) | 320 | 62 | ||
Gobiidae | 0.5 (0-12) | 56.1 (22-80) | 37 | 3 | 2.0 (0-44) | 56.8 (27-94) | 152 | 42 | 2.0 (0-96) | 59.9 (28-98) | 152 | 8 | 1.4 (0-29) | 56.3 (28-90) | 104 | 20 | ||
0.0 (0-3) | 56.3 (41-76) | 3 | 0 | 0 | - | 0 | 0 | 0.1 (0-9) | 67.8 (49-95) | 9 | 0 | 0.1 (0-3) | 78.6 (67-94) | 5 | 1 | |||
0.2 (0-7) | 56.7 (41-77) | 14 | 1 | 0.1 (0-2) | 54.5 (44-64) | 4 | 1 | 0.2 (0-15) | - | 15 | 1 | 0 | - | 0 | 0 | |||
0.1 (0-3) | 69.6 (63-78) | 5 | 0 | 1.1 (0-36) | 65.3 (28-90) | 81 | 23 | 1.8 (0-54) | 71.7 (48-95) | 136 | 7 | 0.3 (0-9) | 64.8 (35-81) | 24 | 5 | |||
0 | - | 0 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-2) | 71.5 (65-78) | 2 | 0 | |||
0 | 76.0 (76-76) | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-1) | - | 1 | 0 | 0.1 (0-2) | 59.0 (54-61) | 4 | 1 | |||
0 | - | 0 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.1 (0-3) | 77.8 (60-95) | 5 | 1 | |||
0.3 (0-9) | 37.2 (20-60) | 21 | 2 | 0.1 (0-7) | 36.1 (28-49) | 11 | 3 | 0.3 (0-20) | 31.7 (24-42) | 20 | 1 | 0.0 (0-3) | 38.3 (37-41) | 3 | 1 | |||
Marine | Ariidae | 0 | - | 0 | 0 | 0.5 (0-34) | 66.1 (54-144) | 37 | 2 | 0.1 (0-6) | 64.0 (56-74) | 6 | 0 | 0 | - | 0 | 0 | |
Bothidae | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-2) | 70.5 (70-71) | 2 | 0 | 0.0 (0-1) | 94.0 (94-94) | 1 | 0 | ||
Carangidae | 0.1 (0-6) | 57.6 (47-71) | 7 | 0 | 0.0 (0-4) | - | 3 | 0 | 0.1 (0-3) | 81.0 (53-115) | 4 | 0 | 0.1 (0-4) | 70.0 (51-102) | 10 | 1 | ||
0.0 (0-1) | 110.0 (61-164) | 1 | 0 | 0.1 (0-3) | 126.5 (90-196) | 4 | 0 | 0.1 (0-4) | 171.6 (89-447) | 5 | 0 | 0 | - | 0 | 0 | |||
0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-1) | 64.5 (64-65) | 2 | 0 | 0 | - | 0 | 0 | |||
Gerreidae | 0.0 (0-1) | 85.0 (85-85) | 1 | 0 | 0 | - | 0 | 0 | 0.1 (0-7) | 68.6 (55-95) | 7 | 0 | 0 | - | 0 | 0 | ||
Haemulidae | 0.7 (0-14) | 118.1 (34-290) | 55 | 2 | 1.1 (0-51) | 88.8 (46-430) | 80 | 3 | 4.2 (0-171) | 121.9 (14-295) | 316 | 15 | 2.0 (0-56) | 83.1 (34-330) | 149 | 15 | ||
1.0 (0-47) | 66.1 (42-143) | 72 | 3 | 1.4 (0-37) | 61.7 (38-105) | 106 | 4 | 0 | - | 0 | 0 | 0.0 (0-3) | 28.3 (22-40) | 3 | 0 | |||
Leiognathidae | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-1) | 80.0 (80-80) | 1 | 0 | 0 | - | 0 | 0 | ||
Lutjanidae | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-1) | 37.8 (32-41) | 1 | 0 | 0 | - | 0 | 0 | ||
Monodactylidae | 0.2 (0-15) | 30.8 (21-40) | 15 | 1 | 0.0 (0-2) | 23.5 (19-28) | 2 | 0 | 0.1 (0-3) | 48.4 (25-74) | 8 | 0 | 0 | - | 0 | 0 | ||
0 | - | 0 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-2) | 68.0 (57-76) | 3 | 0 | |||
Mugilidae | 1.7 (0-43) | 136.9 (50-256) | 129 | 6 | 0.9 (0-50) | 197.0 (30-310) | 68 | 3 | 0.3 (0-9) | 155.3 (78-280) | 20 | 1 | 0.0 (0-1) | 92.0 (92-92) | 1 | 0 | ||
0.7 (0-25) | 98.5 (43-175) | 55 | 2 | 1.1 (0-27) | 116.7 (35-210) | 84 | 3 | 0.4 (0-26) | 77.1 (60-247) | 28 | 1 | 0.1 (0-4) | 95.5 (56-180) | 5 | 0 | |||
0.3 (0-13) | 177.6 (55-281) | 23 | 1 | 0.5 (0-14) | 146.5 (30-390) | 40 | 2 | 0.1 (0-7) | 99.3 (61-108) | 8 | 0 | 0.1 (0-6) | 119.4 (47-312) | 11 | 1 | |||
0.1 (0-2) | 94.5 (64-127) | 4 | 0 | 0.1 (0-4) | 75.1 (45-176) | 7 | 0 | 0.1 (0-5) | 63.6 (59-67) | 5 | 0 | 0.0 (0-1) | 181.0 (181-181) | 1 | 0 | |||
0.0 (0-2) | 143.0 (128-158) | 2 | 0 | 0 (0-1) | 144.5 (33-256) | 2 | 0 | 0.0 (0-1) | 126.0 (126-126) | 1 | 0 | 0.0 (0-2) | 101.7 (50-200) | 2 | 0 | |||
0.1 (0-4) | 98.5 (62-175) | 7 | 0 | 0.8 (0-37) | 99.2 (28-142) | 62 | 3 | 0.3 (0-9) | 149.4 (23-286) | 24 | 1 | 0.0 (0-1) | 41.0 (34-48) | 3 | 0 | |||
Paralichthyidae | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-1) | 72.0 (72-72) | 1 | 0 | 0 | - | 0 | 0 | ||
Platycephalidae | 0.0 (0-1) | - | 1 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-2) | 430.0 (350-510) | 2 | 0 | ||
Pomatomidae | 0 | - | 0 | 0 | 1.8 (0-135) | 74.5 (60-84) | 138 | 6 | 0.0 (0-1) | 67.0 (67-67) | 1 | 0 | 0.0 (0-2) | 94.7 (88-106) | 3 | 0 | ||
Rhinobatidae | 0.0 (0-2) | 720.0 (720-720) | 2 | 0 | 0.0 (0-1) | 760.0 (760-760) | 1 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | ||
Sciaenidae | 0.0 (0-1) | 263.3 (234-286) | 3 | 0 | 0 | - | 0 | 0 | 0.1 (0-4) | 221.9 (109-405) | 8 | 0 | 0.1 (0-5) | 182.5 (64-390) | 11 | 1 | ||
Siganidae | 0.0 (0-2) | 46.5 (46-47) | 2 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | ||
Soleidae | 0.7 (0-23) | 47.7 (37-60) | 50 | 2 | 0.3 (0-6) | 55.6 (23-164) | 21 | 1 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | ||
1.5 (0-47) | 55.2 (33-80) | 109 | 5 | 0.6 (0-16) | 57.0 (35-97) | 46 | 2 | 0.9 (0-24) | 57.3 (23-95) | 71 | 3 | 1.0 (0-19) | 52.6 (26-77) | 74 | 7 | |||
Sparidae | 0.8 (0-19) | 50.1 (38-62) | 58 | 3 | 1.7 (0-75) | 59.7 (21-151) | 125 | 5 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | ||
0.0 (0-1) | 116.0 (54-178) | 2 | 0 | 0.1 (0-5) | 57.6 (41-125) | 7 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | |||
0.0 (0-2) | 177.5 (165-190) | 2 | 0 | 0.0 (0-1) | 45.0 (45-45) | 1 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | |||
0.2 (0-14) | 90.8 (36-170) | 14 | 1 | 0.1 (0-4) | 78.8 (50-110) | 4 | 0 | 0 | - | 0 | 0 | 0 | 80.0 (65-95) | 0 | 0 | |||
21.4 (0-487) | 86.2 (23-207) | 1607 | 72 | 16.1 (0-549) | 83.3 (11-420) | 1210 | 50 | 21.1 (5-406) | 74.9 (12-184) | 1586 | 75 | 9.8 (0-423) | 62.9 (28-130) | 733 | 72 | |||
0 | - | 0 | 0 | 4.2 (0-312) | 71.1 (58-90) | 313 | 13 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | |||
0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-1) | 310.0 (310-310) | 1 | 0 | 0.0 (0-1) | 287.0 (287-287) | 1 | 0 | |||
Sphyraenidae | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-2) | 97.3 (90-110) | 2 | 0 | ||
Terapontidae | 0.0 (0-2) | 166.0 (148-184) | 2 | 0 | 0 | - | 0 | 0 | 0.1 (0-3) | 68.0 (25-120) | 5 | 0 | 0.0 (0-1) | 75.0 (75-75) | 1 | 0 | ||
Tetraodontidae | 0.0 (0-1) | - | 1 | 0 | 0.1 (0-4) | 56.0 (32-93) | 6 | 0 | 0 | 74.0 (74-74) | 0 | 0 | 0 | 55.0 (55-55) | 0 | 0 | ||
0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.0 (0-1) | 67.0 (67-67) | 1 | 0 | 0 | - | 0 | 0 | |||
0 | - | 0 | 0 | 0.5 (0-21) | 107.4 (61-170) | 34 | 1 | 0.0 (0-1) | 45.0 (43-47) | 2 | 0 | 0 | - | 0 | 0 | |||
Torpedinidae | 0.0 (0-1) | - | 1 | 0 | 0.0 (0-2) | 301.0 (33-440) | 3 | 0 | 0 | - | 0 | 0 | 0 | 321.0 (308-348) | 0 | 0 | ||
Trichiuridae | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0 | - | 0 | 0 | 0.1 (0-4) | 4 | 0 |
Within the estuarine guild, significant differences in size were recorded among estuaries for
In the marine guild, significant differences in length were recorded for
Fish community composition and habitat use
Species diversity showed a decreasing trend from the mouth towards the upper reaches within each estuary, but there were no significant differences among estuaries or among habitats (
Nahoon Estuary had the greatest habitat complexity at the sites sampled, with a total of five different habitat types. The Qora and Xhora estuaries each had four recorded habitat types, while the Gonubie Estuary showed the lowest habitat complexity with only three habitat types. The sand and mangrove habitat in the Nahoon Estuary had the greatest mean species diversity (H’=1.8), followed by the sand and mud habitats in the Gonubie and Qora estuaries, respectively (H’=1.6). Mud habitat had the most consistently high mean species diversity across all four estuaries sampled.
There was no significant difference in the catch per unit effort (CPUE) of fishes within both the estuarine and marine guilds among habitat types (
The habitats contributing the highest percentage of the total CPUE (combining fishes from all guilds) included mud and
When the CPUE of dominant species from the estuarine guild was compared between habitat type, it was found that mud and rock habitat contributed the highest percentage of the overall CPUE of dominant species (13%), followed by mud (8%) and mud and
Within the marine guild, mud and
Relationship between fish abundance and environmental variables
Generalized additive models were used to explore the influence of physico-chemical parameters and habitat type on species distributions. The presence of mangroves was included as a factor in all generalized additive models. The response variables analysed were species richness of all taxa and within guilds, as well as the abundance of dominant species. Species diversity was excluded from the models due to the high similarity among estuaries and habitat types.
The abundances of fishes within the marine guild were best described by a model using conductivity (
Grouping (no. species) | Deviance explained (%) | Significant variable(s) |
---|---|---|
Abundance | ||
All taxa (52) | 31.1 | con* |
Estuarine (12) | 65.1 | temp* sal** hab*** |
Marine (30) | 17.5 | con** |
Dominant species | ||
70.6 | temp** DO** hab*** | |
79.4 | temp*** sal*** DO*** hab*** | |
68.1 | NTU** TDS* | |
65.9 | temp*** sal*** hab* | |
44.5 | sal* | |
63.7 | temp** sal*** | |
75.6 | con* | |
13.4 | temp* | |
81.5 | temp*** sal** | |
74.2 | NTU*** DO *** hab** | |
23.3 | sal** | |
14.5 | temp** |
For the dominant species within the estuarine guild, the abundance of
The study aimed to investigate the importance of mangroves for small fishes in warm temperate South Africa. This was the first study of its kind in the region, and it was unknown whether mangroves play an important role in structuring fish communities as their tropical and subtropical counterparts do. Physico-chemical variables measured showed little variation among the four estuaries sampled, because the estuaries fall within the same climatic region (
The regulation of freshwater flow into estuaries has been identified as a potential threat to estuarine ecosystem structure and function, and to the productivity of fisheries in particular (
Although numerous studies note that the abundance and diversity of fishes is greater in mangrove habitats in tropical areas (
Additionally, all four estuaries sampled had a number of shallow mud and sand banks, both of which made large contributions to the overall percentage of CPUE of dominant species. It is thought that these shallow habitats could offer a refuge for young fishes that are vulnerable to predation.
Ultimately, estuaries with an availability of a variety of nursery habitat types are more beneficial to the survival of young fishes and are able to support a greater abundance and diversity of ichthyofauna than estuaries with low habitat complexity (
Although no significant differences in abundance and diversity were found between mangrove and non-mangrove estuaries in the study, it is important to not write off mangroves as significant refuge and habitat providers in warm temperate regions. The loss of habitat provided by mangroves has been found to significantly reduce the abundance and diversity of ichthyofaunal assemblages, which could potentially have cascading effects at higher trophic levels, leading to severe consequences for fisheries and food production (
The present study provides preliminary insights into the use of warm temperate mangroves by small fishes. In the event of future studies, it is suggested that a greater number of warm temperate mangrove estuaries be sampled to provide a larger database to assess the importance of mangroves in warm temperate regions in greater depth. Knowledge on the function of habitats and their role in enhancing fish survival in estuarine nursery areas is a crucial asset for fish conservation.
The authors extend their thanks to Eugin Bornman for imparting his statistical knowledge, and to Cuen Muller, Eugin Bornman, Kyle Hewett and Taryn Smit, who assisted with data collection and fieldwork. Thanks are also given to the Nelson Mandela University for providing infrastructural support and equipment for the duration of this study.