Spatio-temporal dynamics and population biology of the Fly River Herring (Nematalosa papuensis): implications for freshwater lake management in Papua New Guinea

Figa, Boga Soni (2014) Spatio-temporal dynamics and population biology of the Fly River Herring (Nematalosa papuensis): implications for freshwater lake management in Papua New Guinea. PhD thesis, James Cook University.

PDF (Thesis)
Download (3MB) | Preview


In the face of continuous threats to the freshwater systems of the world from waste of anthropogenic origins and climate-induced environmental changes, the productivity of large floodplain ecosystems in virtually every continent is under serious threat of survival. Fish distributions and temporal dynamics are in part functions of habitat structure and conditions. Riverine fish population biology and dynamics have been studied extensively worldwide and described under various river productivity models that explain community dynamics and structure according to a range of spatial and temporal factors. Fish distribution and movements have been described in four dimensions – longitudinal, lateral, vertical, and temporal (seasonal) – that reflect the dynamic spatial and temporal nature of fish movements and habitat requirements in freshwater systems.

Much of what is known about the fish fauna of large tropical floodplain systems of the world is the result of work undertaken in the Amazonian, Asian and African wetlands. Very little information, however, is available on floodplain fisheries ecology of the Indonesian archipelago, the island of New Guinea or the South Pacific. This has lead to this study being undertaken to understand the floodplain ecology of the Fly System in Papua New Guinea (PNG) by studying a key clupeoid species.

Though cold-water species dominate clupeoid fisheries globally, the greatest diversity of clupeids is found in the warmer marine and freshwaters of the tropics. The greatest diversity is found in the Indo- West Pacific region, although this constitutes a very small percentage of the world clupeoid catch, probably because their commercial potential have not have been fully explored, or because, in most cases, the focus of warm water fisheries is more on food security than cash economy.

The Fly River herring (Nematalosa papuensis), a clupeid of the Indo-Pacific region, is from the genera Nematalosa (subfamily Dorosomatinae) of freshwater systems in Papua New Guinea (PNG) and Australia. This species, a primary consumer, is a key species in the fisheries ecology of the heavily polluted Fly River system, making up to 38% of the fish biomass of the Fly River off-river water bodies (lakes), yet its natural history in the largest wetland system in Australasia and Pacific is largely unknown. This study investigated the habitat of N. papuensis, its population dynamics and structure, its diet, and its growth and reproduction biology in seven off river water bodies (ORWB) representing different habitats along the Fly River system. The habitats studies included 3 forested oxbows, a mixed forest and grassed fringed oxbow, a grassed floodplain oxbow, and 2 grassed floodplain blocked valley lakes. Six sampling trips were undertaken, 3 in 2008 (2 wet and 1 dry season) and 3 in 2009 (2dry and 1 wet season). Three replicate sites within each ORWB were sampled for fish, water quality, and aquatic macrophytes during each trip. No significant seasonal variations (p>0.05) were detected in most of the parameters that were studied; however, there were significant habitat-influenced differences in the biology of this species, most probably in relation to impacts from the Ok Tedi copper mine, upstream.

Water quality analyses revealed significant spatial variations at all sites. Water quality at lakes nearest to the Ok Tedi mine were more degraded with higher dissolved metal concentrations and total suspended solids in the water column. Significant habitat differences were observed in pH, temperature, dissolved oxygen, conductivity, turbidity and total suspended solids. There were no seasonal differences in pH, conductivity, and TSS; but significant seasonal differences in temperature, dissolved oxygen and turbidity were detected. Dissolved metal levels were generally low; with the exception of copper (Cu) where mean Cu concentrations exceeded guidelines (of 1.4 μg/L) at mineimpacted sites, being particularly high at sites closest to the mine. Forty-five aquatic macrophyte species were recorded. Grassed floodplain lakes had greater vegetation density and species richness than forested oxbows, and density and richness of macrophytes increased with distance from the mine.

During the study 17,361 adult and 2,164 juvenile N. papuensis were collected, with a combined total weight of 1008.1 kg. Fish biomass was significantly lower in forested oxbows than in grassed floodplain lakes. Most lakes showed a trend of higher catches during the dry than wet season; however, these seasonal variations were only significant for two grassed floodplain lakes. There were significant spatial but not seasonal variations in juvenile catches. Changes in habitat quality were reflected in the biology of N. papuensis populations, as population biomass increased with distance from the mine, indicating an effect of mine pollution. A lagged phase analysis of the long-term data against the El Niño Southern Oscillation Index (ENSO) showed that short-term climatic events such as El Niños were not significant contributors to the long-term declining trend in biomass.

Length-weight relationships (LWR) varied significantly between lakes and sexes. Separate analysis for each sex revealed a weak but significant difference in male LWR slope between lakes (p<0.05), and a highly significant difference in female LWR slope between lakes (p<0.0001). Fish in forested lakes matured earlier than those found in mixed vegetation and grass-vegetated lakes. Males were consistently smaller than females in all lakes (p<0.0001) but there were no differences in sex ratio. Sexually identifiable fish found in the forested lakes fell within the 110-185 mm size range, whilst most in the mixed and grassed floodplain lakes were within the 140 to >200 mm size range, with females being larger and more abundant in the grassed floodplain habitats and with distance from the mine. In all the lakes, the majority of the individuals and adults were found in the 110-215 mm size classes, comprising 84.1% of the total population sampled. The juveniles from 35-85 mm made up 13.7 %, and those in size classes greater than 215 mm were 2.2% of the population. There were no clear seasonal differences in size classes; however, there were significant differences in size-class distributions between the lakes.

The study of growth of this species was undertaken by tracking cohorts in wild population, while aging of this species in the wild was not possible by using otoliths, where 37 otolith samples showed between 3 to 14 checks. The tracking of cohorts in the lakes showed overlapping cohorts year round clearly indicating continuous spawning. Data from pond fish showed that it is not possible to use otolith or scale checks to estimate the age of N. papuensis as several checks are formed on otoliths every year, not corresponding to fixed time periods. The study showed that N.papuensis is a fastgrowing species with a mean growth rate in the Fly River lakes of 13.69 mm/month, whilst that of fish grown in ponds was 11.86 mm/month. A Von Bertalanffy fitted age-length equation of L (t) = 434.73 [-exp (-2.43604*(t-0))], was derived for pond-raised N. papuensis the Fly River system.

Understanding the diet of species which play a key role in an ecosystem is important because of its links to higher order consumers. This aspect of the study was undertaken by studying gizzard contents as well as stable isotopes. A total of 280 gizzard samples (140 during wet and 140 during dry season) were collected of which 202 had food while 78 were empty, and 280muscle tissue samples were collected for stable isotope analysis.N.papuensis was shown to be predominantly planktivorous and detritivorous through all seasons, however its gut contents and fullness is habitat and season dependant (p<0.0001) with gizzards being fuller in the wet than in the dry season. The identifiable microalgae in the gizzard comprised 32 genera from the Chlorophyta (50 %), Baccillariophyta (25%), Euglenophyta (9.4%), Cyanophyta (6.3%), and Dinophyta, Cryptophyta and Chrysophyta (3.1%each). The largest percentages of phytoplankton were found in fish from the grassed floodplain lakes while detritus content was higher in fish from forested lakes. Results showed clear interaction between lake, season, and food category (p<0.0001). The stable isotope results showed that fish from the grassed floodplain lakes were more δ¹³C enriched than those from the forested lakes (p=0.0001). There was also a significant difference in δ¹⁵N between lakes (p<0.001). There was a clear transition between the carbon signatures of fish from forested (very depleted δ¹³3C values close to those typical of C3 plants) to the strictly grassed floodplain lakes (relatively enriched δ¹³C values close to those typical of C4 plants). Fish from lakes that have a mixed riparian vegetation of forests and open grass - aquatic macrophytes had δ¹³C values in the intermediate range. Generally, N. papuensis had C and N signatures that were closer to phytoplankton, periphyton, zooplankton and particulate organic matter.

Maturity stages were present in N.papuensis populations across a range of sizes, and throughout the year. In general, where both males and females were found in either season, stages 3 and 4, the most fertile stages, were the most common, and occurred more frequently in the grass floodplain lakes than in the forested reaches. N.papuensis was shown to be a highly fecund species with 22,688 ± 5,033 to 126,812 ± 14,916 eggs per fish, and with oocyte diameters ranging from 0.41 ± 0.05 mm to 0.58 ± 0.01 mm. There was considerable variation in male and female GSI in all lakes and between seasons despite N. papuensis being a year-round spawner. Female GSI in these lakes fell within a range of 1.31 ± 0.20 to 4.38 ± 0.30, and male GSI was within a range of 0.59 ± 0.08 to 2.47 ± 0.21. There was strong evidence that higher levels of Cu in the mine-impacted lakes have negatively affected the reproductive ability of N.papuensis. The results suggested that with increase in dCu in the water column there is a corresponding decrease in GSI. This relationship was observed in both male and female fish.

This study is only the second species-specific study of the life history of Fly River fishes, despite there being over 120 different fish and macroinvertebrate species in this river system – the largest wetland system in Australasia and the south Pacific – and despite this species being a keystone species as food for predatory fish such as barramundi. However, of greatest concern is that the future of this system is under serious threat from mining and climate change. There is a serious need to research further the implications of these issues on fisheries ecology, proposed fishery developments, and community livelihood.

Item ID: 46220
Item Type: Thesis (PhD)
Keywords: floodplain ecology, floodplains, Fly River Herring, freshwater lake, freshwater management, gizzard shad, mine impact, Nematalosa papuensis, Ok Tedi, Papua New Guinea, PNG, population analysis, water quality
Date Deposited: 01 Nov 2016 22:33
FoR Codes: 06 BIOLOGICAL SCIENCES > 0602 Ecology > 060205 Marine and Estuarine Ecology (incl Marine Ichthyology) @ 100%
SEO Codes: 96 ENVIRONMENT > 9605 Ecosystem Assessment and Management > 960507 Ecosystem Assessment and Management of Marine Environments @ 100%
Downloads: Total: 743
Last 12 Months: 99
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page