Is the Bellinger Crisis is a National Crisis?

The Bellinger River

Turtles have been part of my life for over 20 years now. They have taken me from the smelliest wetlands at the mouth of the Murray River to the vast lakes of the Mississippi River, but there is one small river on earth that has truly captured my heart and is a place that I eventually want to call home.

The Bellinger River is a small River on the east coast of Australia that is nestled between Macksville to the south and Coffs Harbour to the north. Most people driving between Sydney and Brisbane encounter the River at the river mouth, where the Kalang and Bellinger Rivers merge into one before flowing into the sea. As a seaside community, it is a lovely area, but not necessarily distinct from the 'umpteen' other east coast rivers that start in the great dividing range and make the short journey down to the sea.

However, if you take the exit off the Pacific Highway, and head up to Bellingen, then you will begin to understand the uniqueness and magic of the River. Bellingen is a busy little town. It has it's own uniqueness, but has certainly changed significantly over the last 15 years. The bakery is still there with their wildberry dampers (best fieldwork food ever), but there are plenty more cafes cashing in on the 'indie' atmosphere now- that is progress. Lavender Bridge across the Bellinger River (sometimes) connects the two sides of the town. The Bellinger River has no regulation structures on it at all (rare in Australia these days) and when it rains, the water roars down the mountains and the River floods, often three or four times a year. It is often said that you only need to pee in the River to make it flood. The area has a mix of agricultural activities from cattle and milk production to organic orchards and nuts.

The ocean tidal flows extend up to Bellingen, but upstream of Bellingen, the River is very clear and as interesting to snorkel as the Great barrier Reef (in my biased opinion). When the River is not flooding it is largely a series of deep waterholes surrounded by riffles- ideal for swimming and ideal for a taking your kids canoeing. The riparian zone is a mix of old agriculture (mainly dairy) and dense, steep woodland extending into temperate rain forest. Many of the waterholes are bordered by steep volcanic stone and the bed of the River are rocks and stones of the same material. Between riffles, large beds of ribbon reed are present (Valisnaeria sp). When you snorkel around, you will see a range of fish, but certainly the most charismatic are the eel-tail catfish and freshwater eels. Snorkeling is dynamic, every waterhole is different up and down the River and even day to day.

When a Turtle is not what it seems
Sometimes while snorkelling around, you may see a 'rock' on the bottom of the River move. Turtles may move slowly on the land, but in the water you can easily run out of the breath while chasing one of these moving rocks. They a very quick. Once you get your eye in though, you can usually pick them of the bottom of the River before they realise that you know that they are more than a rock. There are supposedly three species in the River, but in reality there are only two. One of them from 1998-2008 was potentially one of Australia's most endangered. The Bellinger River Emydura had only been caught in ~3 locations along the River and less than a handful had actually been captured. It was actually considered a unsubscribed 'sub-species'  of Emydura macquarii (Vulnerbale Listing), which is a common short-necked turtle found from Qld to the mouth of the Murray River. In 2007, we cracked the puzzle that is the Bellinger River Emydura. Over a three week period, we captured 76 of them. How did we go from 4-5 samples in 20 years to 76 in three weeks? It all came down to how to sample them. While we normally will capture turtles by snorkelling for them or throwing a few traps in, we caught most of these turtles using a dipnet from a boat at night. Now this is where the story gets interesting.

DNA analysis revealed that the endangered Bellinger River Emydura was in fact made up of individuals that had probably been put there by people thinking they were doing the right thing picking up turtles on the road as they came home from Coffs Harbour or from the south. In most surrounding catchments, short neck turtles are common, but the geology of Bellinger river catchment may have stopped them from ever getting in there. That is one theory at least. Introduction of a feral, but native Australia turtle- no big deal? The intentions were good. I guess the moral of this story, if any, is that good intentions can help a species go extinct and the rest of this story will demonstrate why.

This figure essentially shows how the DNA of the every Bellinger River Emydura fits against other turtles from neighbouring catchments. Nearly all of the individuals originally came from a neighbouring catchment. The 2nd figures highlight that the Emydura populations up and downstream and in the Kalang have all different haplotypes meaning that up stream populations were probably picked up from the Macleay and Clarence catchments, whereas the Kalang population originated entirely from the Macleay. The lower Bellinger population originated almost entirely from the Clarence catchment. (from Georges et al. 2011).

People saving turtles should be no big deal, but the story becomes more intriguing after I introduce the next turtle that inhabits the River.

The George's Snapping Turtle
Myuchelys (nee Elseya- but NOT Wollumbinia) georgesi, is a pretty plain old freshwater turtle. It is not terribly unique to look at. Like the Emydura, it is a short neck turtle. They are pretty grumpy when you catch them too. Takes a bit to actually get bitten by them, but they will certainly open their mouths and be all unhappy about being picked up. Where they become important and critical to this story lies with their distribution and role in this unique ecosystem.

The only place on earth where this turtle is found, is from the tidal zone at Bellingen up-stream to the mountains. It is also found in the Kalang River, which actually joins with the Bellinger River down near the Pacific Hwy exit. It is the true Bellinger River turtle and should represent a symbolic 'totem' species for the community. Given its limited distribution, regardless of population status, it deserves to be given some conservation status (eg. vulnerable). But the populations were at risk, even before the mass death event that occurred in February 2015 (I will talk about this later). Both the Emydura and George's Snapping turtle appear to hybridise with each other. That means they are breeding together.

How to distinguish between the Bellinger River Emydura and Georges Snapping turtle. The bottom photo looks like an Emydura, but look at the head plate, which is suspiciously George's turtle-like.

Now remember, the Emydura are introduced and not native to the Bellinger River. In a species with such limited distribution, breeding between the two species actually has the potential to wipe out the Georges Snapping turtle. It is the same concept of a mongrel dog breeding with the best in show- the offspring are potentially worthless. Dingoes have gone through the same process. Stray and wild dogs breed with dingoes and the pure-bred dingo is almost extinct in Australia. So Hybridisation with Bellinger River Emydura is a major threatening process. There is a more immediate threat to their survival, though and to truly get to the bottom or implications of losing this species, we need to consider their feeding ecology and habitat preferences.

The figure above might be a bit hard to decipher, but it is essentially showing the dietary composition of most short-necked turtles in eastern Australia (Spencer et al. 2014). Several studies have been done on the George's Snapping turtle and they largely fall out as a dietary specialist (relative to other short-neck turtles). Their diet consists predominantly of insect larvae and things that fall on the surface, like berries, figs and insects. As an indicator of biological health of a River, they don't come much better than this- and here is why.

Environmental Indicator

Firstly, the food that they eat are highly sensitive to pollution, increased sedimentation or general water conditions. In a River, that changes almost daily because of rainfall, the insect population is also going to respond rapidly and be affected by natural changes. The George's snapping turtles are probably adapted to boom-bust cycles of the River and resilient to natural shortages of food for periods. However, if there are chronic issues with the food supply, then the turtles will begin to be impacted. In 2001, I surveyed the the turtles and found in Spring that none of the females were gravid. The flood in March was large and none of the aquatic vegetation had returned. Their food supply had definitely been impacted before winter and they forgo breeding that year. A reproductive cycle like this is common with boom-bust organisms and being long-lived, skipping a breeding season is no big deal. The problem comes when there is long-term disruption of the food cycle and being a clear-water water specialist, changes to water quality are magnified.

What might have long-term effects on water quality and disrupt the food cycle? In clear-water systems, increased sedimentation (both deposited and suspended) is a major issue. Reduced clarity, reduces or delays plant growth, as does any sediment that settles. Floods increase sedimentation and floods also strip vegetation, which is required for many insects to breed. But flooding has always occurred in the area and the River returns to clear water conditions relatively quickly. Agriculture and construction are other sources of potential water quality problems. Agriculture can change riparian zone vegetation and structure, reducing its structural integrity.The importance of riparian zones to provide buffer zones to agriculture is significant. Increased nutrients into the River directly leads to algal growth and a change in the invertebrate community. In general, riparian zones are very good and adapting to changing nutrient levels, effectively providing a buffer between agriculture and the River. Other sources of sedimentation include construction related to bridges of roads. The landscape means that there is little riparian zone between the road and the River. Many roads reside on a steep embankment almost directly over the River and maintaining or upgrading these roads would also release sediment or nutrients into the River. During rain events and flooding, the run-off for dirt roads is also significant. All of these factors will reduce the ability of the River to return optimal conditions after a flood and ultimately have long-term effects on plant and invertebrate communities.

Besides changes to ecosystem processes, increased sedimentation may also reduce habitat availability for the George's Snapping turtle. Our study from 2007 showed that this turtle preferred large waterholes significant amounts of large rocks as substrate and bedrock, as opposed to areas with small gravel or sand. They also preferred certain aquatic plants too. 

Multivariate analysis to show properties of waterholes with and without George's Snapping turtle. the turtles preferred larger waterholes with a rocky substrate, as well as large bedrock walls. They also had a preference for ares with the aquatic plant, Hydrilla. 

Coming back to these turtles representing a good indicator of River health. Their value is that their health relates to medium-long term changes in the River, rather than annual or seasonal fluctuations that occur in an environmentally volatile region. They are also long-lived and turtles have been shown to bio-accumulate toxins and regular sampling (shell or nails) of marked individuals can be used to longitudinally monitor increases in toxins in the River- something that snapshot monitoring of water quality may miss. 

February 2015- A Mass Die-Off

People in a canoe/kayak found turtles hauling themselves out of the water in a bad condition. The area was downstream of the Thora bridge and now as many as 100 Georges Snapping Turtles have been found dead or dying in recent days from a mystery illness. They have lesions and growths around the eye. Turtles are prone to eye diseases, it is a common area for pathogens and disease to attack. Fungus, bacteria or viruses can all cause SCUDs (Septicemic Cutaneous Ulcerative Disease), but bacteria is the most common cause. We don't know at this stage is whether it is bacteria, let alone what type of bacteria. The question is why are they now susceptible? Has the recent dry followed by floods in January increased bacterial loads? Possibly, but why were all of these turtles starving and emaciated and other species were not affected? The answer I think lies with changes to water quality and a disruption of their food supply, over a long period, followed by an increase in bacteria loads following the floods. It is highly unlikely that these turtles starved because of the eye ulceration. The bacteria was able to take advantage of food starved turtles that were immuno-compromised. We need the pathology results to know and these will come. What we also know is that this event has probably not happened before in this River or it is extremely uncommon. A resident that has lived in the heart of their territory for 47 years has never seen an event like this before. This resident is reliable and is on the River daily.

Road construction (over several years) south of Thora bridge is an obvious area of interest, but what we must do first is understand the extent of the outbreak and whether turtle numbers declined over the last 15 years that this species has had some research conducted on it (albeit largely indirectly because research was focused on the Bellinger River Emydura). We also need to know the organism that we are dealing with. The worst thing that we can do is go in gung-ho spreading some highly contagious disease throughout the River.

The dead turtles are symptomatic that something is wrong with the River. This is not a seasonal thing. Put it simply, Turtles simply don't die unless something is wrong. Their recovery will be difficult. Foxes destroy 90% of nests in many parts of the Australia (Spencer 2002) and it is likely to be similar here. Natural recruitment is therefore compromised. Do we need to start a breeding program and protect healthy individuals? This depends on the disease and how extensive the problem is. This turtle represents so much more than just another reptile. In fact the role of turtles in ALL river systems is severely underestimated. In most systems, they have the highest vertebrate biomass and they are the major nutrient recyclers (scavengers); a significant herbivore; and the top predator.

Right now, we must wait and see what the pathology tells us, but if we sit on our hands, we will drive another vertebrate to extinction and ignore a looming ecosystem breakdown of the Bellinger River. There are other rivers on the north coast of NSW, where some species of turtle only exist in. Is the Bellinger River a preview of what is to come?

I hope that this is not the last time that my young family see these turtles in my favourite part of the world.

• Blamires, S., Spencer, R., King, P. and Thompson, M. (2005), 'Population parameters and life-table analysis of two coexisting freshwater turtles: Are the Bellinger River turtle populations threatened?', Wildlife Research, vol 32, no 4 , pp 339 - 347.
• Georges, A., Spencer, R., Welsh, M., Shaffer, H., Walsh, R. and Zhang, X. (2011), 'Application of the precautionary principle to taxa of uncertain status : the case of the Bellinger River turtle', Endangered Species Research, vol 14, no 2 , pp 127 - 134.
• Blamires, S. and Spencer, R. (2013), 'Influence of habitat and predation on population dynamics of the freshwater turtle Myuchelys georgesi', Herpetologica, vol 69, no 1 , pp 46 - 57.
• Spencer, R-J, Lim, D., Georges, A., Welsh, M. and Reid, A.M. (2014). The risk of inter-specific competition in Australian short-necked turtles. Ecological Research 29:767-777.

• Spencer, R. (2002), 'Experimentally testing nest site selection: Fitness trade-offs and predation risk in turtles', Ecology, vol 83, no 8 , pp 2136 - 2144.