Population dynamics of the migratory fish Prochilodus lineatus in a neotropical river: the relationships with river discharge, flood pulse, El Niño and fluvial megafan behaviour

The relative importance of flood pulse dynamics and megafan behaviour for the Sabalo (Prochilodus lineatus) catches in the neotropical Pilcomayo River is studied. The Sabalo catches can mainly be explained by decreased river discharges in the preceding years resulting in smaller inundated areas during rainy season floods and thereby in a decreased area of feeding grounds for the fishes. The decreased river discharges and the related decline of Sabalo catches in the 1990's can be linked to the 90-95 El Nino event. In 2007 the Sabalo catches were comparable to the catches before the "El Nino" event. The connectivity (continuity) between the main river and flood plain areas, which is influenced by sedimentation processes, is also of great importance and very probably plays a more important role since the late 1990's.

The South-American Pilcomayo River is strongly influenced by the natural hydrological cycle.The upper and middle parts of the river have as yet not been regulated by dams or hydrological-technical works.Therefore the river provides an excellent case to examine the impact of flood pulse dynamics on the population dynamics of migratory fishes under natural circumstances.
The Pilcomayo River is populated by various migratory fish species, which have an important nutritional and commercial value for the local indigenous human population.The Sábalo, Prochilodus lineatus, (Characiformes, Prochilodontidae) is the most abundant migratory fish species in the Pilcomayo River (Bayley, 1973;Payne & Harvey, 1989;Smolders et al., 2000Smolders et al., , 2002)).Prochilodus lineatus is a detritivorous fish (Bowen, 1983;Bowen et al., 1984;Bayo & Cordiviola de Yuan, 1996;Fugi et al., 1996;Lopes et al., 2007) that feeds in the Pilcomayo floodplains of the lower Chaco (Bayley, 1973;Payne & Harvey, 1989), where as a result of the decomposition of terrestrial and aquatic vegetation during flooding and flood recession the levels of organic detritus tend to be high (Cordiviola de Yuan, 1992;Bayley, 1995).This might explain the dominance of P. lineatus in such flood plains in the Paraná region, where the species may account for up to 90% of the fish biomass (Bonetto et al., 1969;Bayley, 1973;Cordiviola de Yuan, 1992).Prochilodus species fulfil important functional roles in neotropical riverine ecosystems by ingesting and processing large amounts of sediment.Detritus is a crucial pathway in energy and nutrient fluxes in ecosystems and loss of detritivores could therefore strongly influence ecosystem functioning (Bowen, 1983;Flecker, 1996;Taylor et al., 2006).
The considerable decrease of Sábalo catches in the Pilcomayo River during the 1990's was an important concern.In 1993, P. lineatus was even listed as a vulnerable species by the Bolivian government.Various explanations have been put forward to explain the decline of P. lineatus in the river during the 1990's.Of these explanations contamination of the river with heavy metals from mining activities in the upper reaches of the river (near Potosí) has received most attention (Edwards, 1996;Garcia-Guinea & Huascar, 1997;Medina Hoyos, 1998;Hudson-Edwards et al., 2001;Miller et al., 2002;Smolders et al., 2003).The mining region near the city of Potosí forms part of the Bolivian Pilcomayo catchment (Fig. 1).Mining activities started in the 16 th century at 'Cerro Rico de Potosí'.In 1985 the Crushing-Grinding-Flotation method was introduced to recover Lead (Pb) and Zinc (Zn).At present, there are as many as 40 processing plants ('Ingenios'), all working with the flotation process, operating in or immediately adjacent to the city of Potosí.These plants emit slurries containing highly alkaline (pH 10-12) water and high levels of unrecovered metals in the local streams.Ultimately, these streams drain into the Tarapaya River which is a tributary to the Pilcomayo River.Research revealed, however, that it is highly unlikely, that the measured levels of metal contamination of the lower parts of the river (Hudson-Edwards et al., 2001;Smolders et al., 2003) could have been responsible for the decline of the Sábalo in the 1990's (Smolders et al., 2002).The intensity of fishing has strongly increased but can neither explain the decline of the Sábalo in the nineties (Payne & Harvey, 1989;Smolders et al., 2002).In the Pilcomayo River, however, flood pulse dynamics may be expected to have an overriding effect on fish production and dynamics (Krykhtin, 1975;Welcomme, 1979;Welcomme et al., 2006).
The main aim of this paper is to reveal whether the general FPC applies for the abundance and timing of the reproduction of P. lineatus in the Pilcomayo River, where sedimentation processes have interrupted the former direct connection between the lower reaches of the Pilcomayo River and the Paraguay River due to the presence of an active fluvial megafan system.This system is formed as the Pilcomayo River exits the Andes range and enters the adjacent Chaco basin, allowing the river to migrate laterally and to deposit huge fanshaped bodies of sediment.
The main research questions of the present study are: a) Is river discharge a major force for the abundance of Prochilodus lineatus in the Pilcomayo River?b) Does annual variability of river discharge have an effect on the timing of the reproduction of Prochilodus lineatus?

Study area
The Pilcomayo River, a tributary to the large 'La Plata' system, has a catchment basin area of approximately 272,000 km 2 .It arises in the Bolivian Andes (Cordillera Oriental, ca.5000 m a.s.l.) and cuts down through the Andes, through an inaccessible terrain with rapids and narrow canyons, until it reaches the Chaco plain near the town of Villa Montes (ca.400 m a.s.l.).The Andean part of the river has a length of ca.500 km and a mean slope of 10 per mill.The river then flows southward through the Chaco, where it soon forms the natural frontier between Argentina and Paraguay before it eventually joins the Paraguay River just north of the town of Asunción (Fig. 1).This lower stretch of the river has a total length of almost 1000 km and a very low mean slope of only 0.2 per mill.The river is strongly influenced by the hydrological cycle, which is driven by the differences in rainfall between the dry (May to October) and the wet (November to April) seasons (Bayley, 1973;Payne & Harvey, 1989;Smolders et al., 2002Smolders et al., , 2004)).The lower reaches are alternately affected by inundation and drought.Near the town of Villa Montes migrating fishes are exploited commercially by means of seine nets, fish traps (Fig. 2) and throwing nets.

Data analysis
In general fishing activities in river systems are diffuse and it is hard to register catches more or less completely.In the case of the Pilcomayo River, however, the fish have to pass through a narrow canyon with strong currents.It is along the shores of this canyon that fish traps (Fig. 2) are located, which function day and night and are very efficient.Since the 1980's an effort was made by the authorities to make accurate recordings of the amounts of fish extracted from the river, for tax reasons.This was done by registrations of the amounts of fish transported from all the commercial fishing sites in the surroundings of Villa Montes.Therefore the official estimations of Sábalo catches in the Pilcomayo River are remarkably good compared to many other rivers.
Official estimations of total annual Sábalo (P.lineatus) catch in Villa Montes and data on river discharges were obtained from Proyecto Pilcomayo (Tarija, Bolivia).These data were analyzed by multiple linear regression with backward deletion of variables (SPSS version 15).The discharge of the seven preceding years (Y1, Y2, Y3, Y4, Y5, Y6 and Y7) plus the current year (Y0) were included in the analyses as independent variables.Sábalo catch was the dependent variable.The criterion for the probability of F-to-remove was > .100).

Gonadal development
Data on gonad development where derived from Guerrero Hiza (1998).

Water chemistry
From May 1998 until February 1999 samples of the Pilcomayo River were collected regularly at Villa Montes (Bolivia).Immediately after collection temperature was measured and pH and bicarbonate (HCO 3 -) were determined, after which samples were filtered (0.45 μm).Next 1 ml of nitric acid per 100 ml water was added, after which the samples were stored in pre-washed polyethylene containers at 20 o C until analysis.For anion analyses samples were filtered and 60 mg citric acid per 100 ml water was added after which the samples were stored in pre-washed polyethylene containers at -20 o C until analysis (see Smolders et al., 2004).Total dissolved ion concentration was calculated from the analyses by adding up the concentrations of all cation and anions.

Possible effects of the hydrochemical cycle on biota Intra-annual variation and reproduction
At the onset of the dry season, Sábalo (P.lineatus) appear in Villa Montes, because the fish reach the Andean region in their upstream migration.The peak of the Sábalo migration is observed in May.Only adult fish of at least 2 years old migrate.The young Sábalo (young-of-the-year (YOY)) remain in the flood plain area in the lower Chaco in Paraguay and Argentina to feed and had to survive the dry season in pools and oxbow lakes that retain water throughout the dry season (Bonetto et al., 1969;Cordiviola de Yuan, 1992).As there are no suitable food resources in the upper parts of the Pilcomayo River, the fish do not feed during their upstream migration and fully depend on their fat reserves (Bayley, 1973;Payne & Harvey, 1989).When river discharges increase at the onset of the rainy season (October to January) the migrating Sábalo spawn  their eggs in the open water of the main channel of the river in the foothills of the Andes (Bayley, 1973).The peak of reproduction falls in November-January (Mochek & Pavlov, 1998).Similar migratory behaviour with spawning during periods of rising river water levels has been observed for Prochilodus lineatus in the lower Uruguay River (Espinach Ros et al., 1998) and for Prochilodus scrofa (= P. lineatus) in the Mogi-Guassu River (Godoy, 1959).An advantage of this particular spawning behaviour is that eggs and developing alevins are rapidly transported downstream towards the floodplain, where they are distributed by the rising water over the freshly inundated feeding grounds (Bayley, 1973;Cordiviola de Yuan, 1992;Araujo-Lima & Oliveira, 1998;Welcomme et al., 2006).
In 1998, development of the Sábalo gonads was monitored in the Pilcomayo River by Guerrero Hiza (1998).Fig. 3 shows that the development of the gonads followed the gradual increase in aqueous total dissolved solids.The period in which spawning was observed was characterized by low dissolved solids contents following the onset of the rainy season (Fig. 3).Many other species from rivers that show a strong seasonal variation in discharge also show synchronized reproduction at the early wet season (e.g.Winemiller, 1989;Paugy, 2002;Welcomme et al., 2006).
In a highly dynamic system such as the Pilcomayo River, river discharge and aqueous major ion compositions and total dissolved solid concentrations typically show large variations during the year (Fig. 3).Current velocity is assumed to be an important trigger for spawning and migration for stream fish species (Lucas & Barras, 2001;Welcomme et al., 2006).Macroion concentrations, however, might also play an important role by serving as a trigger for processes related to the reproduction of fish.For example, a strong drop of the conductivity at the onset of the rainy season has been shown to be responsible for the induction of spawning in the neotropical fishes Eigenmannia virescens (e.g.Kirschbaum, 1979) and Hoplosternum littorale (e.g.Ramnarine, 1995).Furthermore in Hoplosternum littorale increasing total dissolved solids serve as a stimulus for gonad development during the preceding dry season (Hostache et al., 1993).We propose that changes in water chemistry may trigger gonad development and spawning in P. lineatus in the Pilcomayo River.Interestingly, gonad development in 1998 was two months ahead of observed maturation indices in preceding years (Guerrero Hiza, 1998); this could be explained by the fact that during the very dry El Niño year 1998 (Smolders et al., 2000) high dissolved solid concentrations were reached much earlier than in 'normal' years.Water temperature did not appear to correlate with gonad development (Fig. 3b).We suggest that research regarding the effect of dissolved solid concentrations on gonad development should be considered for rivers with large fluctuations in river discharge.

Inter-annual variation
The Pilcomayo River is not only characterized by strong intra-annual variations but also by strong inter-annual variations in river discharge (Fig. 4a).Models predict that production in a particular year depends on the amount of water that remains in the flood plain after a flood (Welcomme, 1979), since the area of suitable fish nursery grounds is directly related to the flooded surface area (Benke, 2000).Hence, fish catches in floodplain rivers are frequently related to the extent of the floods in the preceding years (Krykhtin, 1975;Welcomme, 1979;Laë, 1992;Gomes & Agostinho, 1997;Smolders et al., 2002;Welcomme et al., 2006).Dry years not only decrease the availability of feeding grounds but may also result in a lack of connectivity (continuity) between the elements of the flood plain (e.g.Junk et al., 1989;Welcomme et al., 2006).For instance, in dry years young fish are unable to migrate from the permanent lagoons and pools, where they survive the dry season, towards the highly nutritious waterland transition zones.Furthermore, the migration of adult fish towards the main channel is hampered, while desiccation and predation of juveniles by birds can also result in high mortality rates (Agostinho & Zalewski, 1995).
By means of multiple linear regression, we obtained a remarkably strong relationship between Sábalo catch and river discharges of preceding years for the period 1980-1996 (adjusted R 2 = 0.94) (Fig. 5, Table 1).Remarkably, the five preceding years had a significant effect on Sábalo catches although the Sábalo caught were two to three years old.Apparently there is a kind of memory effect playing a role as low or high productivity caused by low or high river discharges in the past has a significant effect on future generations.Apparently a strong year class caused by a high production and survival of eggs and alevins, can have a positive effect on the population size in many years that follow while the opposite will be true for weak year classes.
A decline of mean river discharges since the middle of the 1980's was followed by a gradual decrease of Sábalo catches and the first half of the 1990's was characterized by very low river discharges and low corresponding Sábalo catches (Fig. 5a).The 'return of the Sábalo' after its collapse in the 1990's, coincides with the recovery of the mean discharge in the preceding years (Fig. 5a).However, since 1996 the relationship between river discharge and Sábalo catches has changed considerably.Now, only the three preceding years are significantly contributing, but still 70% of the variation can be explained by past river discharges (Fig. 5b).Apparently the memory effect has gone and the system has become less stable.An explanation might be provided by water  management changes.Since the end of the 1990's, more water was directed, towards Argentinean territory causing significant changes in hydrological processes in the lower catchment area.This caused a rapid siltation of parts of the river bed and breakthroughs of river banks, resulting in the inundation of new areas within the floodplain, while formerly flooded areas became devoid of water (Smolders, 2006).Smolders et al. (2000Smolders et al. ( , 2002) ) demonstrated that since the mid 1970's the mean annual discharge of the Pilcomayo River is closely related to the large-scale climatic "El Niño" Southern Oscillation (ENSO) system.ENSO is known to influence river discharges throughout the world by its influence on rainfall patterns (Depetris et al., 1996;Stone et al., 1996;Poveda & Mesa, 1997;Sun & Furbish, 1997;Puckridge et al., 2000;Camilloni & Barros, 2003).The two worst "El Nino" years of the last thirty years, 1982/1983 and 1997/1998, were characterized by very low mean discharges in the Pilcomayo River (Fig. 4).In the 1990's, five consecutive years with relatively low or very low mean river discharges coincided with a strong decline in the fish catches in Villa Montes.The low river discharges in the first half of the nineties also coincided with a five-year period in which the SOI (Southern Oscillation Index) was consistently low.Some consider this period as the longest "El Niño" event on record (Trenberth & Hoar, 1996).In 2007 the Sábalo catches were comparable to the catches before the "El Niño" event (Fig. 5b).
It is remarkable, however, that river discharge values were comparably low in the 1960's although in these years no alarming Table 1.Results of the stepwise multiple linear regression with backward selection (SPSS v. 15.0).The seven preceding years (Y1-Y7) plus the current year (Y0) were included in the analyses as dependent variables.For each model (step) the unstandardized coefficients with degree of significance and the basic statistics are presented.Sábalo catch was the dependent variable.The criterion for the probability of F-to-remove was > .100).Dependent variable: CATCH.reports of low numbers of Sábalo have been reported for the river (Fig. 4).We will come back to this in the following section.

Stream self-blockage/megafan river behaviour
A major problem in the Pilcomayo River are the high erosion rates in the Andean region.The mean sediment load of the Pilcomayo River (10.6 g L -1 ) is amongst the highest in the world (Guyot et al., 1990;Iriondo, 1993;Smolders et al., 2002;Depetris et al., 2003;Wilkinson et al., 2006).The main cause of the strong erosion is the susceptibility of the local geological formations to erosion because of their small particle size (Iriondo, 1993).Guyot et al. (1990) calculated that the mean erosion rate for the catchment area upstream of Villa Montes amounted to 890 t km -2 y -1 for the period between 1979 and 1982.It should be realized that the mean annual amount of sediment transported by the river is sufficient to cover as much as 5000 ha of land with one meter of sediment.The sedimentation of this material provokes silting up as well as diversion of the river bed.
Thus, huge amounts of sediment are transported to the lower reaches of the river in the Chaco plain.In this region the suspended sediment is deposited creating an alluvial fan with a total surface area of 210,000 km 2 (Iriondo, 1993).It is assumed that the current situation of erosion and sedimentation has remained unchanged since 1000 BP (Iriondo, 1993).The fan has formed because the river has changed its course several times.During the last centuries, the river discharged into a tectonic depression called Estero Patiño (Cordini, 1947), located 250 km upstream from the Paraguay River.This depression gradually filled up with sediment until it had become more or less completely silted up in the mid 1940's (Pool et al., 1993).Then the river started to fill up its own river channel, resulting in a gradual retreat of the river.The existence in the region of many fully silted up paleochannels of the Pilcomayo demonstrates that historically the river has frequently changed its course due to this kind of sedimentation processes (Iriondo, 1993;Pool et al., 1993).As a consequence, the Sábalo population of the Pilcomayo has become more or less completely isolated from the populations in the La Plata basin.
The overflowing water started to form new swampy areas (Wilkinson et al., 2006) (Fig. 6).The point where the river "disappears" (choke point; Fig. 6) has been moving upstream rapidly ever since, a process called endpoint recession.The amount of sediment transport strongly depends on river discharge (Smolders et al., 2002).Since the mid-1970's high river discharges resulted in high annual sediment transports and to a rapid increase of the endpoint recession rate of the river bed (Smolders et al., 2002).The geomorphologic changes which resulted from the high sediment deposition rates in de Pilcomayo floodplain, also seem to have changed the relationship between river discharge and fish production, in the sense that nowadays higher discharges are required to produce the same amount of fish compared to the 1960's (see Fig. 5).Since the early 1990's a further silting up of the river bed was more or less successfully prevented by the construction of channels that try to divide equal parts of the river water between Paraguay and Argentina.
The connectivity between the main river and flood plain areas, can be very dynamic, a phenomenon which seems to play an even more important role since the late 1990's.Due to a less consistent maintenance by the Paraguayan channel, relatively more water (and thus sediment) was directed, towards Argentinean territory.This resulted locally in a rapid siltation of the river bed and breakthroughs of river banks, resulting in the inundation of new areas within the floodplain (Fig. 6).Such events may in some years strongly increase the connectivity, with areas rich in young Sábalo, resulting in relatively high numbers of migrating Sábalo in the dry season.For instance, reconnection of formerly isolated lakes, liberating fish that had been isolated in these pools, may have played an important role since 1998 and explain the underestimation of the Sábalo catches based on the linear regression established from catches before that period (Fig. 5b).

Conclusions and perspectives
It is generally assumed that the contamination of the Pilcomayo River is responsible for the decline of the Sábalo population during the 1990's (e.g.Medina Hoyos, 1998).However, our results show that we can quite well explain the historic and current Sábalo catches in the river by the hydrodynamic processes in the river basin.
The decline of the Sábalo in the 1990's seemed to be caused by the decreased river discharges in these years resulting in smaller inundated areas during rainy season floods and thereby in a decreased surface of the feeding grounds for the Sábalo.The decreased river discharges can be attributed to the 90-95 El Niño event, one of the severest on record.Over-exploitation of the fish stock in these years has possibly co-contributed to this decline of the Sábalo.This river discharge hypothesis was confirmed by observations in recent years showing a recovery of the Sábalo catches in Villa Montes as a result of the reestablished river discharges since the late 1990's.The connectivity (continuity) between the main river channel and flood plain areas, are also of great importance and very probably play an even more important role since the late 1990's.
At the moment a master plan is being developed for the entire Pilcomayo basin (Coccato et al., 1999;Smolders, 2006).The enormous amounts of sediment that are transported every year to the lower reaches of the river provoke enormous problems.The resulting unpredictable desiccation and inundation can harass the human population that are planned to inhabit these areas in the near future.However, proposed measures meant to tackle the sedimentation processes in the lower reaches of the Pilcomayo River may have adverse effects on Sábalo populations.One of the measures being considered is the construction of a transversal dam a few hundred kilometers downstream of Villa Montes.This dam should retain an important part of the sediment load and also divide the water equally between Paraguay and Argentina.It is clear that such a measure would strongly affect flood-plain dynamics in the lower reaches and certainly the connectivity between the middle and the lower reaches of the river, and will thus strongly affect Sábalo catches in Bolivia.An alternative for such a dam is the construction of a series of reservoirs upstream of Villa Montes.These reservoirs should contain the sediment load of the river and would enable the regulation of the river discharge throughout the year.The construction of reservoirs in the middle range of the river will also have a negative effect on fish migration and river dynamics but moreover would strongly affect the intra-annual variation of stream velocity and total dissolved solid concentrations of the river water.This might affect the timing of the reproduction and also the distribution of alevins and eggs within the Pilcomayo floodplains (Fig. 3).
For floodplain rivers irregular dynamic phenomena are essential for the ecosystem functioning.Hence, all measures to regulate those rivers must be considered as an unnatural disturbance from an ecological point of view (Bayley, 1991(Bayley, , 1995;;Welcomme et al., 2006).The problems that occur in naturally functioning river floodplains are rarely real ecological problems but frequently based on social and economic demands that the river is required to satisfy.The establishment of large nature reserves in the area might prevent the further development of agricultural activities in the lower reaches of the Pilcomayo River and decrease the necessity to combat the highly irregular flooding events and deposition of sediments.Ecotourism may be an important alternative to the development of agricultural activities in this still largely pristine region.
Simple multiple linear regressions from limited time series can accurately predict mono-species catches from hydrological values (Fig. 5).Such equations can be used to predict periods of fish scarcity or abundance without the need for complex modeling.However, major environmental alterations, for instance produced by hydrological changes, will alter the relationship.Furthermore, the world-wide increasing abstraction of water for agriculture, growing numbers of dams and dikes and at the same time anthropogenic climate change by global warming and El Niño events may pose serious threats for freshwater fisheries in floodplain rivers by diminishing the amount of water reaching the floodplain areas.
On six different dates in 1998 (May 22, June 19, August 7, August 25, September 11 and September 30) 25 fishes were obtained from fish traps in the Pilcomayo canyon near Villa Montes, and the stage of gonadal maturation was analyzed for each fish.To each fish one of the following stages (I to V) was attributed.Stage I (inactive stage): Ovaries are small.No oocyte is visible to the naked eye.Stage II (recovering stage): Few oocytes are visible to the naked eye.There are few yolks.Stage III (early maturing stage): Most of the oocytes are visible to the naked eye.There are some yolks in these oocytes.Stage IV (late maturing stage): All the oocytes are visible to the naked eye.There are many yolks in the oocytes.Stage V (mature stage): Many mature oocytes are released from their follicles into the ovarian cavity.The oocytes could flow out through the genital opening when the abdomen is gently pressed.The mean value for the gonadal development stage of the 25 fishes was calculated.

Fig. 3 .
Fig. 3. Total dissolved solids concentration (TDS) in Pilcomayo River water (a) or water temperature (b) and gonadal maturation indices of Sábalo (Prochilodus lineatus) fish versus time (May 1998 until February 1999).The gonadal maturation indices are scaled from 1 to 6 in which 6 represents spawning.

Fig. 4 .
Fig. 4. (a) Mean annual discharge for the Pilcomayo River since 1960.The values were calculated for the hydrological year, which runs from October of the previous year until September of the current year.Data were obtained from Proyecto Pilcomayo (Tarija, Bolivia).(b) Mean monthly values of the Southern Oscillation Index (dots) and mean annual discharges of the Pilcomayo River (open circles), since 1976.Mean annual discharge values were calculated from data obtained from Proyecto Pilcomayo (Tarija, Bolivia).The values were calculated for the hydrological year, which runs from October of the previous year until September of the current year.

Fig. 5 .
Fig. 5. (a) Mean annual discharge and Sábalo catches over the years in the Pilcomayo River near Villa Montes.(b) Calculated and observed Sábalo catches based on the data presented in Fig. 4a.Correlations were obtained by stepwise multiple linear regression with backward selection (SPSS v. 15.0).The river discharge of the seven preceding years (Y1-Y7) plus the current year (Y0) were used in the analyses.The solid line is based upon the years 1980-2006.The dashed line is based upon the years 1997-2007.(c) Observed Sábalo catches plotted against the calculated Sábalo catches for the years 1980-1996 and 1997-2006.Data of Sábalo catches and mean river discharges were obtained from Proyecto Pilcomayo (Tarija, Bolivia).

Fig. 6 .
Fig. 6.Above: Retreat of the Pilcomayo River and dynamic creation of new flood plains due to self-blockage (silting up) of the river channel.This caused a retreat of hundreds of kilometers of the choke point in a few decades (indicated by the black arrow) and an upstream migration of the flood plains.Bullets indicate migrating Sábalo population in the Pilcomayo River (white) and Sábalo population in the La Plata basin (black).Below: Breakthrough of Pilcomayo River bank inundating new areas in the Chaco floodplain area.