Bogong Moths Agrotis infusa, soil fertility and food chains in the Australian alpine region, and observations concerning the recent population crash of this iconic species.
The annual migration of Bogong Moths Agrotis infusa to the alpine region is an important ecological and cultural (Indigenous and European) natural phenomenon in south-east Australia. Such is the magnitude of this concentrated food resource that Indigenous groups seasonally migrated to the alpine region to exploit it (Flood 1980). Its periodic invasions of Parliament House in Canberra, and its presence at the Sydney Olympic Games (2000), have added to the iconic status of the species. In the Alps, caves and interstitial spaces in peri-glacial boulder fields and rock outcrops are favoured aestivation sites and a variety of mammalian and avian predators exploit the concentrations of moths and disperse nutrients across the alpine environment. The moth is also a major component of the diet of the endangered Mountain Pygmy-possum Burramys parvus, particularly breeding and lactating females, when it may be the exclusive prey (Mansergh et al. 1990; Smith and Broome 1992).
Common (1954) published the first comprehensive biological study on the species, augmented in Common (1989), and Green (2011) estimated the amount of nutrients and their energy content that A. infusa brought to the Snowy Mountains. Warrant et al. (2016) provide a current review. In the alluvial plains of south-east Australia a female may lay c. 2000 eggs and is capable of multiple broods within a season (Common 1954; Warrant et al. 2016). Agrotis infusa spend the autumn-winter in their egg-pupae stage (as cutworms) and are recognised as an agricultural pest, feeding on crops of annual dicotyledons and broad-leafed crops (Common 1954). As a sub-adult, the moth is c. 25 mm long, and the annual migration (up to 1000 km) to the Alps in late September-November is accomplished by exploiting the prevailing NW winds and their recently discovered internal magnetic-assisted navigation, the first insect known to exhibit this attribute (Dreyer et al. 2018). On arrival in the Alps, a moth weighs about 0.33 g (dry weight 0.155 g) and its abdomen contains 57% fat (dry weight) in females and 66% in males (Common 1954; Green 2011). Green (2011) found the gross energy content per moth to be 4.86 KJ and estimated they bring a massive 4929 GJ of energy to the Snowy Mountains.
Despite its iconic status as an agricultural 'pest' and its role as a key dietary item for the endangered B. parvus, there has been no systematic monitoring of A. infusa abundance: perhaps the moth migration was perceived as 'too big to fail'. However, the need for such information has been identified in national and state recovery documents for B. parvus (e.g. Department of Environment, Land, Water and Planning 2016). Consequently, information about annual variation in moth numbers (in natal and alpine areas) is derived mostly from anecdotal observations and limited surveys. Good and poor seasons have been observed but never the collapse noted in 2017.
From our first observations in 1978 and probably for millennia, sounds from B. parvus habitat in Victorian boulder fields could be heard from over a kilometre away--large flocks of Little Ravens Corvus mellori and currawongs Strepera spp. called as they fed on the moths. In 2017-19, there was an unprecedented dearth of moths. The boulder fields were silent! In spring-early summer of 2017 and 2018, widespread loss of pouch young litter was observed (DH) in B. parvus populations due to starvation. This was unprecedented (never a single incidence in more than 10 000 observations) in 40 years of annual monitoring.
Here we present data collected on A. infusa in alpine boulder fields (Victoria and NSW) during a PhD study on B. parvus (Mansergh 1988, Box 1). The study sought to quantify the contributions of the moths to: (a) B. parvus populations; (b) alpine soil fertility within boulder fields; and (c) the scale of nutrient transfer to food chains in the alpine region. Given the collapse of A. infusa migration to the Alps (breeding failure, lack of juvenile recruitment), these data are augmented with observations of B. parvus sites monitored over a 41-year period in an effort to highlight little-studied implications of moth decline for the health of alpine ecosystems. We trust this will contribute to understanding the urgent need for further research and monitoring.
The broad study area is the Bogong and Hotham High Plains in Victoria's alpine region (Fig. 2). Details (maps, photographs, grid sites) of specific collection locations, including those in NSW, are provided in Mansergh (1988) (dig-itised--http://hdl.handle.net/1959.9/567012) and in Mansergh and Broome (1994). Our soil samples were collected within basalt boulder fields where their depth was between 0.5 and 1.0 m--litter was deep and moist. In granite boulder fields, soil samples were collected at the base of boulders--litter was less abundant than in the basalt sites.
A. Soil fertility and organic matter
A sample consisting of 10 two-cm diameter cores of soil was taken to depths of 25 mm and 100 mm over an area of approximately 10 [m.sup.2] within 3 m of known B. parvus capture sites. At each site, three samples were taken at altitudes above 1760 m on both western and eastern aspects. Samples were collected in February 1985 on Mt Higginbotham, Victoria (basalt), and in February 1986 in Kosciuszko National Park, NSW (granite) (hereafter KNP--western aspect Charlottes Pass; eastern aspect Upper-Lower Blue Cow). Basalt boulder fields were 0.3 m deep at the margins and were often > 3 m deep in the centre. In contrast, granite boulder fields had a tor structure with much more limited interstitial spaces (Box 1; photographs of sites in Mansergh 1988; Mansergh and Broome 1994). Samples were examined by the (then) State Chemistry Laboratory of Victoria (SCLV) using standard tests for: potassium (K); Phosphorus (P); pH value and oxidisable carbon. A state-wide context was provided by AT Brown (pers. comm., SCLV, 1986: full text in Mansergh 1988).
B. Abundance of Agrotis infusa in litter at different aspects and altitudes
In each of three altitude bands (1600-1650 m, 1650-1750 m, and [greater than or equal to] 1780 m) on both easterly and westerly aspects of Mt Higginbotham, six 15 X 15 X 2 cm litter samples were collected in February 1985. Sites were chosen along B. parvus trapping lines (visited [greater than or equal to] 7 times) where A. infusa were most abundant. Each collection site was within the boulder field, at the base of boulders, and was at least 2 m distant from any other. In one instance (westerly site lowest altitude) a trapping site was not available. Similarly, in February 1986, samples were collected in KNP at sites used in Method section A (above), and a single high altitude westerly site on Mt Bogong (Victoria). All samples were examined and all wings and anatomical parts counted. A minimum number of moths was estimated from the larger count, by dividing major wings by two or the number of all wings by four.
These data were analysed with B. parvus abundances at the same sites (mean number per visit up to June 1985--B. parvus trapping records). The association between A. infusa and B. parvus counts was examined using the product-movement correlation coefficient calculated from its six pairs of transformed values from sites with a wide range of A. infusa counts.
C. Accumulation of Agrotis infusa over time
A masonite board (15 X 15 cm) was placed on the ground at six sites systematically chosen within each of 11 Burramys trapping replicates (each 0.2 ha) on Mt Higginbotham (n=66: 18 west and 48 east). Boards were visited between November 1984 and January 1986 and litter and A. infusa debris collected.
D. Observations of fauna in boulder fields 1978-2019
Apart from studies reported here, the abundance of A. infusa was not studied or monitored in Victoria as the sheer regularity of the abundance of the species indicated (wrongly as it turns out) that it was unlikely to be a limiting factor in the abundance of B. parvus, recognising the natural differences between the basalt habitats of Victoria and those of granite in NSW. Notwithstanding, during the annual B. parvus monitoring program (> 20 days and > 4000 trap-nights p.a.) good and poor years for A. infusa were generally noted; observations on the general health of the habitat and mammalian and other faunal abundance and use were made (e.g. Mansergh and Broome 1994; Heinze et al. 2004).
A. Soil fertility and organic matter
The results of the six soil parameters examined are provided in Table 1. Both sites had high levels of total nitrogen (0.57-1.81%) and organic carbon (9.7-27.2%). Basalt-derived soils are inherently more fertile than those from derived granite: however, the aestivation of A. infusa in Victorian basalt boulder fields makes soil fertility significantly richer than granite soils in every nutrient (N, K, P and Oxidisable C) measured (Table 1). Agrotis infusa formed a large portion of the organic residue at Mt Higginbotham, whereas in KNP most organic residue was of vegetation alone (Table 1). In general, as expected, the 0.25 mm samples have a higher K, P, N, and C than the 0-100 mm samples at both sites, except for available P at Mt Higginbotham, which was similar at both depths (Table 1). Due to the abundance of A. infusa remains (see Methods section B), Mt Higginbotham soils are more fertile than KNP soils, especially in relation to available P, which is very high in the former locality but marginal at KNP.
B. Abundance of Agrotis infusa in litter at different aspects and altitudes
Agrotis infusa remains were markedly more common in basalt than granite sites. The total number of moths collected at Mt Bogong (12 on granite) and in KNP (23) provided insufficient data for analysis. In contrast, A. infusa remains were recorded at 14 000 per [m.sup.2] at one site in the high-altitude basalt boulder field of Mt Higginbotham. After a variance-stabilising transformation to logarithms, a two-way analysis of variance of the 36 samples showed that the frequency of A. infusa varied with both altitude and aspect (interaction effect F = 4.57 on 2 and 30 D.F., P < 0.02) (Table 2). High altitude westerly aspects were favoured (Table 2) whilst there appeared to be little difference between aspects at lower altitudes (< 1650 m), although it is suspected that this is due to the relative paucity of A. infusa generally at this altitude (Table 2). There is a very strong positive association between the abundances of A. infusa and B. parvus (r = 0.9087, P < 0.01, Fig. 3).
As remains (dead moths) were recorded at 14 000 per [m.sup.2] (using only samples to 25 mm depth) and, at higher altitudes, nutrient availability remains similar down to 100 mm due to the moths (Table 1), it is concluded that A. infusa bio-transports a most significant nutrient input into the alpine environment.
C. Accumulation of Agrotis infusa over time
The method failed to detect moth accumulation over time, probably because of air movement and/or animal activity. The boards were visited an average of five times (range 2-7) with 330 total visits over the period. A total of 48 moths were recorded (47 on high western slopes) on 10 occasions (3%). This method is not recommended without refinement.
D. Observations at Agrotis infusa aestivation boulder fields
Victorian boulder fields where Agrotis infusa aestivates:
1. The arrival time of the moths was usually late September to early October and moths tended to arrive en masse, aided by the NW winds. In contrast, the departing period was more extended--segments of the population could begin leaving in mid-February and few moths were observed in the field by mid-April. In still weather, regular sunset flights may have allowed evaluation of the less reliable SE winds to aid return to natal sites.
2. Aestivating moths favour, and help create, a micro-climate out of the wind and sun, hence their attraction to boulder fields and caves. The structure of the deep basalt boulder fields allows three-dimensional aestivation sites, most of which are inaccessible to observation (see Mansergh 1988; Box 1; Fig. 1). However, at many sites and during many years, aestivating moths could be observed 100 mm deep (> depth than index finger) in the interstitial spaces. In November 1987, moth congregations were so large that suboptimal space in afternoon sunlight had to be occupied (Fig. 1). Interstitial spaces (aestivation sites) are 30-50% of the volume of basalt boulder fields (Box 1, Fig 1B). Densities of moths as shown in Fig 1A suggest that a cubic metre of boulder field could support >50 000 aestivating A. infusa.
3. Victorian Agrotis infusa aestivation sites provide a concentrated food resource for small mammals and support the highest densities of small mammals in any alpine and subalpine environment. These moths are a major component of the spring-summer diets of the Bush Rat Rattus fuscipes (frequency in scats > 51 % of scats), Dusky Antechinus Antechinus swainsonii (49 %) and B. parvus (69 %) (Mansergh 1988; Mansergh et al. 1990). Adult female B. parvus remain sedentary in these basalt sites, and densities of 95 per ha have been recorded, whereas 24 adults (male and female) per ha is the highest recorded in granite sites (Blue Cow) (Mansergh and Broome 1994). Agrotis infusa may be the exclusive prey item of females during breeding in both basalt and granite sites (Mansergh 1988; Smith and Broome 1992; Heinze et al. 2004). However, increased abundance of moths in basalt A. infusa aestivation sites increases densities of B. parvus four-fold compared to the granite sites. High concentrations of moths were also found in Red Fox Vulpes vulpes scats.
4. During the day (November-March), flocks of ravens and currawongs (often 50 per ha) feed on the moths, and could be 'heard' from a kilometre away. From 1978 this was a consistent feature of the sites. From 2017 to May 2019 the sites were silent and these species were not observed.
5. Microchiropteran bats were often foraging during the twilight emergence of A. infusa. Over several years, a variety of bats has been recorded (heard, trapped) (pers. obs.; Martin Schulz, pers. comm., April 2019) adjacent to the boulder fields. They include Gould's Wattled Bat Chalinolobus gouldii, Lesser Long-eared Bat Nyctophilus geoffroyi and the White-striped Free-tailed Bat Austronomus australis.
6. Although no quantified monitoring of A. infusa numbers was undertaken, their numbers were abundant during 'wetter' decades (late 1970 to mid 1990s), but began to lessen after 1998/99 and were substantially diminished during the Millennium Drought up to mid-2010. Notable abundant years recorded for A. infusa were: 1978/79, 1979/80, 1982/83, 1985/86, 1986/87 and 2013/14. After the high rainfall years of 2010 and 2011, an exceptional Bogong Moth season occurred in the Alps in the summer of 2013/14. Subsequently, A. infusa was very abundant in the Victorian lowlands in the winter of 2014 (G McDonald, pers. comm., October 2019). In the Alps, 2014-16 were low abundance years, with a total collapse in the summer of 2017 that continued yearly to May 2019. Pouch young starvation was first observed at Timms Spur in December 2016 (last population examined late in breeding season) but was widespread in 2017-18.
7. Over the 41-year period, individual and small clusters of A. infusa were observed across alpine environments under rocks, rocky outcrops and in various types of vegetation, e.g. heathland.
The results emphasise the importance of A. infusa for the following reasons: (a) it is an important food source for B. parvus and other mammalian and avian species in the alpine region; (b) it brings with it a massive bio-transfer of nutrients (input to the alpine environment); and (c) it has been an abundant resource available to Indigenous peoples over millennia. These highlight the magnitude of the loss to the alpine environment caused by the recent unprecedented collapse in A. infusa populations. The significant nutrient transfer from the annual A. infusa migration and the geographic extent of basalt boulder fields/rocky outcrops across alpine Victoria (Fig. 2) indicates that A. infusa makes a massive input to the soil nutrient health of the region.
From aerial photography, O'Brien and Gowans (1985) identified 1454 ha of boulder fields in Victorian alpine, subalpine and montane areas above 1200 m ASL and a further 350 ha with appropriate geology where boulder fields could not be identified. The most extensive areas occur between Mt Bogong and Mt Higginbotham with outliers around Mt Buller and Mt Wombargo-Cobberas--the first two support known B. parvus populations. Many of the boulder fields occurred in patches < 3 ha. Evidence of A. infusa was invariably noted at boulder field sites trapped for B. parvus, even when unsuccessful. All sites with permanent B. parvus populations (Mt Bogong, Timms Spur, Mt McKay, Pretty Valley, Bundarra, Mt Loch, Mt Higginbotham and Little Higginbotham, Mt Buller) had aestivating A. infusa in abundance (e.g. Heinze et al. 2004; Fig. 2). Although very abundant at aestivation sites, individual and small clusters of A. infusa were also observed in small clumps of basalt rocks above 1500 m (Fig. 2).
The presence of A. infusa remains, including in scats, make the soils of boulder fields much more fertile than the surrounding landscapes. This is markedly so in basalt boulder fields. The high levels of total nitrogen (0.57-1.81%) and organic carbon (9.7-27.2%) far exceed that for typical 'virgin' bushland (0.1-0.2% and 2-4% respectively, Brown 1986). Kirkpatrick et al. (2014) recognised eight soil classes across the mainland alpine area: the variation in the mean extractable P (ppm), excluding soils in bogs (160 ppm) was 6 to 45 ppm; total N (%)--1.2 to 0.3%; and, organic C (ppm)--6 to 30 ppm (c.f. Table 1, see also Rowe 1970). The abundance of these critical components at A. infusa aestivation sites indicate the scale of the bio-transfer of nutrients (Table 1).
Most published data on A. infusa aestivation derives from caves in NSW (e.g. Common 1954; Warrant et al. 2016) and these may be more favoured and productive than granite boulder fields available in NSW (Table 1). Indeed, Common (1954) found moths on cave walls and ceilings at 17 000 per [m.sup.2]. In contrast, basalt boulder fields are a common type on Bogong and Hotham High Plains (Fig. 2) and appear to be as productive, if not more so, than NSW caves. Basalt boulder fields, as opposed to caves, provide both more geographic spread and vectors (diversity and amount) for the secondary bio-transfer of nutrients. The recent discovery that the moths navigate using magnetic fields (Dreyer et al. 2018) may also assist location of the basalt boulder fields and upslope flows which have high magnetic signatures (Morand et al. 2005).
Victorian alpine caves have not been well studied. However, since December 1967, Monika Baker (pers. comm., 2019) has annually recorded the presence of A. infusa in caves at three altitudes on Mt Buffalo. This remarkable record indicates that moths were always abundant in the higher caves and regularly used the lower caves. She found few moths in January 2018 and none in January 2019. The Mt Buffalo collapse--a 52-year record--mirrors observations on the Bogong High Plains documented here.
The in situ small mammal fauna respond to availability of A. infusa. The arrival of various Microchiroptera in the Alps coincides with moth arrival and is frequently observed when the moths emerge at sunset over summer. Although Green and Osborne (1994) noted that the quantitative extent of predation is unknown, our observations would suggest that A. infusa could be a major food item particularly for the medium to larger species (e.g. White-striped Free-tailed Bat, Eastern False Pipistrelle Falsistrellus tasmaniensis) and those common in adjacent Snow Gum woodlands, (e.g. Gould's Wattled Bat: Norris et al. 1983). Microchiroptera typically eat up to half their body weight per night. Churchill (1998) and Vestjens and Hall (1977) observed A. infusa as the major food component in the stomachs of Gould's Wattled Bat and Lesser Long-eared Bats in alpine NSW.
Consistently dense distributions of large bird species leave guano on site but also disperse nutrients through their droppings further afield. Similar to droppings of birds and Red Foxes, those of bat species would disperse nutrients in a secondary bio-transfer away from aestivation sites. We did not quantify the mothderived guano dispersal by these species across the alpine area during the study, but contend it would be significant. In KNP, female feral pigs Sus scrofa have recently been shown to exploit alpine cave aestivation sites (Caley and Welveart 2018). Agrotis infusa have also transferred arsenic to the Alps (Green et al. 2001; Green 2008; Love 2010). The bio-transfer of nutrients (and contaminants) through seabird guano is environmentally very significant and well-studied (e.g. Ellis et al. 2006).
Microchiroptera, fox and avian predation would disperse nutrients further afield in a secondary bio-transfer. This dispersal of nutrients is important as a large proportion of the available plant nutrients of alpine soils remain concentrated in the top 5-10 cm of the soil profile (Rowe et al. 2017). Dust storms at 10 to 20 year intervals provide nutrients to soils in the Snowy Mountains; however, the contribution of aeolian dust and the in situ P cycle remains unclear (see Kirkpatrick et al. 2014). Other sources of nutrient input into the alpine landscape would come via rain (N) and weathering. We suggest that the annual moth migration, over millennia, has added substantial critical nutrients (e.g. P) to the alpine region.
In the globally unique 'soil' mountains of Australia a large proportion of the nutrients is concentrated in the surface 5-10 cm of the soil and the cycling of these nutrients is key to vegetation health/type and moisture retention capacity (Rowe et al. 2017; Kirkpatrick et al. 2014). From predator consumption, parasitism and adverse weather, Green (2011) estimated that the remains of a billion moths provided an annual 5000 gigajoules of energy to the Alps in New South Wales. Our study observed the same suite of predators. We support Green's conclusion that A. infusa makes a very significant nutrient input into the health of the alpine environment, its vegetation and water-holding capacity (see Rowe et al. 2017). The magnitude of the annual A. infusa migration helps avoid nutrient entropy in the alpine region.
The abundance of A. infusa outside caves helps explain much of the difference between the densities of B. parvus observed in Victorian basalt sites compared to the densities observed in KNP (Mansergh and Broome 1994; Heinze et al. 2004; Fig. 3; Method A). At the KNP sites (see Methods section A) where A. infusa are not as abundant as at Mt Higginbotham (see Results section A), moths may still be the only prey item in female scats during breeding even if nightly movement to congregations is required (Smith and Broome 1992). Burramys parvus is relatively long-lived (females up to 12 years, males up to four years) (Mansergh and Broome 1994).Breeding/lactating female B. parvus and their young are highly dependent on the moths for nutrition. The collapse of the A. infusa population has had an immediate adverse effect causing starvation among B. parvus pouch young across its Victorian range (Australian Broadcasting Company 2019; Readfearn 2019; DH, pers. obs., 2016 to present). Disturbing though this is, the first-year recruitment to the breeding population remains consistent with a 35-year data set. At this stage, direct interventions in the population (e.g. supplementary feeding, removal/return) are to be avoided as this would further stress the wild population. Perverse outcomes are more likely than any perceived benefits (e.g. Weeks et al. 2017--for captive release). Continued absence of moths will lead to population decline, particularly where high densities of B. parvus are maintained by a huge abundance of moths (all Victorian basalt sites).
Although we did not monitor the micro-climate of the boulder fields, we have observed no adverse changes (except absence of A. infusa) in the aestivation sites of A. infusa that could help explain the recent population crash. Thus, the cause of the collapse would appear to derive from either interruption to the migration pattern or disturbance at the natal sites. Change in the strength and/or timing of winds that aid migration, NW spring and SE late summer, may have affected migration, particularly the latter. However, this would appear unlikely to have been the major cause of such a sudden collapse as moths can keep their direction even against wind (Warrant et al. 2016).
Radical environmental change at the natal sites is therefore suspected. Although much is known of cutworms in agricultural landscapes, knowledge of their pre-European habitat (and use of remnant vegetation) remains surprisingly sparse although they were common in saltbush country that has high seasonal dicotyledon abundance in the Riverina (Common 1954; Warrant et al. 2016; David Cheal, pers. comm., April 2019). There is a strong correlation between the range of A. infusa and the historical range of Yam daisies Microseris spp. in eastern Australia (see A. infusa cutworm map in Green 2011; Walsh 2016). Within that broad range the favoured habitat of both species (alluvial self-mulching soils) is more restricted. We suspect Yam daisies were an important part of A. infusa natal areas but with decline of the daisies the moths were able to exploit introduced pastures/crops and become a recognised pest. Drought refugia utilised by A. infusa (presumably moister, more fertile nodes in the landscape) would be a much smaller subset of their broader habitat, which itself is a subset of the national range.
Declines, but not collapse, in the abundance of A. infusa in the alpine region were observed throughout the Millenium Drought (1996 to mid-2010) that occurred across their natal sites, presumably because they were able to utilise drought refugia. Record rains in 2010-11, and consequent increase in moisture content and plant growth, may explain peak A. infusa numbers in 2013-14. Populations recovered from the Millenium Drought. Subsequently, drought conditions resumed and have intensified to the present (July 2019). Agriculture has intensified over these areas in the last 20 years, e.g. water use in cotton cultivation (Cotton Australia 2019). Agrotis infusa habitats in the self-mulching soils of the alluvial plains are favoured agricultural areas, with competition magnified in refugia areas during drought.
The declines in A. infusa abundance since the early 2000s correlate with drought conditions, land-use change and also the widespread adoption of neonicotinoids as a pesticide across natal areas (Nash et al. 2019). Available from the mid-1990s, neonicotinoids, a group of insecticides colloquially called neonics, are now the most widely used insecticides in the world, but only about 5% of the chemical remains in the plant--the remainder is dispersed and persists in the wider environment (Wood and Goulson 2016). In 2013, neonicotinoids were partially banned in Europe due to their adverse effects on bees in certain crops (Wood and Goulson 2016). Wood and Goulson (2016) conducted a review of the effects on a range of biota, but unfortunately no studies were conducted on lepidoptera. They concluded that there is a growing body of evidence that persistent, low levels of neonicotinoids can have a negative impact on a range of free-living organisms. Neonicotinoid use in Australia was relatively stable from 1999 to 2016, but then jumped 34% in 2016-17 and has subsequently remained at these higher levels (Nash et al. 2019) (i.e. contemporaneous with the collapse of A. infusa). The Australian Pesticides and Veterinary Medicines Authority (APVMA) recommended residue monitoring of neonicotinoids in 2014, yet despite increased usage, no monitoring appears to have been undertaken to date (https//apvma.gov.au/pubcris). We strongly suspect that neonics are partially responsible for the collapse of the A. infusa population.
The recent rapid and unprecedented population crash of the annual A. infusa migration is a cause for national concern, particularly as it occurred after a peak year in 2014. There is evidence of worldwide decline in insect populations (Thomas et al. 2004); however, the dramatic collapse of the A. infusa population demands specific attention. The observed effects on endangered species (widespread starvation of B. parvus pouch young) and potential effects on the ecological health of alpine areas, including plant nutrients, vegetation health (and thus water-holding capacity), are now well documented. Are B. parvus pouch young 'canaries in the coal mine' that warn of dramatic environment change over huge areas of our agricultural landscape? The main scientific and conservation action should be concerned with the recovery of the A. infusa population. There is an urgent need for national action: for APVMA and Environment Australia to implement their own recommendations and for immediate additional research and monitoring of the A. infusa population in both their alpine summer habitats as well as in their natal habitats in agricultural areas to identify and protect these natal areas in terms of their native vegetation and drought refugia.
Constructive comments and ideas on the draft were generously provided by John Morgan, Peter Green, Martin Schulz, Harry Parnaby and David Cheal. Michael Nash provided information on neonicotinoides and his recent publication. Monica Baker generously provided access to her remarkable 52-year data set on Bogong Moths. Neville Rosengren generously compiled Fig. 2 and geomorphological information. Garry McDonald (CESAR, Parkville) provided information on Bogong Moth abundance in the lowlands. Many thanks to the two peer reviewers who significantly sharpened the text.
This contribution is humbly dedicated to the late Alan Yen, a fine entomologist and a great supporter of FNCV and conservation in Victoria.
Australian Broadcasting Corporation (ABC) (2019) Report on Bogong moth collapse and death of Burramys parvus pouch young (Dean Heinze). Televised 30 March 2019. <https://iview.abc.net.au/show/abc-news-vic>
Brown AT (1986) Report on the soil analysis of 24 samples taken in B. parvus habitat on Mt Higginbotham and Koscuisko National Park. State Chemistry Laboratories, Victoria, Melbourne. (Full text in Mansergh 1988).
Caley P and Welvaert M (2018) Aestivation dynamics of bogong moths (Agrotis infusa) in the Australian Alps and predation by wild pigs (Sus scrofa). Pacific Conservation Biology 24, 178-182.
Churchill S (1998) Australian Bats. (Reed New Holland: Sydney).
Common IFB (1954) A study of t ecology of the adult Bogong Moth, Agrotis infusa, (Boisi.) (Lepidoptera: Noctuidae), with special reference to its behaviour during migration and aestivation. Australian Journal of Zoology 2, 223-263.
Common I (1989) Moths of Australia. (Melbourne University Press: Melbourne).
Cotton Australia (2019) Australian cotton industry review. <https://cottonaustralia.com.au/cotton-library/factsheets> (accessed 10 October 2019).
Department of Environment, Land, Water and Planning (DELWP) (2016) National Recovery Plan for Mountain Pygmy-possum, Burramys parvus. (Environment Australia: Canberra).
Dreyer D, Frost B, Mourtisen H, Gunther A, Green K, Whitehouse M, Johnsen S, Heinze S and Warrant E (2018) The Earth's Magnetic Field and Visual Landmarks Steer Migratory Flight Behaviour in the Nocturnal Australian Bogong Moth. Current Biology 28 (13), 2160-2166.
Ellis JC, Miguel Farina J and Witman JD (2006) Nutrient transfer from sea to land: the case of gulls and cormorants in the Gulf of Maine. Journal of Animal Ecology 75, 565-574.
Flood J (1980) The moth hunters: Aboriginal prehistory of the Australian Alps. (Australian Institute for Aboriginal Studies: Canberra).
Green K (2008) Migratory bogong moth (Agrotis infus) transport arsenic and concentrate it to lethal effect by estivating gregariously in alpine regions of the Snowy Mountains of Australia. Arctic, Antarctic and Alpine Research 40, 74-80.
Green K (2011) The transport of nutrients and energy into the Australian Snowy Mountains by migrating Bogong Moths Agrotis infusa. Australian Ecology 36, 25-34.
Green K, Broome L, Heinze D and Johnston S (2001). Longdistance transport of arsenic by migrating Bogong moths from agricultural lowlands to mountain ecosystems. The Victorian Naturalist 118, 112-116.
Green K and Osborne W (1994) Wildlife of the Australian snow-country. (Reed: Sydney).
Gullan P and Norris K (1984) The habitat of the Mountain Pygmy-possum (Burramys parvus) in Victoria. In Possums and Gliders, pp. 417-21. Eds AP Smith and ID Hume (Surrey Beatty & Sons in association with the Australian Mammal Society: Chipping Norton, NSW).
Heinze D, Broome L and Mansergh I (2004) An overview of the ecological studies on the Mountain Pygmy-possum Burramysparvus (Broom 1896). In The Biology of Australian Possums and Gliders, pp. 254-267. Eds RL Goldingay and SM Jackson. (Surrey Beatty & Sons: Chipping Norton, NSW).
Kirkpatrick JB, Green K, Bridle KL and Venn SE (2014) Patterns of variation in Australian alpine soils and their relationships to parent material, vegetation formation, climate and topography. Catena 121, 186-194.
Love P (2010) Spatial and temporal characteristics of arsenic in the Bogong Moth (Agrotis infusa). (Unpublished PhD thesis, La Trobe University, Bundoora).
Mansergh I (1988) The conservation and management of the Mountain Pygmy-possum (Burramys parvus) in Victoria with reference to New South Wales. (PhD thesis, La Trobe University, Melbourne) <http://hdl.handle.net/1959.9/567012>
Mansergh IM, Baxter B, Scotts D and Brady T (1990) Diet of the Mountain Pygmy-possum, Burramys parvus (Marsupialia: Burramyidae) and other small mammals in the alpine environment at Mt Higginbotham, Victoria. Australian Mammalogy 13, 167-177.
Mansergh I and Broome L (1994) The Mountain Pygmypossum of the Australian Alps. (University of NSW Press: Sydney) (Reprinted in 1996).
Morand VJ, Simons BA, Taylor DH, Cayley RA, Maher S, Wohlt KE and Radojkovic AM (2005) Bogong: 1:100 000 map area geological report--GeoScience Victoria. Geological Survey of Victoria Report 125. <http://earthresources.efirst.com.au/product.asp?pID=10&cID=39> (accessed October 2018).
Nash M, Severtson D and Macfadyen S (2019) New approaches to manage invertebrate pests in conservation agriculture systems--uncoupling intensification. In Australian Agriculture in 2020: From Conservation to Automation, pp. 191-204. Eds J Pratley and J Kirkegaard. (Agronomy Australia and Charles Sturt University: Wagga Wagga).
Norris KC, Mansergh IM, Ahern LD, Belcher CB, Temby ID and Walsh NG (1983) Vertebrate fauna of the Gippsland Lakes Catchment. Victorian Ministry of Conservation, Fisheries and Wildlife Division, East Melbourne, Occasional Paper Series 1, p. 158.
O'Brien M and Gowans RM (1985) The distribution of boulder streams and Podocarpus heath, components of the preferred habitat of Burramys parvus, in alpine and sub-alpine areas of eastern Victoria. Arthur Rylah Institute Environmental Research Technical Report No. 8, Heidelberg.
Readfearn G (2019) Decline in bogong moth numbers leaves mountain pygmy possums starving. The Guardian, 25 February 2019. <https://www.theguardian.com/environ-ment/2019/feb/25/decline-in-bogong-moth-numbers-leaves-pygmy-mountain-possums-starving?CMP=share_btn_link> (accessed March 2019).
Rowe K (1970) A study of the land in Mount Buffalo National Park. (Soil Conservation Authority: Melbourne)
Rowe K, Gibbons F and Anderson H (2017) High Mountain soils. In The alpine ecology course notes--2017, pp. 71-94 (Research Centre for Applied Alpine Ecology, La Trobe University: Bundoora, Victoria).
Smith A and Broome L (1992) The effects of season, sex and habitat on the diet of the Mountain pygmy-possum (Burramysparvus). Wildlife Research 19, 755-767. <https://doi.org/10.1071/WR9920755> (accessed 16 June 2019).
Thomas JA, Telfer MG, Roy DB, Preston CD, Greenwood JJD, Asher J, Fox R Clarke RT and Lawton JH (2004) Comparative Losses of British Butterflies, Birds, and Plants and the Global Extinction Crisis. Science 303, 1879-1881.
Vestjens WJM and Hall LS (1977) Stomach contents of fortytwo species of bats from the Australasian region. Australian Wildlife Research 4, 25-35.
Walsh N (2016) A name for Murnong (Microseris: Asteraceae: Cichorioideae). Muelleria 34, 63-67.
Warrant E, Frost B, Green K, Mourtisen H, Dreyer D, Adden A, Braubuger K and Heinze S (2016) The Australian Bogong moth Agrotis infusa: a long distance nocturnal navigator. Frontiers in Behavioural Neuroscience, 21 April 2016. <https://doi.org/10.3389/fnbeh.2016.00077> (accessed April 2019).
Weeks A, Heinze D, Perrin L, Stoklosa J, Hoffmann A, Rooyen A, Kelly T and Mansergh I (2017) Genetic rescue increases fitness and aids rapid recovery of an endangered marsupial population. Nature Communications 8, 1071.
Wood TJ and Goulson D (2016) The environmental risks of neonicotinoid pesticides: a review of the evidence post 2013. Environmental Science and Pollution Research 24, 17285-17325.
Received 25 July 2019; accepted 7November 2019
Ian Mansergh (1) and Dean Heinze (2)
(1) Research Centre for Applied Alpine Ecology, La Trobe University, Bundoora, Victoria 3084
(2) Ecologist, Collinsvale, Tasmania 7012
Box 1. Interstitial spaces in granite and basalt boulder fields
Bogong moths Agrotis infusa seek cool, sheltered microclimates for aestivation in the alpine region. Caves and interstitial spaces in boulder fields provide such habitat. The geomorphology of boulder fields of granite (NSW) and basalt (Victoria) provide different types and amounts of sheltering sites that affect A. infusa abundances. Peri-glacial freezing and thawing exploited fractures in outcropping basalt, characteristically breaking angular blocks 0.3-0.6 m (longest axis) that formed boulder fields to depths of > 4 m. Interstitial spaces may be 30-50% of the volume (N Rosengren, pers. comm., October 2019). Mountain plumpine Podocarpus lawrencei establishes when the boulder fields are c. 2 m deep and, as soil and litter accumulates, a unique ecological vegetation community of P. lawrencei heathland evolves (see Gullan and Norris 1984). Granite boulder fields support P. lawrencei heathland but typically have larger rounded boulders (remnant corestones 2-3 m) resulting in relatively fewer interstitial spaces than for basalt boulder fields (Fig. 1A/1B; Mansergh and Broome 1994).
Table 1. Soil analysis of six standard fertility parameters in A. infusa aestivation boulder fields on Mt Higginbotham (Vic) and in Koscuiszko National Park (NSW) (n = 24). Analysis courtesy of State Chemistry Laboratories, Victoria (Brown 1986). Locality Altitude (m) Aspect Sample Depth (1) (0 mm) Mt Higginbotham 1760-1800 West 25 dark grey-brown peat 25 basalt 25 100 100 100 East 25 25 25 Koscuiszko NP 100 Charlottes Pass 100 granite 100 dark grey-brown peat 1820-1950 West 25 25 25 100 100 100 Koscuiszko NP 1800-1900 East 25 Blue Cow 25 granite 25 grey-brown peat 100 100 100 Significance (4) -Locality -Aspect -Aspect x Locality ** -Depth Locality Altitude (m) Total Reaction(pH) Nitrogen (2) ([H.sub.2]O Mt Higginbotham 1760-1800 1.81 4.9 dark grey-brown peat 1.47 4.9 basalt 1.78 4.9 1.36 5.1 1.55 4.9 1.46 4.9 1.38 5.6 1.72 5.8 1.59 5.0 Koscuiszko NP 1.29 5.8 Charlottes Pass 1.49 5.1 granite 1.58 4.8 dark grey-brown peat 1820-1950 0.75 5.0 1.06 4.4 0.55 5.0 0.68 4.8 0.75 4.7 0.57 4.9 Koscuiszko NP 1800-1900 1.24 4.6 Blue Cow 1.11 4.7 granite 0.89 5.1 grey-brown peat 1.07 4.6 0.90 4.9 1.15 4.6 Significance (4) -Locality *** ** -Aspect * -Aspect x Locality * -Depth Locality Altitude (m) Reaction(pH) Electrical Ca[Cl.sub.2]) Conductivity ds/m Mt Higginbotham 1760-1800 4.1 0.16 dark grey-brown peat 4.1 0.12 basalt 4.3 0.10 4.4 0.08 4.2 0.10 4.2 0.09 5.1 0.14 5.4 0.09 4.4 0.12 Koscuiszko NP 5.2 0.08 Charlottes Pass 4.5 0.95 (3) granite 4.2 0.07 dark grey-brown peat 1820-1950 4.1 0.06 4.1 0.08 4.3 0.07 4.1 0.15 4.0 0.08 4.2 0.07 Koscuiszko NP 1800-1900 3.9 0.15 Blue Cow 3.9 0.08 granite 4.4 0.08 grey-brown peat 4.2 0.10 4.0 0.08 4.0 0.11 Significance (4) -Locality -Aspect -Aspect x Locality -Depth Locality Altitude (m) Skene(K) Olsen P Potassium Phosphorous ppm ppm Mt Higginbotham 1760-1800 540 115 dark grey-brown peat 477 83 basalt 529 97 454 81 405 112 382 110 788 71 534 63 595 71 Koscuiszko NP 565 76 Charlottes Pass 450 64 granite 469 64 dark grey-brown peat 1820-1950 417 20 410 16.5 243 13.5 307 16.1 323 15.7 249 12.7 Koscuiszko NP 1800-1900 387 18.7 Blue Cow 308 13.2 granite 310 11.5 grey-brown peat 330 12.1 258 8.2 257 9.3 Significance (4) -Locality *** *** -Aspect *** -Aspect x Locality ** -Depth ** Locality Altitude (m) Oxidisable Carbon (%) Mt Higginbotham 1760-1800 23.4 dark grey-brown peat 19.2 basalt 23.0 16.9 19.1 17.7 19.1 27.2 24.5 Koscuiszko NP 16.6 Charlottes Pass 19.4 granite 21.1 dark grey-brown peat 1820-1950 12.2 18.8 9.7 9.9 13.2 9.3 Koscuiszko NP 1800-1900 17.8 Blue Cow 16.6 granite 15.2 grey-brown peat 15.7 12.4 17.0 Significance (4) -Locality *** -Aspect * -Aspect x Locality -Depth * (1) At Mt Higginbotham A. infusa remains up to 50 mm depth (mean 25 mm) plus vegetation residue; at Mt Koscuiszko most organic residue was vegetation alone. (2) Dumas % method. (3) Chloride as NaCl in this sample was 0.015% and is a normal, low level. (4) 3-way ANOVA significance levels: *** = < 0.001; ** = < 0.01; * = < 0.05. Table 2. Log means of the number of Agrotis infusa collected at three high altitude bands on westerly and easterly aspects on Mt Higginbotham, January 1985. Aspect Altitude band (m) >1750 1750-1650 <1650 West 4.118 2.617 0.5972 East 2.341 1.647 0.1832
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Research Report|
|Author:||Mansergh, Ian; Heinze, Dean|
|Publication:||The Victorian Naturalist|
|Date:||Dec 1, 2019|
|Previous Article:||The secret world of the Net-casting Spider Deinopis subrufa.|
|Next Article:||Assessing consumption of Curiosity[R] feral cat baits by Blotched Blue-tongued Lizards.|