(C) Common Dreams
This story was originally published by Common Dreams and is unaltered.
. . . . . . . . . .



Field evidence of bird poisonings by imidacloprid-treated seeds: a review of incidents reported by the French SAGIR network from 1995 to 2014 [1]

['Millot', 'Florian.Millot Oncfs.Gouv.Fr', 'Research Department', 'National Game', 'Wildlife Agency', 'Office National De La Chasse Et De La Faune Sauvage', 'Oncfs', 'Saint Benoist', 'Decors', 'Mastain']

Date: 2017-02-05

Since 1995, incidents have been detected almost every year and related to imidacloprid-treated seeds of different crops with different sowing periods. Furthermore, mortality due to poisoning by treated seeds was ranked as at least “likely” in 70% of incidents. Various environmental and anthropogenic factors affect the collection of wildlife mortality data (e.g. Berny 2007; de Snoo et al. 1999; Vyas 1999; and below). As a result, it is well acknowledged that the actual number of wildlife mortality largely exceeds the number of carcasses actually recovered (Vyas 1999). Consequently, the first major contribution of this article is to confirm that in real conditions granivorous birds are regularly exposed to imidacloprid-treated seeds that can result in acute lethal or sublethal effects. This work also shows that the two main factors (seed burying and avoidance of treated seeds) supposed to mitigate the risk for granivorous birds are not completely efficient in natura.

Efficacy of mitigation measures

Seed burying

For the incidents reported by the SAGIR, we are not able to estimate to what extent these incidents may be attributed to the noncompliance of use instructions (e.g. spillage not removed). The presence of piles of spilled seeds has been reported in very few incidents. However, a precise assessment of the amount of seed available at the soil surface of sown fields around the incident discovery sites was not systematically performed. In addition, distinguishing between a clear noncompliance of good agricultural practices and an actual technical inability for farmer to strictly follow the use conditions may be sometimes tenuous.

In France, to protect wild birds and mammals, label instructions are to both remove any spilled seeds and incorporate all dressed seeds into the soil. Yet, in routine use, seed burying is rarely 100% effective, especially on the headland (i.e. the place where the drilling implement is turned when sowing) and various factors affect the proportion of seeds actually buried (e.g. Pascual et al. 1999a, b; de Snoo and Luttik 2004). For example, without taking spillage into account, de Snoo and Luttik (2004) found that 9.2% of drilled winter wheat seeds remain on the soil surface on the headland. In France, the winter wheat seeding rate varies from 150 seeds/m2 (in both best soil condition and sowing date) to 450 seeds/m2 (in both worst soil condition and sowing date). So, according to de Snoo and Luttik (2004)‘s results, between 14 and 41 seeds/m2 would remain on the soil surface. These estimations are consistent with our own results. We estimated that an average of 8 to 96 seeds/m2 remained on the soil surface on the headland of 15 winter wheat and barley fields in autumn 2013 (unpublished data). Winter wheat seeds are treated with 0.7 g of imidacloprid per kilogramme of seeds (https://ephy.anses.fr/). With a thousand grain weight of wheat seeds ranging from 40 to 60 g, one seed is thus treated with 0.028–0.042 mg of imidacloprid. Consequently, for a grey partridge weighing approximately 390 g the median lethal dose of 13.9 mg/kg (Gibbons et al. 2015) is reached with the ingestion of between 129 and 194 wheat seeds. For a feral pigeon (C. livia) weighing approximately 300 g (http://app.bto.org/birdfacts/results/bob6650.htm), the median lethal dose of 25 mg/kg (Gibbons et al. 2015) is reached with the ingestion of between 179 and 268 seeds. So, according to de Snoo and Luttik (2004)‘s estimates, these amounts would be found on the soil surface of areas ranging from approximately 3 to 14 m2 for grey partridges and 4 to 19 m2 for pigeons. As a consequence, the amount of seeds that would remain on the soil surface in a routine use would be high enough to cause mortality.

However, according to the optimal foraging theory, areas with spillage (i.e. local high seed density) may be more attractive for birds than areas with only irregularly scattered seeds (de Leeuw et al. 1995). Nevertheless, Murton et al. (1963) found that only food densities <2 grains/m2 were too low for the wood pigeon to exploit successfully. Thus, even without spillage spots, densities of grains in cereal fields in autumn are sufficient to attract wood pigeons. On the contrary, Moorcroft et al. (2002) found that autumn grey partridges rarely feed on stubble fields where cereal grain density was <50 seeds/m2. For grey partridge, thus, fields with only scattered seeds may have only a low food value. Nevertheless, some other factors such as the density of other seeds in sown fields (e.g. weed seeds) or the availability of other fields with higher food values may affect the attractiveness of autumn cereal sowing. Furthermore, even when seeds are completely buried into the soil, bird exposure can still occur. For instance, skylarks bring seeds to the surface by uprooting seedlings (Green 1978).

Avoidance of Imidacloprid-treated seeds

Imidacloprid-treated seed avoidance, observed in captive birds, is due to a conditioned aversion mediated by sickness. This aversion occurs after a first experience of treated seeds ingestion (Avery et al. 1993; Lopez-Antia et al. 2014). The efficiency of this kind of learned avoidance requires that sublethal effects leading to rejection occurred well before a lethal dose is ingested. The amount of toxic needed to cause rejection (in relation to lethal dose) and the speed at which it occurs, as well as the feeding rate of birds are, thus, important parameters. Yet, various factors that may happen in the wild, as starvation (Pascual et al. 1999c), predation risk (Avery et al. 1994), availability and unpredictability of alternative food (Lopez-Antia et al. 2014; Murton and Visozo 1963), or competition (Mckay et al. 1999) may alter feeding behaviour to such an extent that lethal doses of pesticide could be ingested before avoidance occurs, or even overcome avoidance.

Furthermore, the amount of ingested imidacloprid-treated seeds can widely vary among individuals. As a result, even in “optimum” captivity conditions (i.e. availability of alternative food and without food shortage) the avoidance of imidacloprid did not prevent the occurrence of nervous disorders (Avery et al. 1993) or death (Lopez-Antia et al. 2014). Although the observed nervous disorders are transitory, in the field they could cause the death of wild birds for example by making them more vulnerable to predation or collision with vehicles, as well as favouring falling in flight. Such associated incidents could explain the haemorrhagic lesions observed in many birds. These secondary effects may also suggest a higher risk for birds in nature, in relation to indirect mortality.

Another study (Soyez 1998 in ANSES 2011) showed the variable nature of avoidance of imidacloprid-treated seeds that may be reduced when seeds are leached by rain or “aged” a few hours before being accessible for captive birds. This decrease of repellent effect could result from the dissipation of imidacloprid residues on seeds. Indeed, the concentration of imidacloprid in seeds affects the bird avoidance response (Avery et al. 1993, 1994).

The density of treated seeds at the soil surface may also influence the avoidance response of birds by modifying their rate of seed consumption. The intake rate of grains by birds increases with the density available on the soil surface (Baker et al. 2010; Murton et al. 1963). For example, the seed intake rate varies from about 4 peck/min at low seed densities (2 seeds/m2) to about 30 pecks/min at high densities (>150 seeds/m2) for grey partridges (Baker et al. 2010), and from about 30 seeds/min at low densities (<20 seeds/m2) to about 60 seeds/min at high densities (200 seeds/m2) for wood pigeons (Murton et al. 1963). Thus, in a high seed density situation a lethal amount of treated seeds could be ingested before post-ingestional distress happen. As a result, spots of spilled seeds pose a very high risk for granivorous birds.

Variability in risk factors, incident detection and reporting

The majority of incidents were reported in autumn. Grey partridge and pigeons—especially feral pigeon and to a lesser extent wood pigeon—are the major bird species reported in incidents. These results may reflect specific features in the probability of incidents being detected and reported but also in risk factors.

Detection and reporting

Carcass density and morphology (i.e. size and colour), ground-vegetation composition and structure, as well as the level of human activity where mortalities occurred, affect the probability of carcasses being found (Vyas 1999). In addition, once a dead animal is found, other factors such as public awareness of registration scheme, or the affective value of the species found, can also affect incident reporting. For example, the SAGIR mainly relies on hunters. Hunters (and their dogs) are more likely to find wildlife carcasses than other people as jogger or walker. They are generally well informed of the existence of the reporting scheme but their focus is often limited to game species (Berny 2007).

Thus, given these sources of variability, autumn incidents may have been more detected than spring incidents. In autumn, after crop harvest, cereal-growing areas are mainly composed of stubble, bare ground and recently sown crops while in spring, the vegetation is actively growing. In a farmland, small game hunting period is mainly from mid/end September to November. In addition, more individuals were involved in autumn incident. Similarly, grey partridge and pigeon carcasses may have been more detected and reported. They are relatively large game birds compared to other nongame farmland birds such as small passerines. Pigeon incidents may have been even more detected than partridge ones since they involved more individuals and they were detected more frequently close to human activity area (e.g. village, barn, farm). In addition, smaller birds are scavenged more quickly and at a higher proportion than larger birds (Ponce et al. 2010). Thus, the fact that no imidacloprid incidents involving smaller farmland birds were reported in the SAGIR does not mean that they do not occur in the field. For instance, Emberizidae, Fringillidae, Passeridae and Paridae species represent only about 5% of the approximately 15,000 bird data reported in the SAGIR database. Fringillidae mortality was specified in one incident also involving pigeons. Unfortunately, no post-mortem examination and residue analyses could be done on carcasses of this species.

Is the exposure to treated seeds higher for some crops?

The majority of incidents were related to winter cereals. Yet, with a mean of 0.035 mg of imidacloprid by seed, cereals are not the most hazardous seeds compared to beet or maize seeds that are treated with 0.9 mg and ~1 mg of active ingredient per seed, respectively (Goulson 2013). So, some factors may have increased the bird exposure to imidacloprid-treated cereal seeds sown in autumn.

First, in France approximately 6.5 millions of hectares of winter cereals are cultivated compared to 2.5 millions of hectares of spring crops (spring cereals, beets, maize, and sunflower) for which imidacloprid is/was used as seed treatment. Besides, successive use restrictions of imidacloprid as seed treatments of different spring crops have occurred since 1995 (sunflower in 1999, maize in 2004, spring cereals in 2014). Thus, although we do not know the yearly share of each crop with imidacloprid seed dressing, the acreage of field sown with imidacloprid-treated seeds was certainly much larger in autumn than in spring.

Second, the proportion of seeds remaining on the soil surface is higher for cereals sown in autumn than in spring, probably due to the unfavourable soil conditions in autumn, and for crops sown with standard drill (as cereals) than crops sown with precision drill (as sugar beet, maize and, sunflower; de Snoo and Luttik 2004). For instance, de Snoo and Luttik (2004) found an average of 0.03 seeds/m2 on precision-drilled crops (maize, onion, and sugar beet), that is far below the 2 grains/m2 for wood pigeons to exploit these fields. As a result, precision-drilled crop fields may be not attractive to birds due to too low densities of surface grain. Birds are thus probably more exposed to treated seeds of autumn sown cereals than to seeds of spring sown crops.

In addition, other factors as bird preference for some seeds may also influence the degree of exposure to treated seed according to crop. For example, pelleted sugar beet seeds are poorly attractive (Prosser and Hart 2005).

Variability of species sensitivity

Grey partridges and pigeons are particularly sensitive to imidacloprid. Indeed, according to the USEPA classification, imidacloprid is highly acutely toxic for both grey partridge and feral pigeon while it is moderately toxic for Mallard (Gibbons et al. 2015). Feeding habits of these species may also increase their exposure to imidacloprid-treated seeds. Pigeons feed on crop sowings, especially cereals (Inglis et al. 1990; M’Kay et al. 1999; Murton and Westwood 1966; Murton et al. 1963), all the more when other preferred food sites, as stubble, are scarce (Inglis et al. 1990; Murton and Vizoso 1963). In autumn, cereal grains represent about 50% of the diet of the grey partridge (Birkan and Jacob 1988) and partridges prefer to forage in field edges. Thus, they will probably be exposed to a higher number of imidacloprid-treated seeds due to the higher number of surface seeds on the headland. Furthermore, the grey partridge and the pigeons store food in their crop from the mid/end of day for digestion during the night (Murton et al. 1963; Nikiforov 1992; Rashotte et al. 1997), but, to the best of our knowledge, whether and how this feeding behaviour could affect the avoidance response has never been studied.

On the other side, smaller sized birds are more exposed to pesticides due to higher daily energy expenditure. As a result, the risk of imidacloprid-treated seed poisoning could be greater for them. Mineau and Palmer (2013) estimated that the ingestion of less than four imidacloprid-treated wheat seeds would have a 50% probability of killing a bird weighing 15 g. Given the number of imidacloprid incidents involving grey partridge and pigeon reported here, and the number of wheat seeds sufficient to reach the LD50 for both the bird species, we can claim that this toxic amount of seeds (for small birds) is commonly and largely available at the soil surface of wheat sown fields. This is also supported by our estimation of surface seeds in fields. However, it is acknowledged that some birds dehusk seeds and that this behaviour is mainly observed in small species (body weight < 50 g) and chiefly in the specialized granivores (finches, sparrows and buntings) (Avery et al. 1997; Prosser and Hart 2005), Thus, the ingestion of imidacloprid together with the consumption of treated seeds may be reduced by this behaviour for small granivorous farmland birds.

Residue analysis and diagnosis of poisoning

Overall, in more than half the cases when residue analyses were performed on both crop/gizzard and liver the systemic absorption of imidacloprid was not confirmed since imidacloprid residues were not detected in the liver. This may reflect the occurrence of indirect mortalities (e.g. traumatic death due to a falling in flight, as can be suspected from the detection of haemorrhages in dead birds) caused by imidacloprid sublethal poisoning. This assumption is supported by the fact that we detected no imidacloprid residues in the liver of the nine individuals found moribund for which residue analyses were performed in both crop/gizzard content and the liver.

However, Lopez-Antia et al. (2015) found mean concentrations of imidacloprid of 55.3 μg/g and 82.6 ng/g respectively in crop and the liver of 19 dead red-legged partridges exclusively exposed to wheat seeds treated at recommended application rate (0.7 mg of imidacloprid/g of seeds) for 25 days. Their crop concentrations are similar to ours while their liver concentrations are just below our detection limit in the liver (i.e. 100 ng/g). These results suggest that our nondetection of imidacloprid in the liver of some individuals could be sometimes due to a lack of sensitivity in our analytical method rather than an actual absence of imidacloprid residues. Their findings also indicate that mortalities can be associated with very low imidacloprid liver levels.

In very few cases, we observed a great variability in imidacloprid concentration (crop/gizzard and liver) among individuals of the same clustered incident (see Online Resource). This could be explained by the great individual variability of treated seed consumption, observed in avoidance studies. Furthermore, we found no relation between crop/gizzard concentration and liver concentration. Factor as regurgitation might have been affected the results of crop/gizzard content analyses, but does not necessarily prevent mortality (Pascual et al. 1999d). Further works are required to better understand the relation and variation in both crop/gizzard and liver imidacloprid concentration and, thus, to improve the diagnosis of imidacloprid poisoning.

Impact of imidacloprid poisoning on bird populations

How the imidacloprid-related mortalities affect farmland bird populations is not known yet. The impact at the population scale is likely to depend upon the status of the population and the timing of mortality.

First, the rate of mortality attributable to imidacloprid poisoning in different bird populations is still unknown. Second, compensatory mortality or natality (see for instance Boyce et al. 1999) may mitigate the effects of these mortalities on populations. Density-dependent overwinter survival and/or natality were found, for example, in grey partridge (e.g, Bro et al. 2003; Panek 1997; Rotella et al. 1996) and pigeons (e.g. Hetmański and Barkowska 2007; Kautz and Malecki 1990; Murton et al. 1974). Seasonal timing in these anthropogenic mortalities and density dependence are important factors determining the nature of the demographic response (Boyce et al. 1999; Kokko 2001). Anthropogenic mortalities are more likely to be compensated for when mortalities are time-limited and happen before a seasonal density-dependent mechanism. Thus, with respect to density-dependent overwinter survival, autumn imidacloprid incidents are more likely to be compensated for than spring incidents. In addition, spring incidents occur principally in the late winter/early spring, at the early beginning of reproduction season of the majority of farmland bird species. Additive mortality is indeed more likely at this stage of the year, and spring pesticide exposure may potentially impact the reproductive success.

The bird population status is another relevant factor that underlies the demographic response to these anthropogenic mortalities. Mortality is more likely to be additive when populations are in decline or at low-density (e.g. Bartmann et al. 1992). In France, short-term (2001–2012) breeding population trends are considered to be fluctuating for both grey partridge and stock dove, increasing for wood pigeon and unknown for feral/rock pigeon (Comolet-Tirman et al. 2015). Thus, these imidacloprid casualties may have limited effects on at least wood pigeon, grey partridge and stock dove global population. However, many other farmland bird populations still decline in France (Comolet-Tirman et al. 2015) and Europe (EBCC 2015). Thus, even though imidacloprid-related casualties are probably not the primary cause of this continuing decline, under some circumstances, they could be an aggravating factor.

Moreover, imidacloprid poisoning of juveniles of some bird species may be a possibility, depending upon the timing of hatching and sowing, and the diet of chicks (while the fledglings of many granivorous birds are mostly insectivore especially in the first weeks of their life). If it occurred, this type of incidents would go totally undetected in the SAGIR Network.

[END]
---
[1] Url: https://link.springer.com/article/10.1007/s11356-016-8272-y

Published and (C) by Common Dreams
Content appears here under this condition or license: Creative Commons CC BY-NC-ND 3.0..

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/commondreams/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.