1. Introduction
When the Krefeld Entomological Society, together with colleagues, published their famous long-term study (the “Krefeld study“) in 2017, it became known that the biomass of flying insects had declined by approx. 75% over the past three decades in over 60 nature protection areas of Germany [
1]. Another long-term study (the “Munich study“) confirmed the Krefeld results, providing evidence that strong abundance declines of insect populations had occurred in farmland and forests across vast areas of Germany and Switzerland [
2]. Similar abundance declines of insect populations have been documented in other long-term studies of other European and North American regions [
3,
4,
5,
6,
7,
8,
9]. It is generally accepted that we now live in an era of global insect meltdown [
9,
10,
11,
12,
13]. This has dramatic ecological implications: the fact that insects comprise the basis of many food chains and food webs [
14,
15] means that in a world without insects, countless insectivorous species will ultimately become extinct due to starvation [
16].
In this context, aerial web-spinning spiders are a group of utmost interest. These spiders (including the orb-weaving species in the families Araneidae) trap flying insects with the aid of aerial webs [
17,
18,
19]. In the temperate climatic zone, aerial web-spinning spiders feed for the most part on small dipterans, aphids, and hymenopterans (
Figure 1), which are exactly the types of insects known to have most dramatically decreased in abundance and biomass in recent decades [
1]. Aerial web-spinning spiders, because of their reliance upon flying prey, are therefore a highly vulnerable predator group in regions characterized by a high loss of flying insects, such as Germany and Switzerland.
While a larger number of studies on population declines of various groups of insects has been conducted in recent years (see references in [
12]), potential cascading effects have only been documented so far for insectivorous birds [
6]. No studies of this type have, to date, considered potential declines due to trophic cascades among arthropods, like aerial web-spinning spiders. Here we used the European garden spider
Araneus diadematus as a model system to address whether evidence for population declines in aerial web-spinning spiders in the Swiss midland can be found (
Figure 2). The female, 10–18 mm in length, reaches adulthood in late summer, at which time it spins (≈30 cm diameter) orbs [
20,
21,
22]. The webs are built 0.5–2 m off the ground [
21]. This animal, with its conspicuous white, cross-shaped mark on the upper side of the abdomen, is one of the best known spider species in western Europe.
By comparing historical abundance data (20th century) with present-day data (2019), we examined whether the abundance of
Araneus diadematus has changed over the past few decades. This work is based on extensive experience gathered during graduate research in the Swiss midland in the 20th century [
20,
21] and supplemented by a present-day population density survey in the same geographical area. To make the comparison more robust, the data base was expanded by including published population density values extracted from the scientific literature.
4. Discussion
This study revealed that the large orb-weaving spider
Araneus diadematus occurred in extraordinarily low population densities in the Swiss midland in 2019 (
Table 1).
Araneus diadematus spiders take down their webs during the night and rebuild them early in the morning. However, as laboratory experiments and field observations have revealed, not all individuals rebuild their webs the following day [
30,
36]. Some spiders, after consuming exceptionally large amounts of food at particularly favorable hunting sites, may cease feeding and rebuilding their webs for one or more days [
36]. Thus, the percentage of spiders found with webs within an
A. diadematus population is, in general, lower than 100% and usually in the range of 66% to 88%, depending on the availability of food at a particular time and location [
30,
36]. The question arises of whether the method of web counting applied in this study might have resulted in underestimation of the true population densities by overlooking a certain percentage of satiated spiders which had not rebuilt their webs at the time of the surveys. On this issue, it may be noted that densities assessed in the past century were for the most part also based on web counts [
18,
20,
21,
28,
29,
30,
31,
32,
33,
35], so the same methodological bias was involved at both time periods. Furthermore, since unsatiated spiders tend to rebuild their webs more frequently than fully satiated spiders [
36,
37], and since present-day spiders living in the era of insect loss are more likely to be unsatiated compared with half a century ago (
Table S2), the frequency of web building nowadays might be expected to be rather higher. Thus, the possibility that the extraordinarily low present-day population densities were underestimations due to this particular type of methodological bias is highly unlikely.
In 2019, the webs appeared to be rather fragile compared with the stronger webs from previous decades, as is the case when malnourished spiders make webs with thinner threads (Nyffeler, pers. observations). This is in good agreement with laboratory experiments, in which the amount of thread produced is reduced if
Araneus diadematus spiders are kept deprived of food [
36] and it agrees with our observation that the webs contained significantly fewer prey compared to previous studies (
Table S2). The overall impression gained during this survey was that there has been a paucity of prey throughout the entire study area in recent decades (
Table S2), which has been confirmed by the observations of other researchers [
2,
38,
39].
The abundance decline we show here of this once common spider in the Swiss midland is evidently revealing a bottom-up trophic cascade in response to prey scarcity recently documented in the same area, and more widely across western Europe [
1,
2,
38,
39]. The hypothesis that the availability of flying insect prey in the study area had drastically declined over the past decades is confirmed by the “windshield phenomenon” noticed throughout the Swiss midland (i.e., compared with previous decades, many times fewer flying insects are nowadays killed on the front windshields of cars [
9,
40]). This is in sharp contrast to the situation a few decades ago, when fairly frequent “wasteful killing“ (and, coupled with it, “partial consumption”) of insects in
Araneus diadematus webs at favorable sites could be observed (with capture rates of sometimes up to 1000 prey web
−1 day
−1; [
41]). The reduced food intake of
Araneus diadematus in recent years (
Table S2) is likely to have negatively impacted the fecundity and survival of this spider [
42,
43], which in turn may have led to the abundance decline documented in this study (
Table 1). Sublethal effects of chemical pollution may have additional negative impacts on this spider’s survival [
44,
45,
46,
47,
48,
49].
Aside from food scarcity due to insect population declines and chemical pollution, other negative environmental developments have additionally contributed to the decreasing population densities of large orb-weaving spider species in the western European landscape. These include adverse effects of modern forestry practices (e.g., removal of forest understory and of shrubs at forest margins, large-scale aerial spraying for the purpose of caterpillar or bark beetle control), the transformation of traditional farmland into large-scale depauperate monocultures (accompanied by the loss of weedy field margins and hedges), the transformation of fallow grassland patches into construction sites, and the removal of shrubs and trees in urban and suburban gardens. That habitat conversion and degradation can have a strong detrimental effect on large orb-weaving spider abundance is well demonstrated by the example of the fate of a plot of land located near Zurich-Oberengstringen (in between Highway A3 and the Limmat riverbank). In the 1970s, tall grassland interspersed with shrubs covered this plot, and at that time, large orb-weaving spiders occurred there in high numbers (up to 6 m
−2 in small, local patches [
50]). During a visit to this site, 40 years later, it became apparent that the land had been converted to a highway rest area with short-cut lawns and, as a result, the density of large orb-weaving spiders had declined to a very low value of 0.002 m
−2 (Nyffeler, pers. observations). Similar habitat conversions resulting in the destruction of suitable habitats for large orb-weaving spiders have been noted elsewhere.
We of course realize that the recorded declines were not based on systematic long-term monitoring and that they could be attributed to normal population cycling. However, while we are not here able to provide a smoking gun, several elements support a systematic decline. First of all, the dramatic decline up to < 1% of the reference baseline from 40 years ago has never been observed at shorter temporal time frames, and reaches far beyond the typical natural fluctuations in population sizes (e.g., [
51]), even when they are known to be linked to resource pulses (e.g., [
52]). Second, records from other locations show that this decline is occurring at such large spatial scales that it can only be explained by large-scale, more global environmental drivers like land-use change, climate change, or the global use of pollutants. Indeed, the extraordinarily low population densities of
Araneus diadematus in the Swiss midland observed in the 2019 survey (
Table 1) are supported by statements of biology students at the University of Basel, according to whom generally very few large orb-weaving spiders have been spotted in recent years during walks through forests and fields in the Basel region. Furthermore, in mid-June 2017, a group of over 40 biology experts conducted an extensive faunistic survey in the grounds of the Merian Gardens, a park-like area covering 180,000 m
2 located on the outskirts of Basel [
53]. In the course of this survey, only three specimens of
A. diadematus could be found over a time period of 24 h suggesting that nowadays this once “abundant garden spider” [
54] must have become rare in that area. The only large orb-weaver species still found in high densities during the 2019 survey in the Swiss midland was the bridge spider
Larinioides sclopetarius (Nyffeler, pers. observations). This nocturnal species, which lives near water and frequently builds its webs on street lights and illuminated bridge railings, still gets sufficient amounts of food in the form of flying adults of semiaquatic insects (chironomids, ephemeropterans, trichopterans, etc.) attracted in large numbers to the artificial light [
55,
56,
57]. Because of its capability to exploit these artificially high prey densities,
L. sclopetarius is—in contrast to other large orb-weavers—extremely successful in colonizing urban habitats in high density even in the 21st century, not only in Switzerland, but also in other parts of western Europe [
56].
Similar extraordinarily low population densities of
Araneus diadematus (0.0004 adult females m
−2 at a landscape scale) were also recorded in an extensive survey in nine landscapes of the Ghent region, northern Belgium, in the summer of 2014 [
49]. Here, likewise, densities were one order of magnitude or more lower than those recorded one decade earlier (e.g., during a survey in 2004–2006, densities at two locations were 10–20 times higher compared to those ten years later [
58]; Bonte, pers. observations). Interestingly, while other orb-weaver species—especially the larger ones—showed dramatic declines in both abundance and species richness along an urbanization gradient [
59,
60],
A. diadematus was found to reach similar low densities across this land-use gradient. In other words, local densities at spatial scales of approx. 100 km were alarmingly low in both the more natural and human-impacted regions, demonstrating regional declines beyond the scale of local land use. Thus, extraordinarily low present-day population densities of
Araneus diadematus have been recorded in two different regions of western Europe, ≈500 km apart. These declines extend beyond local environmental changes at small scales and suggest common negative impacts of intensive urbanization, climate change, or any other large-scale stressor across the entire landscape over the past half century [
61,
62].
The present study and that from northern Belgium suggest that the notion that
Araneus diadematus is an abundant spider, as pointed out in much of the literature [
63,
64], has these days turned out to be a myth—at least in highly urbanized western European landscapes such as the Swiss midland and northern Belgium. So far, the apparent abundance decline of this species in western Europe has been ignored by the faunists in charge of compiling local Red Lists for spiders (e.g., [
63,
64]). Red Lists are based on the number of different locations within a landscape in which a species has been recorded, and not on absolute population densities. The fact that
Araneus diadematus is still found in many places (although in highly reduced densities) may explain why it is still labeled an “abundant species” (e.g., [
63,
64]).
In contrast, present-day population densities of
Araneus diadematus in some areas outside of continental western Europe still appear to be rather high. For instance, exterminators of the pest control firm Senske Services removed more than 75
Araneus diadematus spiders from the exterior of a “2500 square foot home“ (232.5 m
2) in Seattle (USA) in August 2016 [
65]. Evidently, the spiders at this particular location were still capturing sufficient food in the form of flying insects to enable them to build up a population density roughly 290 times that of the present-day Swiss overall mean density. Future detailed assessments of the population densities of this spider species in different parts of its geographic range are urgently needed, and may provide important information on the extent of the loss of flying insects in various regions of western Europe and further afield. We would in this respect like to promote the use of new citizen-science tools such as SpiderSpotter [
66] to achieve these highly needed insights, and especially to guide short-term action to mitigate alarming declines in these sentinel arthropod top predators.