I have had very little success trapping bees using orange traps with almost no UV and plateauing at about 600nm, but I was trapping mostly for flies. A study on Hopliini beetles by Colville et al. (Species turnover of monkey beetles (Scarabaeidae: Hopliini) along environmental and disturbance gradients in the Namaqualand region of the succulent Karoo, South Africa. Biodiversity and Conservation 11: 243–264, 2002)) found that beetles were mostly attracted to either short wavelength or long wavelength pan traps.

Im not sure about bees, but Chittka et al. (Chance and adaptation in the evolution of island bumblebee behaviour. Popul Ecol 2004. 46:243–251) documented colour preference to vary with location and bee species. From their results, bee blue and bee UV-blue seem to be most attractive overall, at least for bumblebees.

I have data from a two large colour trapping experiments with dozens of colours, where we also recorded bees, separately for various taxa. Spectral reflectance was always measured. The work is not published yet but I am happy to share the data privately for the time being. However, in comparison with other insects, pan trapping does not really work with bees.

With regard to UV-bright: Many paints including those of commercial yellow pan traps have a low reflectance in the UV, therefore it seems that authors writing "UV-bright" just emphasise their traps are different in having a high reflectance across 300-400 nm.

Do a Google search for "sam droege pan trapping" and look at some of the links to resources he has made available about pan trapping.  He also has a source for pre-painted plastic bowls with standardized colors.

Dear Klaus, we have a data for this so called "uv-bright" pan trap we can share and talk along with Adrian.

Mani

Klaus Lunau, you can get information about pan trapping of bees and the most attractive colours from literature and some Authors who have worked on such and then carry out preliminary study and if possible further study on it to compare observations and results. The observations and results could be determined by species diversity, location and time of the day. What ever the observation and result lets have a share it by publication.      

Thanks to all the answerers of the pan trapping question. I think that "UV-bright" might refer to the colour's fluorescence under ultraviolet light. If this is true, it means that the colours might reflect more than 100% in the visible range of wavelength. This was found by own research in which we checked the spectral reflectance of Spavar Leuchtfarben. Besides reflectance >100% in one range of wavelength, we also found that these fluorescent colours absorb ultraviolet light and trap more bees than similarly coloured non-fluorescent colours. The answers helped me to understand how misleading information about the attractiveness of UV-bright and UV-reflective colours for bees might have developed.

Hi Klaus!

I collected data for my master's thesis with pan-traps and hand-netting, and based on the results solitary bees do fly to pan-traps but bumblebees don't. I used pink, white, yellow and blue traps and yellow had the best catch rate when looking at species and individual numbers. Yellow was also the only trap differing in its reflectance spectra from the others. Blue, pink and white were blue to bees, while the yellow plate was green (information about honeybees photoreceptors used). However, hand-netting was still yielding the most individuals, although species numbers were about the same for the yellow pan-trap and hand-netting. But even though the species numbers were about the same for these two methods, the species themselves were not the same; bumblebees were caught with hand-net and solitary bees with pan-traps.Unfortunately these results are not published yet.

Interestingly we have used the same pan traps painted with UV-reflecting paint in different habitats/regions and they systematically failed to attract wild bees in tropical regions (DRCongo, Singapore), whereas the same pan traps attact hundreds or thousands of specimens in temperate areas (France, Belgium, UK, etc.). This phenomenon can not be explained only by differences in major groups of bees since some genera occur both in temperate and tropical areas (Megachile, Coelioxys, Thyreus, Amegilla, etc.). There is still a lot to learn from these and other experiments and although we are still using the same pan traps everywhere to get our setup standardised across regions and habitats, I would not mind changing colours or protocol if the science suggests that it might be worth it. I look forward to reading your thoughts, Klaus!

Dear Nico, I have no explanation for your finding that pan traps in tropical reasons are far less effective for bees. As I have no experience with pan trapping of bees in the tropics I can only speculate that the height of exposure or other factors are more important,

Just one comment on your wording "UV-reflecting paint": at least white UV-reflecting paints might be very inattractive for bees, since this colour provide little colour contrast against the background and of low bee-subjective spectral purity.

Dear Klaus,

Thank you for your reply. Check out the attached Venn Diagram with the species counts per sampling technique — you will see that the white pan traps are quite attractive actually. There's still a lot to discover!

Dear all,

Thank you very much for all your opinions.

I was wondering what would be the difference, in terms of light reflectance spectrum, between flourescent colors and UV-bright colors in the context of pan-trapping. I think it is a methodologicaly interesting question, as in America I found no studies using UV-bright colors (they use mostly fluorescent colors), but in Europe they use mostly UV-bright colors. This makes difficult standardization/comparing results, quantitatively and qualitatively.

I have read a little bit about physics on it, and I found out that UV-bright colors have a high reflectance across UV (300-400 nm), but fluorescent colors absorb UV energy but it reflect it as visible, so they have high brigthness in the visible spectrum, but low or no UV reflectance. Am I right?

Does anybody know if there exist any study comparing UV-bright and fluorescent colors in terms of light reflectance spectrum and/or in terms of pan-trapping results (effectiveness in abundance or richness of bee captures)?

Thank you very much in advance!

Sergio

Dear all,

my impression is that UV-bright colours and fluorescent colours are synonymous. Some spectrometers are able to register fluorescence, other are not; that might have caused confusion in the naming of colours. I would like to get a clear definition of UV-bright, UV-reflective and fluorescent pantrap colours and information about the first mentioning of the attributes in the pan trap literature.

Klaus

The first mention of pan traps I know of in the literature is:

Ssymank, A. (1991). "Die funktionale Bedeutung des Vegetationsmosaiks eines Waldgebietes der Schwarzwaldvorbergzone für blütenbesuchende Insekten - untersucht am Beispiel der Schwebfliegen (Diptera, Syrphidae)." Phytocoenologia 19(3): 307-390.

The functional importance of the vegetation of three forest areas in Southwest-Germany for flower flies (Diptera, Syrphidae), as a major group of pollinating insects has been investigated. Both mesophilic and acidophilic beech forests and hygrophilous ash-forests with their homologous vegetation of clear cuttings and the accompanying seamcommunities on the verges of the forest roads are phenologically and biocoenologically characterized. Syrphid flies have been caught with coloured water dishes, their flower preferences and diurnal activity patterns have been studied to reveal habitat preferences and to analyse the community structure of flower visiting insects and its determining ecological factors in relation to a mosaique of well-defined vegetation units.

The first study I know that considered pan color preferences by bees is:

Leong, J. M. and R. W. Thorp (1999). "Colour-coded sampling: the pan trap colour preferences of oligolectic and nonoligolectic bees associated with a vernal pool plant." Ecological Entomology 24: 329-335.

1. Pan traps or water traps have been used widely to sample agricultural insect pests, but no formal studies have assessed the utility of these traps as sampling devices for bees.

2. Coloured pan traps, used as flower models, can efficiently and selectively sample an oligolectic bee, Andrena (Hesperandrena) limnanthis, and other bees associated with white‐flowered Limnanthes douglasii rosea.

3. Females and males of A. limnanthis unexpectedly exhibit different colour preferences. Females are strongly attracted to white and blue traps, but discriminate against yellow traps. Males prefer white traps over blue and yellow traps. Consequently, blue traps are selective for females only, while white traps are selective for both sexes.

4. Non‐A. limnanthis bees were caught in significantly greater numbers in yellow than in blue or white traps. These bees included generalists, as well as specialists that are oligolectic on mostly yellow‐flowered species.

5. Colour of traps had a significant effect on the numbers of A. limnanthis females and males, and non‐A. limnanthis bees caught in traps. These results indicate that quantitative sampling of bees by pan trap methods can be highly sensitive to trap colour.

Dear all,

I just found a link to a similar discussion which might be intersting to some of you followers:

https://www.researchgate.net/post/Should_pan_traps_for_insect_sampling_be_painted_with_UV_paint

Here, Lynn V Dicks, asked:

Should pan traps for insect sampling be painted with UV paint?

Are pan traps more effective if coated with ultraviolet reflective paint? If so, what brands do people use or recommend? We have found these:http://www.uvgear.co.uk/product/product165.htmhttps://translate.google.co.uk/translate?hl=en&sl=de&u=http://www.spraycolor.de/&prev=search Any advice or thoughts welcome.

I think the term UV-paint communicates confusiuon about the corresponding colour at least for non-native speakers like me.

Klaus

Klaus, thank you very much for sending this discussion.

Dear Sergio,

Here is a link of our recent paper. Be cautious while using fluorescent pan traps or UV bright.

Insects 2019, 10, 40; doi:10.3390/insects10020040

UV bright colour or florescent pan traps affect the insect capture so it will be biased. We only presented order level categories but it also affect genus and species level in our experience/ experiment.

Does this help?

REgards

Thank you very much Mani!

You defined (in your paper) "fluorescent" colors, those that showed >90% of reflectance.

I would think that "fluorescent" should be differentiated from "UV-bright", as I understand that "UV-bright" define colors which have some reflectance in the UV spectra (200-400 nm). Do you agree with this "UV-bright" definition? (I thought that this kind of UV-bright colors are used trying to take advantage of the fact that bees are capable to see UV-light, and highly attracted by UV-patterns of flowers).

In this sense, some of your results (in your Figure 2B) agree wiht such a differentiation (but other not, or it's confusing):

-your yellow "UV" fluorescent paint, although little, show a reflectance throught 300-400nm spectrum, but your yellow (non-fluorescent) did not show reflectance in UV spectrum (just a little bit almost at 380-400nm). In addition, yellow "UV" fluorescent has a peak with >90% of reflectance around 500 nm, therefore is considered "fluorescent". So, this paint is fluorescent (~500nm) and UV-bright (300-400 nm). Does it make sense to you?

-contrarily, in the blue color, your blue "UV fluorescent" paint did not show reflectance in UV spectrum (nor in visible spectrum), but, curiously, your blue (not UV fluorescent) paint did show some UV reflectance (in UV-A, between 370-400nm).

So I find the names "fluorescent" and "UV fluorescent" or "UV-bright" confusing.

-in the case of white paint, I have a doubt (I'm not sure if it is a confusion of mine or there is a mistake in the Table1 or in Figure 2B): you consider in your Table 1 that W3 and W4 are "fluorescent" (>90% reflectance, following your definition), and W5 "non-fluorescent" (90%, and W4 have reflectance

Dear all,

thanks for interesting discussion.

I have the same doubts as Sergio.

Regards,

Kátia

Our 3-year sampling (https://academic.oup.com/ee/advance-article-abstract/doi/10.1093/ee/nvz056/5506571 ) using fluorescent pan trap colors (blue, white, and yellow) showed no differences among colors in bee abundance, diversity, and community composition.

Dear Subodh,

Thank you very much for adding your research to this discussion!

I would be very grateful if you could clarify to me some a couple of questions:

1. What kind of fluorescent colors did yoy use? Commercial ones that call to itselves as "fluorescent"? Or did you make your own mix with some flouresecent colors (in the last case, which ones did you use?)

2.Do you refer to "fluorescent" in the sense of Mani Shrestha & col. (see above in this discussion) as colors that showed >90% of reflectance in visible spectrum? Or in the sense that they have some reflectance in the UV spectrum (200-400 nm)?

Thank you very much again!

Regards!

Sergio

Hi Sergio,

You have a very good question. 1. You are right, I am referring "fluorescent" to the commercial ones (I didn't make a mix). 2. I think my answer for this will be based on my first point, that I can't be sure about the reflectance of my pan traps (since I didn't measure it). I would follow Mani Shrestha & col's paper on this matter.

All the best to your research.

Dear Subodh,

I think measuring fluorescence could be a problem, since some sprectophotometers give one wavelength on to probe and measure only the light of the same wavelength reflected from the probe. This type of spectrometer will not register fluorescence.

Klaus

Dear Subodh, Thank you very much for your reply!

Dear All, Let me make a little summary of all information in this discussion:

Following wavelength plots in papers referenced by Klaus Lunau & cols. and Mani Shrestha & cols., it seems that all paints whose reflectance spectra have been studied, called "fluorescent" or "luminous" or "UV-bright (Sparvar) have a high luminosity (>90% or >100% reflectance), but very little or no reflectance in UV spectrum (200-400 nm). Curiosly, in the Mani Shrestha & cols. study, only the white bowl without painting, and normal blue and white paint (not fluorescent) have some UV reflection in the UV-A (around 400 nm).

In this sense, I agree with Klaus' statement that "UV-bright" colours and "fluorescent" colours seem to be synonymous. If it is so, "UV-bright" should be understood as pigments that absorb UV radiation but reflect it in the visible spectrum, thus increasing total reflectance above 100%.

Finally, we have the term "UV-reflective" that I would set aside for those colors (or materials) that are able to emit (or reflect) in the UV spectrum (200-400 nm). But this is my interpretation, not a statement.

Does it make sense to you all?

Considering all of this, I conclude that this kind of "fluorescent/UV-bright" colors are atractive to bees and other insects NOT due to UV reflection/emission (thus, NOT imitating to flower UV-clues, which is what I thought), but due to high reflectivity or luminosity (>100%). Am I right?

And just a final question: Are anyboty aware of some publication in which is shown a color/material that have a clear and specific emission/reflectance in the UV spectrum?

Sorry for extending so much and than you very much to all!

Sergio

Dear Sergio,

there was a recent publication of a group from Japan which used chorogenic acid as a fluorescent substance which is found in plants and can be used to add fluorescence to artificial flowers.

Klaus

Interesting!

Thank you very much Klaus!

Regards!

Sergio

According to my observation on solitary bees, they attracted more to yellow colour flowers. Maybe u can use this point for your research

Hi, this article maybe Useful

Thank you very much Dr. Sandeep!

You may want to look at:

Preprint A New Model of “Color” Vision for Remote Sensing

Thank you very much Jack!

In a study we used pan traps of capturing bees and found out that yellow and white are more attractive to bees than blue and grey (used as control). Herewith I am sending a file which may be useful to you.

Thank you very much Arati!

The initial question about pan trap colour is getting out of focus. My may concern is that we cannot know the attacting parameter of coloured pan traps as far as we do not consider colour hue, colour saturation AND colour brightness of the pan trap colours.

I completely agree with you Klaus.

The main question is why are certain colours pan traps effective: because their color (i.e. yellow...), because their UV-bright (if they have), because their brightness or luminosity (>100%), or becasue a combination of them all.

And a second question is about the confusing terms that we can find in the literature or in commercial brands, as: "fluorescent" or "luminous" or "UV-bright". Are they synonymous? If not, what are they mean?

Let me ans

Sorry, I was too quick on the button. forget my previous answer. In the paper attached we compare the catch of normal vs fluorescent colours. Fluorescent colours are more efficient than normal ones.

The term "UV-bright" is misleading in my opinion, since these colours absorb UV-light. "Fluorescent" would be a correct term for these colours that absorb light of shorter wavelengths (mostly UV) and reflect it as light of longer wavelengths. The terms "UV-bright" and "luminous" might decribe the property of the colours under UV-light, for example if illuminated with a UVtorch.

Klaus

What might a bee see? Are there any observations of the spectral sensitivities of a bee’s sensors?

Bees see ultraviolet, blue and green, but are insensitive to red light. Some bees exhibit preferences for distinct colour hues, many bees prefer (bee-subjective) saturated colours over less saturated colours. Thus UV-absorbing red (bee-black) and UV-reflecting white (bee-white) are not attractive due to low colour saturation.

It seems that bees are specially attracted by UV-yellow, lilac-pink and purple flowers. You may be interested in taking a look at the following work of our lab:

Article Pollinators show flower colour preferences but flowers with ...

Thanks Sergio Osorio Cañadas

You may be interested in this condensed version of “A New Model of Color Vision for Remote Sensing.”

https://www.researchgate.net/publication/339445986_A_New_Model_of_Color_Vision_for_Remote_Sensing_A_Condensed_Version_1

Thank you very much Jack!

Article I see your false colours: How artificial stimuli appear to d...

Dear Jack,

I agree that is is not easy to find UV-reflective white stimuli seen as white by bees, birds and humans: There are some chemicals BaSO4, PbSO4 that I have used to build artificial flowers for bees or, alternatively, the white feathers of swan covered with a UV-transmittent foil (or UV-absorbing clear floil to make a white UV-absorbing stimulus): By the way, we had similar problems with red and found that the red tail feathers of the grey parrot refect UV.

Here comes the paper with these bird colour stimuli.

Klaus