- Email: [email protected]
Thermal emissivity of avian eggshells
Journal of Thermal Biology 57 (2016) 1–5
Contents lists available at ScienceDirect
Journal of Thermal Biology journal homepage: www.elsevier.com/locate/jtherbio
Thermal emissivity of avian eggshells Lars Olof Björn a,b,n, Sven-Axel Bengtson b, Shaoshan Li a,nn, Christoph Hecker c, Saleem Ullah d, Arne Roos e, Annica M. Nilsson e a Key Laboratory of Ecology and Environmental Science in Guangdong Higher Education, School of Life Science, South China Normal University, Guangzhou 510631, China b Lund University, Department of Biology, Sölvegatan 35, SE-22362 Lund, Sweden c University of Twente, Faculty of Geo-Information Science and Earth Observation (ITC), Department of Earth Systems Analysis, 7500AA Enschede, The Netherlands d Department of Space Sciences, Institute of Space Technology, P.O. Box 2750, Islamabad, Pakistan e Uppsala University, Dept of Engineering Sciences, Solid State Physics, Box 534, 751 21 Uppsala, Sweden
art ic l e i nf o
a b s t r a c t
Article history: Received 15 September 2015 Received in revised form 11 November 2015 Accepted 26 November 2015 Available online 4 January 2016
The hypothesis has been tested that evolution has resulted in lower thermal emissivity of eggs of birds breeding openly in cold climates than of eggs of birds that nest under protective covering or in warmer climates. Directional thermal emissivity has been estimated from directional–hemispherical reflectance spectra. Due to several methodological difficulties the absolute emissivity is not accurately determined, but differences between species are obvious. Most notably, small waders of the genus Calidris, breeding in cold climates on the tundra, and in most cases with uniparental nest attendance, have low directional emissivity of their eggshells, about 0.92 when integration is carried out for wavelengths up to 16 μm. Species belonging to Galloanserinae have the highest directional emissivity, about 0.96, of their eggs. No differences due to climate or breeding conditions were found within this group. Eggs of most other birds tested possess intermediate emissivity, but the values for Pica pica and Corvus corone cornix are as low as for Calidris. Large species-dependent differences in spectral reflectance were found at specific wavelengths. For instance, at 4.259 μm the directional–hemispherical reflectance for galliforms range from 0.05 to 0.09, while for Fratercula arctica and Fulmarus glacialis it is about 0.3. The reflection peaks at 6.5 and 11.3 μm due to calcite are differentially attenuated in different species. In conclusion, the hypothesis that evolution has resulted in lower thermal emissivity of bird eggs being exposed in cold climates is not supported by our results. The emissivity is not clearly related to nesting habits or climate, and it is unlikely that the small differences observed are ecologically important. The spectral differences between eggs that nevertheless exist should be taken into account when using infrared thermometers for estimating the surface temperature of avian eggs. & 2016 Published by Elsevier Ltd.
Keywords: Avian eggs Birds Heat dissipation Egg cooling Thermal emissivity Incubation Seabirds Thermal radiation Waders
1. Introduction Most bird eggs are heated by the bodies of the parents during the development of the embryo, and keeping the eggs at a temperature much above ambient is important in cold climates. For various reasons parents may leave their nests unattended for shorter or longer periods, during which the eggs cool. In some bird species only one parent attends to the eggs, and has to leave the nest for feeding. The little stint (Calidris minuta) may leave its nest unattended for half an hour, during which the egg surface n Corresponding author at: Lund University, Department of Biology, Sölvegatan 35, SE-22362 Lund, Sweden. nn Corresponding author. E-mail addresses: [email protected] (L.O. Björn), [email protected] (S. Li).
http://dx.doi.org/10.1016/j.jtherbio.2015.11.008 0306-4565/& 2016 Published by Elsevier Ltd.
temperature can fall below 10 °C (Tulp and Schekkerman, 2006). During snowstorms eggs of sanderling (Calidris alba) may be abandoned for extended times (Reneerkens et al., 2011). In other cases the parents are forced to leave the nest due to predators. Thus common terns (Sterna hirundo) have been observed to leave nests for hours during night due to owls (Arnold et al., 2006). Even if embryos survive and eggs hatch, low incubation temperature has negative effects on hatchling development and survival (Olson et al., 2006; Hepp and Kennamer, 2012; Carter et al., 2014). Hepp and Kennamer (2012) found that the incubation time was increased from 30.1 to 37.9 days, i.e. by 26% when the temperature was lowered from 37.3 to 35.0 °C. Development in domestic fowl comes to a stop below 25 °C (Funk and Biellier, 1944) and hatching is seriously compromised already at a temperature four degrees below normal (Mortola, 2006). Clearly, it is of survival value for the embryo if the egg is constructed such that cooling takes place as
2
L.O. Björn et al. / Journal of Thermal Biology 57 (2016) 1–5
slowly as possible during these periods. Cooling takes place mainly by radiation and air convection, and although other factors, such as nest construction, probably are more important (Rockweit et al., 2012). Björn et al. (2012) pointed to the importance of eggshell emissivity as a factor that could affect egg temperature and thus embryo or hatchling survival. These authors also list further references related to emissivity and bird egg temperature. A priori, it is likely that there is an evolutionary pressure towards a low emission of thermal radiation for birds nesting in cold climates, and the aim of this work is to test the hypothesis that this has affected emissivity. Measurements of infrared reflectance spectra on eggs of domestic hens have been carried out before (Narushin et al., 2004), but these authors used a different measurement geometry and reported their result as transmission in arbitrary units and with an arbitrary zero line. Therefore these results cannot be compared with ours. A note on terminology: some authors distinguish between emissivity of a substance and emittance of an object, for instance emissivity of tungsten as opposed to the (higher) emittance of a coiled tungsten filament of an incandescent lamp. This has been discouraged by others, since “emittance” is also used in other senses. We here use “emissivity” as recommended in the third edition of IUPS (2003) Glossary of terms for thermal physiology, and we distinguish between directional emissivity in a direction near normal to the surface, and hemispherical emissivity, over 2π steradians.
2. Materials Most of the estimates were carried out with empty eggshells from a collection at Lund University. Since collecting eggs of wild birds has been illegal in many countries for a considerable time, most of the eggshells were several decades old, in some cases over one hundred years of age. To ascertain that infrared reflectance did not change with age some estimates were done using eggshells from fresh eggs, and some control estimates were also taken on intact eggs of quail and domestic hen.
3. Methods Direct measurement of thermal emissivity requires heating of the sample above the surroundings and the measuring instrument, and this might result in changed properties, so an indirect method was used. This method depends on Kirchhoff’s law, stating that the emissivity is one minus the reflectance measured in a certain way, if transmission of radiation through the sample is negligible. The reflectance that has been measured is the so-called directional– hemispherical reflectance: radiation is incident at near normal to the sample surface (10° to the normal), and the reflected radiation is measured over 2π steradians of solid angle. Subtraction of these values from 1, multiplication by a blackbody spectrum for 30 °C, integration over the spectrum, and division by the blackbody spectrum integrated over the same spectral range was used as a proxy for the total directional emissivity. The blackbody spectrum was calculated per wavenumber interval, since the instrument spectral step size is uniform on the wavenumber scale, not the wavelength scale. Using two instruments we could cover wavelengths from 2.5 to 22 μm, but for reasons discussed below we have only used values up to 16 mm. Some data for the instruments are shown in Supplementary Table ST1, and further details are available in Hecker et al. (2011). Various port sizes were used with each instrument, but for reasons given below, estimates with the smaller port sizes (10–17 mm diameter) are considered less
reliable and conclusions are based on estimates with 19 and 20 mm ports. To give an idea of reproducibility we compare reflectance spectra of the four eggs in a clutch of two Calidris species in Supplementary Fig. SF1. It is clear that the variability is greatest in the region around 1500 cm 1.
4. Methodological concerns
1. The instruments used are primarily intended for measurements on flat surfaces. Eggs of different sizes have different curvatures. It was therefore important to investigate the effect of curvature on the results. We have done this in several ways: (a) by measuring paper balls of different sizes painted with the same paint. (b) By measuring the same eggs with differently sized apertures (instrument ports). (c) By measuring both the blunt and the pointed end of eggs for which the curvature at the two ends is very different. As expected for mostly specular surfaces, for the smallest eggs as well as the pointed end of small eggs, smaller ports give slightly higher emissivity values than 19 and 20 mm ports, because radiation is restricted to a smaller incidence angle. However, because beam intensity is not constant over the port aperture, an accurate correction for finite port size is not possible, and we have only included values for ports of 19 and 20 mm diameter, i.e. 9.5 and 10 mm radius, in our final computations, in order to make all results comparable. Likewise, although we have done estimates also of the pointed end, we have only included data for the blunt end of the eggs, except for the age comparison in Supplementary Table ST2. The strongest indication that there exist differences in emissivity, independently of any curvature artifact, is that eggs of similar curvature (such as those of quail, Coturnix coturnix on one hand, and sandpipers, Calidris sp. or magpie, Pica pica on the other) can have very different emissivity. The average blunt end radii of curvature in mm were for Cournix coturnix 11.5, Lagopus muta 13.4, Lagopus lagopus 12.3, P. pica 11.3, C. coronae cornix 13.1, and for Calidris species ranging from 10.1 (Calidris temminckii) to 13.1 (Calidris melanotos). These radii were determined from photographs of eggs against a background of mm-ruled paper. When the radius of curvature comes close to the radius of the entrance port of the instrument, below 11 mm in our case, the emissivity values obtained are likely to be lower than true directional (perpendicular) emissivity. This is the case here only for C. minuta and C. temminckii. 2. Most of the eggshells used are old museum specimens. It was therefore important to investigate if age would have an effect on the results. The conclusion from five species for which both old and new eggs were measured is that there is no significant age effect (Supplementary Table ST2). In addition a few full-spectra comparing reflectance of old and new eggs are shown in Supplementary Fig. SF2. 3. To rule out the possibility that the results in some way depend on the instrumentation, two different instruments were used. Fig. 1 shows, as an example, spectral emissivity for eggs of C. alba weighted by the 30 °C blackbody spectrum as measured using a 15 mm port for the instrument in Uppsala and a 20 mm port for the instrument in Enschede (also the blackbody spectrum itself is shown). The Uppsala curve is very noisy at the high-wavelength end, especially for this small port, and even extends above the blackbody curve. Therefore the estimates with the smaller ports are considered uncertain, and not
L.O. Björn et al. / Journal of Thermal Biology 57 (2016) 1–5
3
Table 1 Species and groups (obsolete names in parentheses)
Fig. 1. Comparison of spectral directional emissivity multiplied by 30 °C blackbody emission (blue) for eggs of Calidris alba measured with the instruments in Uppsala (15 mm port, red) and Enschede (20 mm port, green).
further used. However, Fig. 1 illustrates the principle for emissivity computation: integrated emissivity is calculated as the area below the egg curve divided by the area under the blackbody curve for the same spectral interval. Because the instruments record values at fixed wavenumber intervals, a wavenumber scale, not a wavelength scale, must be used. 4. It has not been possible to cover the whole thermal spectrum with the instruments at hand. A main component of all eggshells is calcite. It can be assumed that emissivity differences between eggshells are, to a large extent, due to differences in shielding of calcite by other substances. To get some idea of how spectral coverage affects the data, we have investigated the effect of the measured spectral range from published values (Long et al., 1993) of calcite reflectance. Calcite is birefringent. As an average for ordinary and extraordinary rays we obtain the following emissivity values for integration to various upper wavelengths: 0.909 (16 mm), 0.812 (22 mm), 0.861 (266 mm, i.e. full thermal spectrum). Thus it seems that our integration to 16 mm gives an overestimation of the emissivity of the eggshells, but that integration to 16 mm comes as close to the true value as integration to 22 mm. Our own estimate of a calcite slab from Moura (Spain) with integration to 16 mm gave the value 0.915 for 10° directional emissivity. 5. According to Kirchhoff's law the emissivity is equal to one minus the reflectance, but applying it directly to our estimates of directional–hemispherical reflectance gives us the directional emissivity, not the hemispherical emissivity that is more relevant for egg cooling rates. In this work we are limited to the directional emissivity as a rough proxy for the hemispherical emissivity, 6. To ascertain that observed differences between reflectance and emissivity calculated therefrom is not due to differences in eggshell thickness that might result in differences in transmission of radiation through the eggshells, some measurements were taken on eggshell thickness. There was no correlation between thickness and reflectance.
5. Results and discussion Detailed results of directional emissivity are shown in Supplementary Table ST3, and an overview of directional emissivities in Table 1. Note that for the individual species all estimates are included in the calculation of standard errors, so estimates on eggs from the same clutch, and in some cases estimates of the same egg (at different occasions, with different instruments) are regarded as different samples. For the bird groups each species is regarded as
Passeriformes Pica pica Corvus corone cornix Procellariformes Fulmarus glacialis Alcidae Alle alle Fratercula arctica Uria lomvia Uria aalge Sternidae and Stercorariidae Sterna paradisaea Sterna hirundo Stercorarius longicaudus Calidris Calidris alba Calidris maritima Calidris melanotos Calidris alpina Calidris minuta Calidris temminckii Galliformes Gallus gallus domest. Coturnix coturnix Lagopus muta (mutus) Lagopus lagopus Lophura leucomelanos Argusianus argus Lyurus (Tetrao) tetrix Tetrao urogallus Francolinus francolinus Anseriformes Anas crecca Anas platyrhynchus Anser brachyrhynchus Anser erythropus Palaeognathae Struthio camelus
Number of measurem.
2 5 3 1 10 4 4 11 8 10 3 2 2 2 6 6 6 3 12 3 8 9 5 9 12 8 3 2 2 2 1 4 4 4 1 1 1 2
Directional emissivity Average
Std dev.
0.918 0.918 0.917 0.936 0.936 0.936 0.930 0.927 0.946 0.942 0.936 0.931 0.941 0.935 0.922 0.927 0.930 0.914 0.926 0.924 0.912 0.958 0.963 0.961 0.957 0.954 0.967 0.954 0.958 0.958 0.954 0.955 0.956 0.953 0.942 0.967 0.932 0.932
0.001 0.005 0.023 0.007 0.009 0.019 0.017 0.006 0.011 0.005 0.009 0.003 0.009 0.007 0.007 0.006 0.003 0.007 0.006 0.003 0.005 0.003 0.004 0.006 0.004 0.004 0.004 0.003 0.003 0.010 0.015 0.020
0.005
The directional emissivity values of the Calidris group differ from the Galliformes group (p ¼ 0.001, two-tailed t-test) whether C. minuta and C. temminckii are excluded or not, and from Anseriformes (p ¼ 0.004). The Lophura species called leucomelanos here was labeled Lophura melanotus in the egg collection; the identification as leucomelanos is tentative.
one sample. Galloanserine birds consistently have high values of directional emissivity, around 0.96 (average 7standard deviation 0.9557 0.010 for Anseriformes, 0.95870.005 for Galliformes), while Calidris species, as representatives of Scolopacidae, have relatively low directional emissivities, 0.922 70.007. Other birds have intermediate values, with the exception of P. pica and hooded crow, C. corone cornix, which also have low emissivity. We have found no strict correlation between thermal emissivity and the nesting conditions of the birds; with reservation for the few taxa investigated, the emissivity appears more related to phylogeny (see Jarvis et al. (2014), for a recent overview of bird phylogeny). In addition to the examples above, it can be mentioned that the fulmar (Fulmarus glacialis) and the guillemots (Uria lomvia and Uria aalge), having open, exposed nest-sites, do not have significantly lower emissivity than the burrow-nesting puffin (F. arctica), and ptarmigan (L. muta) nesting in Arctic and Subarctic regions, not lower than tropical Galliformes. Passerines are known to have fragile egg cuticles, easily exposing calcite, a property which may be related to the low-values for P. pica and C. corone cornix. We want to point out that radiative (as well as convective) heat dissipation does not depend only on mid-infrared emissivity, but also on the shape and size of eggs. Thus calculations indicate that shape and size differences are at least as important as emissivity
4
L.O. Björn et al. / Journal of Thermal Biology 57 (2016) 1–5
wavelengths for calcite reflection peaks around 4.5 and 11.3 mm there are interesting differences between bird species indicating that reflectance differences are due to various factors, not only differences in modulation of calcite reflectance (Fig. 2). There was no consistent relation between visible color and thermal emissivity.
6. Conclusion There are clear differences between egg thermal directional emissivities between bird species. The differences are small, and no ecological importance has been established, but emissivity differences should be considered when estimating egg temperatures using thermal radiation thermometers or infrared cameras. Before drawing any further conclusions, attention should be paid to possible differences between directional and hemispherical emissivity. For some infrared spectral regions, there are large differences in egg reflectance among species.
Acknowledgments
Fig. 2. Details of reflectance spectra. Moura calcite, Calidris melanotos (pectoral sandpiper), Pica pica (magpie), and Coturnix coturnix (quail). P. pica and C. melanotos, although not closely related, are spectrally very similar in both short (left) and long (right) wavelength regions. Pure calcite has much higher reflectance than eggshells in the higher wavelength region (right), but in the lower wavelength region (left) its reflectance is lower and more similar to that of C. coturnix. This indicates that egg reflectance is due to several factors, not just different modulation of calcite reflectance.
differences. In the natural condition, the more conical shape of eggs of Calidris species as compared to, e.g., those of Coturnix, makes it easier for the incubating birds to protect the eggs from predation and cooling when laid in quadruples (Andersson, 1978). At least for some birds the rate of embryo development and required length of the incubation period depend on temperature (Nakage et al. 2003; Hepp et al., 2005). Every day of prolongation increases the risk of predation on eggs and incubating birds, as well as the risk that the young birds breeding in northern regions with a narrow time-window will not be able to complete their development to ensure a successful migration to the overwintering area. Nevertheless, even considering this, our conclusion is that the small differences in egg emissivity are of no relevance for survival, and other circumstances, such as nest construction, are more important for the temperature of the eggs. Differences in eggshell emissivity have implications for scientific methodology. Infrared thermometers and cameras have been used to monitor avian egg temperatures (e.g., Lamprecht and Schmolz, 2004; Hulet et al., 2007). Serious errors may result if correct egg emissivity is not taken into account. The range of emissivity described here corresponds to an apparent temperature range of up to one degree, or even a little more. The reflectance spectra reveal large species differences in particular wavelength regions. One interesting case is a reflectance peak around 3.76 mm. At this peak, reflectance is 3 times as high for F. arctica as for most galliform birds, and high also for F. glacialis, while it has intermediate values for the other tested birds. Also at the
We thank Eva and Lennart Engström for gift of an ostrich egg from their farm, Edward Lantz for a gift of quail eggs from his farm, and Dr. J. Boudewijn de Smeth for gifts of hen’s eggs fresh from a farm and for help and encouragement. We are grateful to Dr. Gudmundur A. Gudmundsson and Mr. Thorvaldur Thor Björnsson, Icelandic Institute of Natural History, Reykjavik for providing eggshells from fresh eggs of species of seabirds. Curator Jonas Ekström of Lund University Biology Museum was helpful with eggs from the collections. Per Vestergren and Stefan Sydoff helped LOB in the workshop. Financial support for LOB’s travel and construction of apparatus for preliminary experiments has been provided by the Gerda och Henry Dunker’s foundation through the Royal Physiographical Society in Lund, the Leading Talent Project of Guangdong Provincial Government, and the Pearl River Scholar Funded Scheme of Guangdong Province, China. Lund University has provided most of the eggshells used in the study. We have followed the Ornithological Council’s Guidelines to the use of wild birds in research, third edition 2010, and EU Directive 2010/63/EU for animal experiments. We thank the reviewers for helpful comments and Dr Helena Björn van Praagh for help with formatting of figure files.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.jtherbio.2015.11.008.
References Andersson, M., 1978. Optimal egg shape in waders. Ornis Fenn. 55, 105–109. Arnold, J.M., Sabom, D., Nisbet, I.C.T., Hatch, J.J., 2006. Use of temperature sensors to monitor patterns of nocturnal desertion by incubating common terns. J. Field Ornithol. 77, 384–391. Björn, L.O., Uvdal, P., Li, S., 2012. Ecological importance of the thermal emissivity of avian eggshells. J. Theor. Biol. 301, 62–66. Carter, A.W., Hopkins, W.A., Moore, I.T., DuRant, S.E., 2014. Influence of incubation recess patterns on incubation period and hatchling traits in wood ducks Aix sponsa. J. Avian Biol. 45, 273–279. Funk, E.M., Biellier, H.V., 1944. The minimum temperature for embryonic development in the domestic fowl (Gallus domesticus). Poult. Sci. 23, 538–540. Hecker, C., Hook, S., van der Meijde, M., Bakker, W., van der Werff, H., Wilbrink, H., van Ruitenbeek, F., de Smeth, B., van der Meer, F., 2011. Thermal infrared spectrometer for earth science remote sensing applications—instrument modifications and measurement procedures. Sensors 11, 10981–10999, available at 〈〈http://www.mdpi.com/1424-8220/11/11/10981〉〉. Hepp, G.R., Folk, T.H., Manlove, C.A., 2005. Nest temperature, incubation period, and
L.O. Björn et al. / Journal of Thermal Biology 57 (2016) 1–5
investment decisions of incubating wood ducks Aix sponsa. J. Avian Biol. 36, 523–530. Hepp, G.R., Kennamer, R.A., 2012. Warm is better: Incubation temperature influences apparent survival and recruitment of wood ducks (Aix sponsa). PLoS One 7 (e47777), 6. Hulet, R., Gladys, G., Hill, D., Meijerhof, R., El-Shiekh, T., 2007. Influence of egg shell embryonic incubation temperature and broiler breeder flock age on posthatch growth performance and carcass characteristics. Poult. Sci. 86, 408–412. IUPS, 2003. Glossary of terms for thermal physiology (3rd ed.). J. Therm. Biol. 28, 75–106. Jarvis, E.D., Mirarab, S., Aberer, A.J., et al., 2014. Whole-genome analyses resolve early branches in the tree of life of modern birds. Science 346, 1320–1331. Lamprecht, I., Schmolz, E., 2004. Thermal investigations of some bird nests. Thermochim. Acta 415, 141–148. Long, L.L., Querry, M.R., Bell, R.J., Alexander, R.W., 1993. Optical properties of calcite and gypsum in crystalline and powdered form in the infrared and far-infrared. Infrared Phys. 34, 191–201. Mortola, J.P., 2006. Metabolic response to cooling temperatures in chicken embryos and hatchlings after cold incubation. Comp. Biochem. Physiol. A 145, 441–448. Nakage, E.S., Cordozo, J.P., Pereira, G.T., Queiroz, S.A., Boleli, I.C., 2003. Effect of temperature on incubation period, embryonic mortality, hatch rate, egg water loss and partridge chick weight (Rhynchotus rufescens). Rev. Bras. Cienc. Avic. 5, 131–135. Narushin, V.G., van Kempen, T.A., Wineland, M.J., Christensen, V.L., 2004. Comparing infrared spectroscopy and egg size measurements for predicting eggshell quality. Biosyst. Eng. 87, 367–373. Olson, C.R., Vleck, C.M., Vleck, D., 2006. Periodic cooling of bird eggs reduces embryonic growth efficiency. Physiol. Biochem. Zool. 79, 927–936. Reneerkens, J., Grond, K., Schekkerman, H., Tulp, I., Piersma, T., 2011. Do uniparental sanderlings Calidris alba increase egg heat input to compensate for low nest attentiveness? PLoS One 6 (e16834), 9. Rockweit, J.T., Franklin, A.B., Bakken, G.S., Gutiérrez, R.J., 2012. Potential influences of climate and nest structure on spotted owl reproductive success: a biophysical approach. PLoS One 7 (e41498), 11. Tulp, I., Schekkerman, H., 2006. Time allocation between feeding and incubation in uniparental arctic-breeding shorebirds: energy reserves provide leeway in a tight schedule. J. Avian Biol. 37, 207–218.
5 Christoph Hecker
Saleem Ullah
Arne Roos
Lars Olof Björn
Annica M. Nilsson
Sven-Axel Bengtson
Shaoshan Li
Contents lists available at ScienceDirect
Journal of Thermal Biology journal homepage: www.elsevier.com/locate/jtherbio
Thermal emissivity of avian eggshells Lars Olof Björn a,b,n, Sven-Axel Bengtson b, Shaoshan Li a,nn, Christoph Hecker c, Saleem Ullah d, Arne Roos e, Annica M. Nilsson e a Key Laboratory of Ecology and Environmental Science in Guangdong Higher Education, School of Life Science, South China Normal University, Guangzhou 510631, China b Lund University, Department of Biology, Sölvegatan 35, SE-22362 Lund, Sweden c University of Twente, Faculty of Geo-Information Science and Earth Observation (ITC), Department of Earth Systems Analysis, 7500AA Enschede, The Netherlands d Department of Space Sciences, Institute of Space Technology, P.O. Box 2750, Islamabad, Pakistan e Uppsala University, Dept of Engineering Sciences, Solid State Physics, Box 534, 751 21 Uppsala, Sweden
art ic l e i nf o
a b s t r a c t
Article history: Received 15 September 2015 Received in revised form 11 November 2015 Accepted 26 November 2015 Available online 4 January 2016
The hypothesis has been tested that evolution has resulted in lower thermal emissivity of eggs of birds breeding openly in cold climates than of eggs of birds that nest under protective covering or in warmer climates. Directional thermal emissivity has been estimated from directional–hemispherical reflectance spectra. Due to several methodological difficulties the absolute emissivity is not accurately determined, but differences between species are obvious. Most notably, small waders of the genus Calidris, breeding in cold climates on the tundra, and in most cases with uniparental nest attendance, have low directional emissivity of their eggshells, about 0.92 when integration is carried out for wavelengths up to 16 μm. Species belonging to Galloanserinae have the highest directional emissivity, about 0.96, of their eggs. No differences due to climate or breeding conditions were found within this group. Eggs of most other birds tested possess intermediate emissivity, but the values for Pica pica and Corvus corone cornix are as low as for Calidris. Large species-dependent differences in spectral reflectance were found at specific wavelengths. For instance, at 4.259 μm the directional–hemispherical reflectance for galliforms range from 0.05 to 0.09, while for Fratercula arctica and Fulmarus glacialis it is about 0.3. The reflection peaks at 6.5 and 11.3 μm due to calcite are differentially attenuated in different species. In conclusion, the hypothesis that evolution has resulted in lower thermal emissivity of bird eggs being exposed in cold climates is not supported by our results. The emissivity is not clearly related to nesting habits or climate, and it is unlikely that the small differences observed are ecologically important. The spectral differences between eggs that nevertheless exist should be taken into account when using infrared thermometers for estimating the surface temperature of avian eggs. & 2016 Published by Elsevier Ltd.
Keywords: Avian eggs Birds Heat dissipation Egg cooling Thermal emissivity Incubation Seabirds Thermal radiation Waders
1. Introduction Most bird eggs are heated by the bodies of the parents during the development of the embryo, and keeping the eggs at a temperature much above ambient is important in cold climates. For various reasons parents may leave their nests unattended for shorter or longer periods, during which the eggs cool. In some bird species only one parent attends to the eggs, and has to leave the nest for feeding. The little stint (Calidris minuta) may leave its nest unattended for half an hour, during which the egg surface n Corresponding author at: Lund University, Department of Biology, Sölvegatan 35, SE-22362 Lund, Sweden. nn Corresponding author. E-mail addresses: [email protected] (L.O. Björn), [email protected] (S. Li).
http://dx.doi.org/10.1016/j.jtherbio.2015.11.008 0306-4565/& 2016 Published by Elsevier Ltd.
temperature can fall below 10 °C (Tulp and Schekkerman, 2006). During snowstorms eggs of sanderling (Calidris alba) may be abandoned for extended times (Reneerkens et al., 2011). In other cases the parents are forced to leave the nest due to predators. Thus common terns (Sterna hirundo) have been observed to leave nests for hours during night due to owls (Arnold et al., 2006). Even if embryos survive and eggs hatch, low incubation temperature has negative effects on hatchling development and survival (Olson et al., 2006; Hepp and Kennamer, 2012; Carter et al., 2014). Hepp and Kennamer (2012) found that the incubation time was increased from 30.1 to 37.9 days, i.e. by 26% when the temperature was lowered from 37.3 to 35.0 °C. Development in domestic fowl comes to a stop below 25 °C (Funk and Biellier, 1944) and hatching is seriously compromised already at a temperature four degrees below normal (Mortola, 2006). Clearly, it is of survival value for the embryo if the egg is constructed such that cooling takes place as
2
L.O. Björn et al. / Journal of Thermal Biology 57 (2016) 1–5
slowly as possible during these periods. Cooling takes place mainly by radiation and air convection, and although other factors, such as nest construction, probably are more important (Rockweit et al., 2012). Björn et al. (2012) pointed to the importance of eggshell emissivity as a factor that could affect egg temperature and thus embryo or hatchling survival. These authors also list further references related to emissivity and bird egg temperature. A priori, it is likely that there is an evolutionary pressure towards a low emission of thermal radiation for birds nesting in cold climates, and the aim of this work is to test the hypothesis that this has affected emissivity. Measurements of infrared reflectance spectra on eggs of domestic hens have been carried out before (Narushin et al., 2004), but these authors used a different measurement geometry and reported their result as transmission in arbitrary units and with an arbitrary zero line. Therefore these results cannot be compared with ours. A note on terminology: some authors distinguish between emissivity of a substance and emittance of an object, for instance emissivity of tungsten as opposed to the (higher) emittance of a coiled tungsten filament of an incandescent lamp. This has been discouraged by others, since “emittance” is also used in other senses. We here use “emissivity” as recommended in the third edition of IUPS (2003) Glossary of terms for thermal physiology, and we distinguish between directional emissivity in a direction near normal to the surface, and hemispherical emissivity, over 2π steradians.
2. Materials Most of the estimates were carried out with empty eggshells from a collection at Lund University. Since collecting eggs of wild birds has been illegal in many countries for a considerable time, most of the eggshells were several decades old, in some cases over one hundred years of age. To ascertain that infrared reflectance did not change with age some estimates were done using eggshells from fresh eggs, and some control estimates were also taken on intact eggs of quail and domestic hen.
3. Methods Direct measurement of thermal emissivity requires heating of the sample above the surroundings and the measuring instrument, and this might result in changed properties, so an indirect method was used. This method depends on Kirchhoff’s law, stating that the emissivity is one minus the reflectance measured in a certain way, if transmission of radiation through the sample is negligible. The reflectance that has been measured is the so-called directional– hemispherical reflectance: radiation is incident at near normal to the sample surface (10° to the normal), and the reflected radiation is measured over 2π steradians of solid angle. Subtraction of these values from 1, multiplication by a blackbody spectrum for 30 °C, integration over the spectrum, and division by the blackbody spectrum integrated over the same spectral range was used as a proxy for the total directional emissivity. The blackbody spectrum was calculated per wavenumber interval, since the instrument spectral step size is uniform on the wavenumber scale, not the wavelength scale. Using two instruments we could cover wavelengths from 2.5 to 22 μm, but for reasons discussed below we have only used values up to 16 mm. Some data for the instruments are shown in Supplementary Table ST1, and further details are available in Hecker et al. (2011). Various port sizes were used with each instrument, but for reasons given below, estimates with the smaller port sizes (10–17 mm diameter) are considered less
reliable and conclusions are based on estimates with 19 and 20 mm ports. To give an idea of reproducibility we compare reflectance spectra of the four eggs in a clutch of two Calidris species in Supplementary Fig. SF1. It is clear that the variability is greatest in the region around 1500 cm 1.
4. Methodological concerns
1. The instruments used are primarily intended for measurements on flat surfaces. Eggs of different sizes have different curvatures. It was therefore important to investigate the effect of curvature on the results. We have done this in several ways: (a) by measuring paper balls of different sizes painted with the same paint. (b) By measuring the same eggs with differently sized apertures (instrument ports). (c) By measuring both the blunt and the pointed end of eggs for which the curvature at the two ends is very different. As expected for mostly specular surfaces, for the smallest eggs as well as the pointed end of small eggs, smaller ports give slightly higher emissivity values than 19 and 20 mm ports, because radiation is restricted to a smaller incidence angle. However, because beam intensity is not constant over the port aperture, an accurate correction for finite port size is not possible, and we have only included values for ports of 19 and 20 mm diameter, i.e. 9.5 and 10 mm radius, in our final computations, in order to make all results comparable. Likewise, although we have done estimates also of the pointed end, we have only included data for the blunt end of the eggs, except for the age comparison in Supplementary Table ST2. The strongest indication that there exist differences in emissivity, independently of any curvature artifact, is that eggs of similar curvature (such as those of quail, Coturnix coturnix on one hand, and sandpipers, Calidris sp. or magpie, Pica pica on the other) can have very different emissivity. The average blunt end radii of curvature in mm were for Cournix coturnix 11.5, Lagopus muta 13.4, Lagopus lagopus 12.3, P. pica 11.3, C. coronae cornix 13.1, and for Calidris species ranging from 10.1 (Calidris temminckii) to 13.1 (Calidris melanotos). These radii were determined from photographs of eggs against a background of mm-ruled paper. When the radius of curvature comes close to the radius of the entrance port of the instrument, below 11 mm in our case, the emissivity values obtained are likely to be lower than true directional (perpendicular) emissivity. This is the case here only for C. minuta and C. temminckii. 2. Most of the eggshells used are old museum specimens. It was therefore important to investigate if age would have an effect on the results. The conclusion from five species for which both old and new eggs were measured is that there is no significant age effect (Supplementary Table ST2). In addition a few full-spectra comparing reflectance of old and new eggs are shown in Supplementary Fig. SF2. 3. To rule out the possibility that the results in some way depend on the instrumentation, two different instruments were used. Fig. 1 shows, as an example, spectral emissivity for eggs of C. alba weighted by the 30 °C blackbody spectrum as measured using a 15 mm port for the instrument in Uppsala and a 20 mm port for the instrument in Enschede (also the blackbody spectrum itself is shown). The Uppsala curve is very noisy at the high-wavelength end, especially for this small port, and even extends above the blackbody curve. Therefore the estimates with the smaller ports are considered uncertain, and not
L.O. Björn et al. / Journal of Thermal Biology 57 (2016) 1–5
3
Table 1 Species and groups (obsolete names in parentheses)
Fig. 1. Comparison of spectral directional emissivity multiplied by 30 °C blackbody emission (blue) for eggs of Calidris alba measured with the instruments in Uppsala (15 mm port, red) and Enschede (20 mm port, green).
further used. However, Fig. 1 illustrates the principle for emissivity computation: integrated emissivity is calculated as the area below the egg curve divided by the area under the blackbody curve for the same spectral interval. Because the instruments record values at fixed wavenumber intervals, a wavenumber scale, not a wavelength scale, must be used. 4. It has not been possible to cover the whole thermal spectrum with the instruments at hand. A main component of all eggshells is calcite. It can be assumed that emissivity differences between eggshells are, to a large extent, due to differences in shielding of calcite by other substances. To get some idea of how spectral coverage affects the data, we have investigated the effect of the measured spectral range from published values (Long et al., 1993) of calcite reflectance. Calcite is birefringent. As an average for ordinary and extraordinary rays we obtain the following emissivity values for integration to various upper wavelengths: 0.909 (16 mm), 0.812 (22 mm), 0.861 (266 mm, i.e. full thermal spectrum). Thus it seems that our integration to 16 mm gives an overestimation of the emissivity of the eggshells, but that integration to 16 mm comes as close to the true value as integration to 22 mm. Our own estimate of a calcite slab from Moura (Spain) with integration to 16 mm gave the value 0.915 for 10° directional emissivity. 5. According to Kirchhoff's law the emissivity is equal to one minus the reflectance, but applying it directly to our estimates of directional–hemispherical reflectance gives us the directional emissivity, not the hemispherical emissivity that is more relevant for egg cooling rates. In this work we are limited to the directional emissivity as a rough proxy for the hemispherical emissivity, 6. To ascertain that observed differences between reflectance and emissivity calculated therefrom is not due to differences in eggshell thickness that might result in differences in transmission of radiation through the eggshells, some measurements were taken on eggshell thickness. There was no correlation between thickness and reflectance.
5. Results and discussion Detailed results of directional emissivity are shown in Supplementary Table ST3, and an overview of directional emissivities in Table 1. Note that for the individual species all estimates are included in the calculation of standard errors, so estimates on eggs from the same clutch, and in some cases estimates of the same egg (at different occasions, with different instruments) are regarded as different samples. For the bird groups each species is regarded as
Passeriformes Pica pica Corvus corone cornix Procellariformes Fulmarus glacialis Alcidae Alle alle Fratercula arctica Uria lomvia Uria aalge Sternidae and Stercorariidae Sterna paradisaea Sterna hirundo Stercorarius longicaudus Calidris Calidris alba Calidris maritima Calidris melanotos Calidris alpina Calidris minuta Calidris temminckii Galliformes Gallus gallus domest. Coturnix coturnix Lagopus muta (mutus) Lagopus lagopus Lophura leucomelanos Argusianus argus Lyurus (Tetrao) tetrix Tetrao urogallus Francolinus francolinus Anseriformes Anas crecca Anas platyrhynchus Anser brachyrhynchus Anser erythropus Palaeognathae Struthio camelus
Number of measurem.
2 5 3 1 10 4 4 11 8 10 3 2 2 2 6 6 6 3 12 3 8 9 5 9 12 8 3 2 2 2 1 4 4 4 1 1 1 2
Directional emissivity Average
Std dev.
0.918 0.918 0.917 0.936 0.936 0.936 0.930 0.927 0.946 0.942 0.936 0.931 0.941 0.935 0.922 0.927 0.930 0.914 0.926 0.924 0.912 0.958 0.963 0.961 0.957 0.954 0.967 0.954 0.958 0.958 0.954 0.955 0.956 0.953 0.942 0.967 0.932 0.932
0.001 0.005 0.023 0.007 0.009 0.019 0.017 0.006 0.011 0.005 0.009 0.003 0.009 0.007 0.007 0.006 0.003 0.007 0.006 0.003 0.005 0.003 0.004 0.006 0.004 0.004 0.004 0.003 0.003 0.010 0.015 0.020
0.005
The directional emissivity values of the Calidris group differ from the Galliformes group (p ¼ 0.001, two-tailed t-test) whether C. minuta and C. temminckii are excluded or not, and from Anseriformes (p ¼ 0.004). The Lophura species called leucomelanos here was labeled Lophura melanotus in the egg collection; the identification as leucomelanos is tentative.
one sample. Galloanserine birds consistently have high values of directional emissivity, around 0.96 (average 7standard deviation 0.9557 0.010 for Anseriformes, 0.95870.005 for Galliformes), while Calidris species, as representatives of Scolopacidae, have relatively low directional emissivities, 0.922 70.007. Other birds have intermediate values, with the exception of P. pica and hooded crow, C. corone cornix, which also have low emissivity. We have found no strict correlation between thermal emissivity and the nesting conditions of the birds; with reservation for the few taxa investigated, the emissivity appears more related to phylogeny (see Jarvis et al. (2014), for a recent overview of bird phylogeny). In addition to the examples above, it can be mentioned that the fulmar (Fulmarus glacialis) and the guillemots (Uria lomvia and Uria aalge), having open, exposed nest-sites, do not have significantly lower emissivity than the burrow-nesting puffin (F. arctica), and ptarmigan (L. muta) nesting in Arctic and Subarctic regions, not lower than tropical Galliformes. Passerines are known to have fragile egg cuticles, easily exposing calcite, a property which may be related to the low-values for P. pica and C. corone cornix. We want to point out that radiative (as well as convective) heat dissipation does not depend only on mid-infrared emissivity, but also on the shape and size of eggs. Thus calculations indicate that shape and size differences are at least as important as emissivity
4
L.O. Björn et al. / Journal of Thermal Biology 57 (2016) 1–5
wavelengths for calcite reflection peaks around 4.5 and 11.3 mm there are interesting differences between bird species indicating that reflectance differences are due to various factors, not only differences in modulation of calcite reflectance (Fig. 2). There was no consistent relation between visible color and thermal emissivity.
6. Conclusion There are clear differences between egg thermal directional emissivities between bird species. The differences are small, and no ecological importance has been established, but emissivity differences should be considered when estimating egg temperatures using thermal radiation thermometers or infrared cameras. Before drawing any further conclusions, attention should be paid to possible differences between directional and hemispherical emissivity. For some infrared spectral regions, there are large differences in egg reflectance among species.
Acknowledgments
Fig. 2. Details of reflectance spectra. Moura calcite, Calidris melanotos (pectoral sandpiper), Pica pica (magpie), and Coturnix coturnix (quail). P. pica and C. melanotos, although not closely related, are spectrally very similar in both short (left) and long (right) wavelength regions. Pure calcite has much higher reflectance than eggshells in the higher wavelength region (right), but in the lower wavelength region (left) its reflectance is lower and more similar to that of C. coturnix. This indicates that egg reflectance is due to several factors, not just different modulation of calcite reflectance.
differences. In the natural condition, the more conical shape of eggs of Calidris species as compared to, e.g., those of Coturnix, makes it easier for the incubating birds to protect the eggs from predation and cooling when laid in quadruples (Andersson, 1978). At least for some birds the rate of embryo development and required length of the incubation period depend on temperature (Nakage et al. 2003; Hepp et al., 2005). Every day of prolongation increases the risk of predation on eggs and incubating birds, as well as the risk that the young birds breeding in northern regions with a narrow time-window will not be able to complete their development to ensure a successful migration to the overwintering area. Nevertheless, even considering this, our conclusion is that the small differences in egg emissivity are of no relevance for survival, and other circumstances, such as nest construction, are more important for the temperature of the eggs. Differences in eggshell emissivity have implications for scientific methodology. Infrared thermometers and cameras have been used to monitor avian egg temperatures (e.g., Lamprecht and Schmolz, 2004; Hulet et al., 2007). Serious errors may result if correct egg emissivity is not taken into account. The range of emissivity described here corresponds to an apparent temperature range of up to one degree, or even a little more. The reflectance spectra reveal large species differences in particular wavelength regions. One interesting case is a reflectance peak around 3.76 mm. At this peak, reflectance is 3 times as high for F. arctica as for most galliform birds, and high also for F. glacialis, while it has intermediate values for the other tested birds. Also at the
We thank Eva and Lennart Engström for gift of an ostrich egg from their farm, Edward Lantz for a gift of quail eggs from his farm, and Dr. J. Boudewijn de Smeth for gifts of hen’s eggs fresh from a farm and for help and encouragement. We are grateful to Dr. Gudmundur A. Gudmundsson and Mr. Thorvaldur Thor Björnsson, Icelandic Institute of Natural History, Reykjavik for providing eggshells from fresh eggs of species of seabirds. Curator Jonas Ekström of Lund University Biology Museum was helpful with eggs from the collections. Per Vestergren and Stefan Sydoff helped LOB in the workshop. Financial support for LOB’s travel and construction of apparatus for preliminary experiments has been provided by the Gerda och Henry Dunker’s foundation through the Royal Physiographical Society in Lund, the Leading Talent Project of Guangdong Provincial Government, and the Pearl River Scholar Funded Scheme of Guangdong Province, China. Lund University has provided most of the eggshells used in the study. We have followed the Ornithological Council’s Guidelines to the use of wild birds in research, third edition 2010, and EU Directive 2010/63/EU for animal experiments. We thank the reviewers for helpful comments and Dr Helena Björn van Praagh for help with formatting of figure files.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.jtherbio.2015.11.008.
References Andersson, M., 1978. Optimal egg shape in waders. Ornis Fenn. 55, 105–109. Arnold, J.M., Sabom, D., Nisbet, I.C.T., Hatch, J.J., 2006. Use of temperature sensors to monitor patterns of nocturnal desertion by incubating common terns. J. Field Ornithol. 77, 384–391. Björn, L.O., Uvdal, P., Li, S., 2012. Ecological importance of the thermal emissivity of avian eggshells. J. Theor. Biol. 301, 62–66. Carter, A.W., Hopkins, W.A., Moore, I.T., DuRant, S.E., 2014. Influence of incubation recess patterns on incubation period and hatchling traits in wood ducks Aix sponsa. J. Avian Biol. 45, 273–279. Funk, E.M., Biellier, H.V., 1944. The minimum temperature for embryonic development in the domestic fowl (Gallus domesticus). Poult. Sci. 23, 538–540. Hecker, C., Hook, S., van der Meijde, M., Bakker, W., van der Werff, H., Wilbrink, H., van Ruitenbeek, F., de Smeth, B., van der Meer, F., 2011. Thermal infrared spectrometer for earth science remote sensing applications—instrument modifications and measurement procedures. Sensors 11, 10981–10999, available at 〈〈http://www.mdpi.com/1424-8220/11/11/10981〉〉. Hepp, G.R., Folk, T.H., Manlove, C.A., 2005. Nest temperature, incubation period, and
L.O. Björn et al. / Journal of Thermal Biology 57 (2016) 1–5
investment decisions of incubating wood ducks Aix sponsa. J. Avian Biol. 36, 523–530. Hepp, G.R., Kennamer, R.A., 2012. Warm is better: Incubation temperature influences apparent survival and recruitment of wood ducks (Aix sponsa). PLoS One 7 (e47777), 6. Hulet, R., Gladys, G., Hill, D., Meijerhof, R., El-Shiekh, T., 2007. Influence of egg shell embryonic incubation temperature and broiler breeder flock age on posthatch growth performance and carcass characteristics. Poult. Sci. 86, 408–412. IUPS, 2003. Glossary of terms for thermal physiology (3rd ed.). J. Therm. Biol. 28, 75–106. Jarvis, E.D., Mirarab, S., Aberer, A.J., et al., 2014. Whole-genome analyses resolve early branches in the tree of life of modern birds. Science 346, 1320–1331. Lamprecht, I., Schmolz, E., 2004. Thermal investigations of some bird nests. Thermochim. Acta 415, 141–148. Long, L.L., Querry, M.R., Bell, R.J., Alexander, R.W., 1993. Optical properties of calcite and gypsum in crystalline and powdered form in the infrared and far-infrared. Infrared Phys. 34, 191–201. Mortola, J.P., 2006. Metabolic response to cooling temperatures in chicken embryos and hatchlings after cold incubation. Comp. Biochem. Physiol. A 145, 441–448. Nakage, E.S., Cordozo, J.P., Pereira, G.T., Queiroz, S.A., Boleli, I.C., 2003. Effect of temperature on incubation period, embryonic mortality, hatch rate, egg water loss and partridge chick weight (Rhynchotus rufescens). Rev. Bras. Cienc. Avic. 5, 131–135. Narushin, V.G., van Kempen, T.A., Wineland, M.J., Christensen, V.L., 2004. Comparing infrared spectroscopy and egg size measurements for predicting eggshell quality. Biosyst. Eng. 87, 367–373. Olson, C.R., Vleck, C.M., Vleck, D., 2006. Periodic cooling of bird eggs reduces embryonic growth efficiency. Physiol. Biochem. Zool. 79, 927–936. Reneerkens, J., Grond, K., Schekkerman, H., Tulp, I., Piersma, T., 2011. Do uniparental sanderlings Calidris alba increase egg heat input to compensate for low nest attentiveness? PLoS One 6 (e16834), 9. Rockweit, J.T., Franklin, A.B., Bakken, G.S., Gutiérrez, R.J., 2012. Potential influences of climate and nest structure on spotted owl reproductive success: a biophysical approach. PLoS One 7 (e41498), 11. Tulp, I., Schekkerman, H., 2006. Time allocation between feeding and incubation in uniparental arctic-breeding shorebirds: energy reserves provide leeway in a tight schedule. J. Avian Biol. 37, 207–218.
5 Christoph Hecker
Saleem Ullah
Arne Roos
Lars Olof Björn
Annica M. Nilsson
Sven-Axel Bengtson
Shaoshan Li