Electronic Supporting Material (ESM)
Conservation through connectivity: can isotopic gradients in Africa reveal winter quarters of a migratory bird?
Thomas S. Reichlin, Keith A. Hobson, Steven L. Van Wilgenburg, Michael Schaub,Leonard I. Wassenaar, Manuel Martín-Vivaldi, Raphaël Arlettaz, Lukas Jenni
Appendix S1
Assignment of geographic origins
Hoopoes were assigned to their geographic origins using likelihood based assignments (Hobson et al. 2009). To accomplish this, we first converted a GIS-based model of mean annual δ2H in precipitation (hereafter MAD; Bowen et al. 2005), into a δ2Hf model (hereafter feather isoscape). In order to calibrate a δ2Hf isoscape, we first obtained data for known origin birds, using all non-waterbird data reported in Bowenet al. (2005), 226 Eurasian reed warblers (Acrocephalus scirpaceus; Petr Procházka, unpublished data) and42 hoopoes (this study). From42 second-year hoopoes ringed as nestlings the previous year, we clipped a piece of a primary flight feather (P4 or P5; feathers are counted from the distal side), which was not moulted during winter and hence carried the isotopic value of the natal site.
Known-origin feather values were regressed against MAD, using General Linear Models (GLM) including MAD as a covariate, and species (or species group, e.g. Tetraonids) as factor, and the interaction between species and MAD. We selected between competing models using AICc (Burnham and Anderson 2002). The top model received 71% of the support based on AICc weights, and included species and MAD. The second best model included the interaction term, and received 29% of the support. Although the second best model received non-trivial support, we chose the less complex model, in part due to a lack of broad scale sampling of hoopoe. The selected GLM explained 75.5% of the variance in δ2Hf. The resulting calibration equation for hoopoe (δ2Hf = 19.53 + 1.27 MAD), was used to create a hoopoe specific δ2Hf isoscape (Fig. S1). For each sample, we then assessed the probability that any given cell within the δ2Hf isoscape represented the origin of an individual by using a Normal probability density function as follows:
(1)
where f(y* | μc,σc) represents the probability that a given cell within the δ2Hf isoscape (c) represents a potential origin for an individual of unknown origin (y*), given the expected mean δ2Hf for that cell (μc) from the calibrated δ2Hf isoscape, and the expected standard deviation (σc) of δ2Hf between individuals growing their feathers at the same locality. The standard deviation of the residuals from the regression equation reported above was used to estimate σc(σ =12.54). For each individual, this resulted in a surface of spatially explicit probability densities (i.e. one surface per bird).
Birds were assigned to their likely origins (one bird at a time) based on the spatially explicit probability densities. In order to accomplish these assignments, we first normalized the probability densities. We normalized the probability densities by dividing the estimated probability density for a given cell in the isoscape by the sum of the probability densities across all cells in the isoscape, resulting in the probabilities across the whole surface summing to 1. We then assigned individuals to the basemap by first determining the odds that any given potential origin was correct relative the odds it was incorrect. Based on 2:1 odds that a given bird had truly originated from within the range defined by the odds ratio, we recoded the raster cells that defined the upper 67% of estimated “probabilities of origin” from the raster surface that was normalized to the maximum, and coded those as 1. All others cells were coded as 0, resulting in one binary map (1=likely origin, 0=unlikely origin) per assigned individual. The results of the individual assignments were then summed over all individuals, by addition of the surfaces. We facilitated this step by rescaling the probabilities relative to the maximum value within the surface prior to applying the odds ratio based reclassification. All assignments were completed using functions written within the R statistical computing environment (R Development Core Team 2009) using the ‘raster’ package.
Fig. S1Feather deuterium isoscape for Africa and Southern Europe derived from the MAD grid calibrated with the regression equation δ2Hf = 19.53 + 1.27 MAD.
1
Appendix S2
Museum specimens
Deuterium values (δ2Hfin ‰) of hoopoe museum specimens from the collections of the Natural History Museum at Tring (T), London, UK; the Muséum National d’Histoire Naturelle in Paris (P), France; and from the Natural History Museum of Basel (B) in Switzerland. Data from birds without coordinates were not included in the analysis.
Country / Coordinates / Subspecies / Collection / δ2Hx / y / (as given on the museum label) / No. / Date
Ethiopia / 8°24'N / 39°00'E / U. epops / T / 1912.10.15.2029 / 18.12.1904 / -34.24
Ethiopia / 9°19′N / 4270'E / U. epops / T / 1903.9.5.57 / 08.01.1902 / -81.30
Ethiopia / 14°00'N / 40°30'E / U. epops / T / 1903.9.5.127 / 02.02.1902 / 16.28
Ethiopia / - / - / U. epops / T / 1900.1.3.372 / 12.01.1899 / 6.66
Ethiopia / 8°11'N / 34°51'E / U. e. senegalensis / T / 1938.5.18.169 / 26.12.1916 / 20.38
Ethiopia / - / - / U. epops / T / 1923.11.10.115 / 12.01.1900 / 29.70
Uganda / 2°20'N / 32°35'E / U. e. somaliensis / T / 1923.8.7.5307 / 01.01.1910 / 3.91
Uganda / - / - / U. senegalensis / T / 1916.12.1.295 / 23.02.1913 / 12.11
Uganda / - / - / U. senegalensis / T / 1916.12.1.294 / 01.02.1913 / -4.47
Sudan / 9°70'N / 31°40'E / U. e. senegalensis / T / 1915.12.24.2111 / 18.02.1908 / -10.75
Sudan / 13°05'N / 30°21'E / U. e. somaliensis / T / 1922.12.8.458 / 17.01.1921 / -0.16
Sudan / 6°30'N / 31°14'E / U. e. epops / T / 1919.12.17.828 / 19.01.1914 / 45.09
Sudan / 1333′N / 33°36′E / U. e. epops / T / 1919.12.17.825 / 18.12.1913 / -14.62
Sudan / 15°16'N / 26°30'E / U. e. epops / T / 1922.12.8.468 / 18.02.1922 / 9.06
Sudan / 21°55'N / 31°18'E / U. e. epops / T / 1965.M.6187 / 11.02.1928 / -17.50
Sudan / 21°55'N / 31°18'E / U. e. epops / T / 1965.M.6086 / 11.02.1928 / 0.22
Gambia / 13°29'N / 14°11' W / U. somaliensis / T / 1929.2.18.187 / 19.01.1929 / -9.63
Gambia / 13°29'N / 14°11' W / U. epops / T / 1929.2.18.186 / 19.01.1929 / -36.18
Nigeria / 13°28'N / 12°57'E / U. e. africana / T / 1911.12.23.853 / 10.01.1905 / 10.19
Mali / 13°26'N / 6°16' W / U. e. somaliensis / T / 1932.8.6.225 / 30.01.1932 / -10.10
Mali / 13°26'N / 6°16' W / U. e. somaliensis / T / 1932.8.6.226 / 24.01.1932 / -3.24
Mali / 16°46'N / 3°00' W / U. e. somaliensis / T / 1932.8.6.223 / 16.11.1931 / -45.23
Mali / 16°46'N / 3°00' W / U. e. somaliensis / T / 1932.8.6.221 / 18.11.1931 / 6.76
Nigeria / 6°26'N / 7°30'E / U. e. senegalensis / T / 1966.16.147 / 31.03.1954 / -1.43
Senegal / 14°21'N / 16°56' W / U. e. somaliensis / T / 1928.5.2.42 / - / -13.60
Cameroon / 5°54'N / 10°30'E / U. e. senegalensis / T / 1961.9.79 / 10.11.1956 / -27.49
Cameroon / 10°35'N / 14°19'E / U. e. somaliensis / T / 1923.10.26.84 / 06.12.1922 / -9.14
Cameroon / 11°02'N / 14°19'E / U. e. somaliensis / T / 1926.8.8.84 / 28.04.1925 / -28.09
Kenya / 1°00'S / 39°56'E / U. e. africana / T / 1912.12.22.24 / 28.02.1912 / -58.12
Kenya / 0°40'N / 36°06'E / U. africana / T / 1901.2.22.465 / 10.01.1900 / 25.02
Kenya / 1°27'S / 36°58'E / U. africana / T / 1901.2.22.464 / 01.11.1899 / 7.44
Kenya / - / - / U. africana / T / 1923.5.7.5209 / 18.02.1912 / -21.40
Kenya / - / - / U. africana / T / 1935.10.16.165 / 20.02.1935 / -45.44
Somalia / 9°30'N / 44°00'E / U. e. senegalensis / T / 1965.M.6107 / 26.02.1949 / 17.63
Somalia / 10°01'N / 45°12'E / U. e. somaliensis / T / 1923.8.7.5304 / 17.12.1918 / 2.29
Somalia / 6°15'N / 47°58'E / U. epops / T / 1898.6.13.121 / 14.11.1894 / -12.89
Somalia / - / - / U. somaliensis / T / 1898.7.27.166 / 14.11.1897 / -16.72
Somalia / - / - / U. e. epops / T / 1923.8.7.5204 / 11.11.1918 / -17.30
Nigeria / 7°40'N / 11°05'E / U. e. somaliensis / T / 1940.2.8.16 / 27.05.1925 / -15.73
Gambia / 13°38'N / 14°58' W / U. senegalensis / T / NA / - / -72.43
Morocco / 35°45'N / 5°48' W / U. e. epops / T / 1951.13.711 / 23.02.1936 / -69.03
Algeria / 32°06'N / 1°14' W / U. e. epops / T / 1951.13.709 / 16.03.1924 / -20.30
Tanzania / 5°06'S / 30°23'E / U. e. africana / B / 8709 / 01.1937 / -3.40
Tanzania / 6°49'S / 38°52'E / U. e. africana / B / 20224 / 22.01.1962 / 1.18
Mali / 12°52'N / 7°33' W / U. epops / P / CG1962-2684 / 11.1956 / -88.46
Senegal / 15°24'N / 15°24' W / U. epops / P / CG1968-861 / 26.09.1967 / -69.28
Senegal / 14°'48'N / 16°31'E / U. e. senegalensis / P / CG1911-1736 / 01.1909 / -35.89
Senegal / - / - / U. e. senegalensis / P / CG2008-411 / 1836 / -50.07
Mali / 12°17'N / 10°58' W / U. e. senegalensis / P / CG1962-3683 / 01.1959 / -23.37
Chad / 21°34'N / 19°13'E / U. e. senegalensis / P / CG1956-45 / 29.09.1958 / -36.00
Chad / 7°39'N / 15°52'E / U. e. senegalensis / P / CG1977-797 / 10.04.1973 / -14.96
Chad / 12°06'N / 15°04'E / U. e. senegalensis / P / CG1934-1309 / 10.1933 / -18.06
Mali / 14°28'N / 4°12' W / U. e. senegalensis / P / CG1966-439 / 1957 / -34.96
Chad / 12°06'N / 15°04'E / U. e. senegalensis / P / CG1934-1310 / 10.1933 / 32.07
Mali / 14°28'N / 4°12' W / U. e. senegalensis / P / CG1966-439 / 1957 / -4.40
Mali / 15°54'N / 3°55' W / U. e. senegalensis / P / CG1966-441 / 22.04.1957 / 13.34
Chad / 21°34'N / 19°13'E / U. e. senegalensis / P / CG1962-253 / 13.07.1959 / -8.08
1
Appendix S3
Geolocated birds
Fig. S2 Likelihood of moult origin for three hoopoes based upon analysis of stable-hydrogen isotope analyses (δ2Hf) of feathers versus observed wintering locations based on light-level geolocators for(a)geolocator6909, (b) geolocator6915, and (c) geolocator6927. The 50% kernel of the wintering site is indicated in red. These kernels have been revised by S. Hahn et al. (unpubl.), based on new calibration knowledge, and in two birds are now slightly further south than those depicted in Bächler et al. (2010).
1
Appendix S4
δ13C, δ15N and δ2H values for European hoopoes.
Summary of stable isotope (δ13C, δ15N, δ2H) data for European hoopoes. Birds were assigned to east versus west Africa based on analysis of δ2Hf. AD = adults, SY = second-year birds
East / Westn / Mean / SD / Min / Max / n / Mean / SD / Min / Max / F / p
δ13C (‰)
AD ♀ / 21 / -18.6 / 3.8 / -24.4 / -8.8 / 39 / -15.7 / 4.1 / -25.3 / -9.4 / 7.45 / 0.008
AD ♂ / 13 / -18.4 / 3.4 / -24.1 / -10.5 / 52 / -16.3 / 3.1 / -23.9 / -10.6 / 4.98 / 0.029
SY ♀ / 47 / -15.6 / 4.1 / -22.7 / -8.3 / 46 / -16.3 / 3.7 / -22.9 / -8.1 / 0.67 / 0.414
SY ♂ / 27 / -15.7 / 3.8 / -22.9 / -7.8 / 54 / -16.5 / 3.7 / -24.1 / -9.7 / 0.82 / 0.367
δ15N (‰)
AD ♀ / 21 / 11.2 / 2.7 / 7.7 / 16.3 / 39 / 10.4 / 2.3 / 5.2 / 16.4 / 1.36 / 0.248
AD ♂ / 13 / 11.6 / 2.4 / 7.2 / 16.9 / 52 / 10.1 / 2.2 / 3.3 / 16.7 / 4.78 / 0.032
SY ♀ / 47 / 10.5 / 2.8 / 6.1 / 21.3 / 46 / 10.1 / 1.5 / 7.1 / 12.7 / 4.17 / 0.368
SY ♂ / 27 / 9.0 / 2.0 / 6.3 / 13.9 / 54 / 10.2 / 2.2 / 6.5 / 17.7 / 5.68 / 0.020
δ2H(‰)
AD ♀ / 23 / 17.2 / 12.2 / 5.3 / 43.8 / 51 / -14.1 / 12.6 / -48.9 / 3.8 / 100.07 / 0.000
AD ♂ / 14 / 17.1 / 6.7 / 7.4 / 27.8 / 62 / -17.6 / 12.7 / -54.1 / 3.7 / 94.06 / 0.000
SY ♀ / 47 / 20.0 / 10.9 / 5.5 / 62.8 / 46 / -14.3 / 13.5 / -58.5 / 4.3 / 182.38 / 0.000
SY ♂ / 27 / 25.4 / 12.7 / 8.0 / 58.9 / 54 / -15.4 / 16.1 / -67.6 / 4.6 / 131.52 / 0.000
1
Reference List
BächlerE, Hahn S, Schaub M, Arlettaz R, Jenni L, Fox JW, Afanasyev V, Liechti F (2010) Year-round tracking of small trans-Saharan migrants using light-level geolocators. PLoS ONE5: e9566
Bowen GJ, Wassenaar LI, Hobson KA (2005) Global application of stable hydrogen and oxygen isotopes to wildlife forensics. Oecologia143: 337-348
Burnham KP,Anderson DR(2002)Model selection and multimodel inference: a practical information-theoretic approach, 2nd edn. Springer-Verlag, New York
Hobson KA, Lormée H, Van Wilgenburg SL, Wassenaar LI, Boutin JM (2009) Stable isotopes (delta D) delineate the origins and migratory connectivity of harvested animals: the case of European woodpigeons. J Appl Ecol46: 572-581
1