Ecology

Study areaThis study was conducted in Jilin Province, which is located in the center of Northeast China (40°52ʹ N–46°18ʹ N, 121°38′ E–131°19ʹ E) and covers an area of approximately 187,400 km2, with an elevation varying from 5 m to 2,691 m (Fig. 1). The study area has a temperate continental monsoon climate and is climatologically humid, semi-humid, and semi-arid from the southeast to the northwest. The annual mean temperature and annual total precipitation form a southeast-northwest gradient; the eastern part is relatively humid and rainy while the western region is dry in the summer months. Generally, 70–80% of the annual precipitation occurs from June to September, with the most abundant rainfall in the east. The long-term average annual temperature and average annual rainfall are 5.8 °C and 687.0 mm, respectively49. Crop cultivation is mostly concentrated in the black soil region50. The soil types of cultivated lands mainly include black soil, sand, and paddy soil, which are suitable for potato growth.Figure 1Spatial distribution of 51 meteorological stations and soil sampling sites in the study area. Soil data were divided into two categories. Soil samples (I): soil mechanical compositions, involving 81 sampling points; soil samples (II): soil physico-chemical properties, involving 79 sampling points. The map was created using ArcGIS v. 10.4.1 (http://www.esri.com/software/arcgis).Full size imagePotato growth is highly dependent on temperature and light. Jilin Province, as one of the main potato-producing areas in China, possesses sufficient sunlight and exhibits large temperature difference between day and night. Generally, potato cultivation occurs from April to May, depending on the lowest temperature (5 °C), and potatoes are harvested from August to October of the same year. Among potato production areas, mid-late maturing cultivars (e.g., Yanshu No. 4, Atlantic, Jishu No. 1, and Summer) account for about 70%, while early maturing cultivars (e.g., Favorita, Youjin, and Fujin) account for 30%51.DataClimate dataClimate data were obtained from the National Meteorological Information Center, China Meteorological Administration (http://data.cma.cn), including 51 national standard meteorological stations in Jilin Province (Fig. 1). The meteorological data contain daily average temperature, daily maximum temperature, daily minimum temperature, daily sunshine hours, and daily precipitation during 1957–2018. Based on the periods of potato sowing and harvesting in Jilin Province, the climate data between April 1 and September 31 each year were selected. To avoid the impact of extreme weather within a single year on the inter-annual climate change, we used 5-year moving average values of climate data rather than single-year values to establish a geo-climate model using regression analysis and evaluated changes in suitable areas for potato cultivation under the influence of climate change.Topography dataTopography data were extracted from the digital elevation model (DEM) sourced from the geospatial data cloud SRTM (http://www.gscloud.cn). Through a series of processes such as adding X–Y axis, splicing, vector data layering, filtering, cropping, and resampling of raster data on the ArcGIS platform, digitized elevation model (90-m resolution) maps were used to derivate layers such as longitude, latitude, slope, and aspect (Fig. 1).Soil dataSoil mechanical composition data (81 sampling sites) were extracted from the National Science and Technology Infrastructure Platform (http://soil.geodata.cn) and soil physico-chemical property data (79 sampling sites) were provided by the Soil and Fertilizer General Station of Jilin Province (http://www.jltf.cn). The sequence number of the occurrence layer is 1, and the thickness is about 20-50 cm. The soil properties extracted included contents of soil sand, silt, and clay, pH, and contents of nutrients such as organic matter (OM), quick-acting potassium (QAK), available nitrogen (AN), and available phosphorus (AP) (Fig. 1; Tables S1-S2).The soil data were rasterized using kriging. First, the soil mechanical composition data were converted into spherical coordinates, and then ordinary Kriging interpolation was used to spatialize the soil mechanical composition data. co-kriging was used to interpolate spatialize the soil physico-chemical property data. Due to limited soil samples and the lack of a continuous dataset in the study area, the soil data in 2018 were selected as a fixed background for the analysis.Analysis of climatic factorsWe used six climatic factors in this study. Usually, potatoes have has different requirements for light, heat, and water in each growth and development stage. We used average daily temperature during the growth period (ADT/°C, mean of daily average temperature from April 1st to September 30th) and active accumulated temperature ≥ 10 °C (AAT/°C d, sum of active accumulated temperature ≥ 10 °C from April 1st to September 30th) from 1961 to 2018 to reflect the temperature conditions of potato growth52,53,54,55,56. ADT at 14–17 °C was evaluated as “Most suitable”; 10–14 °C or 17–20 °C as “Suitable”; 8–10 °C or 20–24 °C as “Sub-suitable”; < 17 °C or > 24 °C as “Not suitable” for potato growth in the study area. AAT for mid-late maturing varieties at 2000–3000 °Cd was evaluated as “Most suitable”; 1,500–2,000 °Cd or 3,000–6,000 °Cd as “Suitable”; 1,300–1,500 °Cd or 6,000–8,000 °Cd as “Sub-suitable”; < 1,300 °Cd or > 8,000 °Cd as “Not suitable”.The average temperature in July (ATJ, mean of daily average temperature in July) and the day/night temperature difference from July to August (DIF/°C, mean of the day/night temperature difference from July 1st to August 31st) are the key climatic factors for the expansion of potato chunks, which have significant correlation with the meteorological yield of potato 53–57. ATJ at 16–20 °C was evaluated as “Most suitable”; 15–16 °C or 20–24 °C as “Suitable”; 12–15 °C or 24–28 °C as “Sub-suitable”; < 10 °C or > 28 °C as “Not suitable”. DIF at 8–12 °C was evaluated as “Most suitable”; 5–8 °C as “Suitable”; 2–5 °C as “Sub-suitable”; < 2 °C as “Not suitable” in the study area.During the growth and development of potato, there is a great demand for water, especially from the budding stage to the swelling stage of potato growth, which are extremely sensitive to water supply52,53,54,55,56. The total precipitation during the growth period (PP/mm, sum of the daily precipitation from April 1st to September 30th) at 700–900 mm was evaluated as “Most suitable”; 600–700 mm or 900–1,200 mm as “Suitable”; 500–600 mm or 1,200–1,500 mm as “Sub-suitable”; < 500 mm or > 1,500 mm as “‘Not suitable” in the study area.Short daylight and appropriate high temperature during the seedling stage are beneficial to promote potato root development, forming strong seedlings and increasing potato formation52,53,54,55,56. The total sunshine duration during potato growth (SD/hours, sum of the daily sunshine duration from April 1st to September 30th) at 900–1,200 h was evaluated as “Most suitable”; 700–900 h or 1,200–1,500 h as “Suitable”; 400–700 h or 1,500–1,800 h as “Sub-suitable”; < 400 h or >1,800 h as “Not suitable”.MethodsFirst, climatic factors were simulated using geo-climate models. Then, the AHP-PCA model was employed for suitability evaluation, and the satellite-based gridded environmental data were applied for suitability mapping. Finally, the degree of changes in climatic factors and suitable geographic ranges were calculated. These data were interpolated into the surface grid data with a spatial resolution of 0.03° × 0.03° (~3 km × 3 km)57,58. All maps and statistical analyses were generated using ArcGIS 10.4.159 and R 3.6.360.Geo-climate model buildingTopographic factors such as longitude, latitude, and altitude dominate the distribution of climate factors, and directly affect the solar radiation budget and atmospheric circulation, which makes the climate resources to demonstrate obvious spatial differences in both vertical and horizontal directions61,62. Based on the meteorological data and geographic information of each meteorological station, we established geo-climate models and used them to calculate the climate distribution of the study area. The difference between the highest temperature and the lowest temperature from July 1st to August 31st was used to calculate the grid layer of DIF. The relationship between climate zoning indicators and geographic factors is expressed as follows:$$ F = fleft( {lambda ,varphi ,h} right) + varepsilon $$
(1)
where, F is the simulated value of grid point of the climate zoning index; λ, φ, and h represent longitude (°), latitude (°), and altitude (m), respectively; f (λ,φ,h) is called climatological equation of regionalization index; and ε is the influence of local small topography and random factors on climate (i.e., comprehensive geographical residual term).Residual correction : Affected by local topography and random factors, the variation of climatic factors is random, which will cause errors in the calculation of geo-climate models. Therefore, the inverse distance weight (IDW) routine in ArcGIS was used to derive the simulated value of the comprehensive geographical residual term ε raster63. The interpolation calculation formula is:$$varepsilon ={sum }_{i=1}^{n}frac{{varepsilon }_{i}}{{d}_{i}^{k}}/{sum }_{i=1}^{n}{d}_{i}^{k}$$
(2)
where, ε is the simulated value of the grid point of the residual term of climatic factors; (n) is the number of meteorological stations; ({varepsilon }_{i}) is the residual value of the climate factor of the (i)-th meteorological station; ({d}_{i}) is the Euclidean distance between the grid point and the (i)-th meteorological station; k is the power of the distance.AHP-PCA and GIS based suitability analysis for potato cultivationThe suitability map for potato cultivation was generated based on identified criteria that are relevant to the climatic, soil environmental, and geophysical conditions considered. Details of the data analysis procedure, model application, and suitability classification are described as follows.

AHP-PCA model
Analytical Hierarchy Process (AHP) is a multi-criterion decision-based approach developed for analyzing complex decisions involving multiple criteria38,64,65. Principal Component Analysis (PCA) is a multivariate statistical data analysis technique that combines all input variables using a linear combination into a number of principal components that retain the most variance within the original data to identify possible patterns or clusters between objects and variables. In this study, we used AHP to calculate the weight of each zoning indicator in the evaluation index system66,67, and then, we explored the comprehensive relationship of suitability evaluation factors using the grid calculator and PCA tool on the ArcGIS platform. The first principal component will have the greatest variance, the second will show the second most variance not described by the first, and so forth. In most cases, the first three or four raster bands of the resulting multiband raster from principal components tool will describe more than 95% of the variance, that is, the cumulative contribution rate of the principal component reaches more than 95%. The variance of the weighted original data becomes larger, leading to more scientific and reasonable evaluation results. In summary, the proposed approach is achieved as follows (Fig. 2):
Step 1: The weight of each index was calculated by using AHP and consistency test;
Step 2: The indicators were standardized using the Z-Score method;
Step 3: The weights calculated in Step 1 were loaded onto the standardized indicators;
Step 4: A standardized matrix was built and the correlation coefficient matrix was calculated;
Step 5: The principal components was filtered and determined;
Step 6: The score for each principal component was calculated;
Step 7: A comprehensive score for all indicators was obtained.

Establishment of indicator system and calculation of weight
The assessment of climate change impacting suitability of potato cultivation has multiple objectives and levels. This paper combined comprehensive and hierarchical principles, relevant literature reviews38,39,40,68, expert opinions, and characteristics of potato cultivation in Jilin Province to establish an index system for evaluation of ecological environment impact, including 18 evaluation indicators: ADT ( °C d), AAT ( °C), PP (mm), SD (h), ATJ ( °C), DIF ( °C), elevation (m), slope (°), aspect (°), hill shade, sand (%), silt (%), clay (%), OM (g/kg), pH, QAK (mg/kg), AN (mg/kg), and AP (mg/kg). These indicators were classified into three categories: climatic conditions, soil environments, and topography (Table 1).
The weight of each evaluation indicator was determined by AHP. According to relevant literatures and expert opinions, we established a judgment matrix for these evaluation indicators. Pairwise comparison was used for obtaining the relative importance score between different indicators. The consistency of pairwise importance scales is one of the important measurements for successful decision-making by AHP, which could be checked using consistency ratio (CR). If CR < 0.10, the degree of consistency is satisfactory, whereas, CR > 0.10 indicates an inconsistency63,69 (Table 1).

Classification and mapping for suitability of potato cultivation

Figure 2Diagrammatic flow of the Analytical Hierarchy Process (AHP) weighted Principal Component Analysis (PCA) model evaluation process.Full size imageTable 1 Weights of all criteria used for estimating suitability of potato cultivation in the study area.Full size tableThe natural breakpoint method in ArcGIS was employed to classify lands of the study area in terms of cultivation suitability. The study area was delineated into 4 zones: zone 1 (Not suitable), zone 2 (Sub-suitable), zone 3 (Suitable), and zone 4 (Most suitable) (Table 2).Table 2 Dimensionless grading of evaluation values of potato cultivation suitability.Full size tableAfter normalizing all indicators, the cultivation suitability index was established as follows:$$I={sum }_{i}^{n}{W}_{i}{X}_{i}$$
(3)
where I is the suitability index for comprehensive evaluation, ({W}_{i}) is the weight of the indicator, ({X}_{i}) is the value after dimensionless treatment of the indicator, i is the comprehensive evaluation value of topography, climatic conditions, and soil environments. The larger the topography value was converted into a negative value for the calculation as the greater its value, the higher its negative impact on cultivation suitability. Meanwhile, the greater the pH value is, the more unfavorable the comprehensive evaluation of soil will be; the pH value was therefore inversed for the calculation.Trends and fluctuations in changes of climatic factors and suitable areasThe fluctuations of various climatic factors over the past 58 years were analyzed by coefficient of variation (CV), which was calculated as CV = (standard deviation/mean) × 100%. Temporal trends in changes of climatic factors and suitable areas were calculated using ordinary least squares linear regression on annual data from 1961 to 2018. Among them, the trend in suitable area changes was calculated based on each grid. The significance of trends was estimated following a method that considers the temporal autocorrelation by reducing the effective sample size of the time series70. And the significance of temporal trends was tested at P < 0.171. More

Mammalia Linnaeus, 175829.Artiodactyla Owen, 184830.Ruminantia Scopoli, 177731.Infraorder Tragulina Flower, 188332.Family Lophiomerycidae Janis, 198733.Included generaLophiomeryx, Zhailimeryx, Krabimeryx, Chiyoumeryx nov. gen.Genus Krabimeryx Métais, Chaimanee, Jaeger, and Ducroq, 200117.EtymologyKrabi—from Krabi Basin, where the fossils were found, and—meryx is the Greek word for ruminant.Diagnosis [modified after Métais et al.17]Small primitive ruminant with lower molars morphologically close to those of Zhailimeryx. Krabimeryx differs from Zhailimeryx in: more laterally compressed lingual cuspids in the lower molars; an entoconid displaced to anterior with respect to the hypoconid; the lack of both a paraconid and a hypoconulid in m1 and m2; a p4 with a mesolingual conid that is located more posterior and less individualized; a p4 without a distinct posterolingual conid. Krabimeryx differs from Lophiomeryx by less selenodont labial cuspids in the lower molars, the presence of a developed external postmetacristid, and by a distinct groove on the anterior side of the entoconid, the entoconidian groove. Krabimeryx can be distinguished from Iberomeryx in having a well-marked entoconidian groove; the lack of a clear external postprotocristid; the third lobe of m3 not forming a complete buckle; and a more transversely compressed hypoconulid in the m3. Krabimeryx possesses a huge notch in lingual view between the entoconid and the third lobe in the m3.Type speciesKrabimeryx primitivus Métais, Chaimanne, Jaeger, and Ducroq, 200117.Included speciesKrabimeryx gracilis nov. comb. (Miao, 198220).Krabimeryx gracilis nov. comb. (Miao, 198220).Figure 1A and Figure S1.Figure 1Dentition of Krabimeryx gracilis nov. comb. (Miao, 1982)20 (A, B, G, H), Chiyoumeryx nov. gen. shinaoensis (Miao, 1982)20 (C, D), Chiyoumeryx nov. gen. flavimperatoris nov. sp. (E) and Iberomeryx miaoi nov. sp. (F–I). Krabimeryx gracilis nov. comb. (Miao, 1982)20: (A) IVPP V 6546-1 (holotype), partial skull with right and left M1–M3; (B) IVPP V 6546-2 (holotype), right fragmented mandible with m2–m3. Chiyoumeryx nov. gen. shinaoensis (Miao, 1982)20: (C) IVPP V 6531 (holotype), right mandible with p2–m3 and tooth socket of p1; (D) IVPP V 6532 (paratype), right fragmented maxillary with P4-M3. Chiyoumeryx nov. gen. flavimperatoris nov. sp.: (E) IVPP V 6547 (holotype), right mandible with p4–m3; Iberomeryx miaoi nov. sp.: (F) IVPP V 6551 (holotype), left mandible with m1–m3 (mirrored); (G) lower molar Lophiomerycidae dental nomenclature (based on the m3 of IVPP 6546-2): 1 internal postmetacristid, 2 metaconid, 3 external postmetacristid, 4 internal preentocristid, 5 entoconidian groove, 6 external preentocristid, 7 entoconid, 8 posthypoconulidcristid, 9 hypoconulid, 10 prehypoconuldicristid, 11 posthypocristid, 12 hypoconid, 13 prehypocristid, 14 ectostylid, 15 postprotocristid, 16 protoconid, 17 preprotocristid, 18 anterior cingulid; (H) upper molar Lophiomerycidae dental nomenclature (based on the M2 of IVPP 6546-1): 1 postmetacrista, 2 metacone, 3 premetacrista, 4 mesostyle, 5 postparacrista, 6 paracone, 7 paraconid labial groove, 8 preparacrista, 9 parastyle, 10 preprotocrista, 11 anterolingual cingulum, 12 protocone, 13 postprotocrista, 14 entostyle, 15 additional cone, 16 premetaconulecrista, 17 metaconule, 18 postmetaconulecrista; (I) lower molar Tragulidae dental nomenclature (based on the m2 of IVPP V 6551, reversed): 1 metaconid, 2 external postmetacristid, 3 Dorcatherium fold, 4 internal postmetacristid, 5 preentocristid, 6 entoconid, 7 postentocristid, 8 posterior cingulid, 9 posthypocristid, 10 hypoconid, 11 prehypocristid, 12 ectostylid, 13 external postprotocristid, 14 Tragulus fold, 15 internal postprotocristid, 16 protoconid, 17 preprotocristid, 18 paraconid, 19 preparacristid. (J) phylogenetic position and stratigraphie of the Shinao/Yangjiachong/Xiaerhete ruminants (topology2). a stem Ruminantia, b Archaeomeryx, c Chiyoumeryx nov. gen. and Krabimeryx gracilis, d crown Ruminantia, e Iberomeryx miaoi nov. sp.; 1 lingual view, 2 occlusal view. Scale bare is 1 cm.Full size image*v pars1982 Lophiomeryx gracilis—Miao: 532, Table 3, Figs. 6 and 720.v non1982 Lophiomeryx gracilis?—Miao: 536, Fig. 820.v pars1987 L. gracilis—Janis: 21133.v pars1997 L. gracilis—Vislobokova: Fig. 321.v pars2000 L. gracilis—Guo, Dawson, and Beard: 247, Table 214.v pars2001 L. gracilis—Métais, Chaimanee, Jaeger, and Ducroq: 239, 24117.v pars2012 L. gracilis—Mennecart: 6234.NeodiagnosisKrabimeryx gracilis has an m2 that is wider than the m3; this is the other way round in K. primitivus. Moreover, the entoconid is less anterior relative to the hypoconid in K. gracilis than it is in K primitivus. The ectostylid is large in K. gracilis, while it is absent in K. primitivus. The cingulum on the upper molars in K. gracilis is more developed than in K. primitivus.HolotypeIVPP V 6546, partial skull with right and left M1–M3 (IVPP V 6546-1) and an associated right fragmented mandible with m2–m3 (IVPP V 6546-2) found in occlusion with the skull.Additional materialIVPP V 6549, right m3 on fragmented mandible; IVPP V 6550 left fragmented mandible with m1–m2; IVPP V 26638, right m1. Measurements are given in Table S1.LocalitiesShinao Basin, Panxian County, Southwestern Guizhou, China; Xiaerhete locality, Jiminay County, Xingjiang, China. Late Eocene.Taxonomical attributionThe herein described specimens were first attributed to the genus Lophiomeryx20. However, the thorough reassessment of the specimens now leads to the conclusion that Lophiomeryx gracilis sensu Miao20 contains three different species and genera, but none of them can be assigned to Lophiomeryx.Based on the presence of a strong lingual cingulum in upper molars and a short anteroposteriorly oriented postprotocrista, as well as the absence of a premetacristid and an anterior fossa widely open in the lower molars, we can conclude that the specimens, IVPP V 6546-1, IVPP V 6546-2, IVPP V 6549, and IVPP V 6550, belong to Lophiomerycidae or Tragulidae35,36. However, the absence of a large paraconid and the absence of an elongated external postmetacristid distinguish the specimens from primitive Tragulidae17,36. In Zhailimeryx jingweni, the cuspids are more slender than in the herein described specimens14, a feature the taxon shares with K. primitivus. In Z. jingweni, m1 and m2 are of relative similar width14, while in K. primitivus and the herein described specimens from Shinao the m2 is clearly bigger than the m117. Similarly to K. primitivus, the herein described specimens differ from Z. jingweni in its lower molar lingual cusps being more laterally compressed, and in an entoconid that is slightly shifted to anterior with respect to the hypoconid, while it is more posterior in Z. jingweni14,17. Furthermore, K. primitivus and the herein described specimens from Shinao both lack the rudimentary paraconid present in Z. jingweni14,17.Like K. primitivus, the here-described specimens differ from Chiyoumeryx nov. gen. (described below) and the Lophiomeryx species L. mouchelini, L. chalaniati and L. angarae by having more massive and more bunomorph lower molars16,17,24,34,37. Furthermore, Zhailimeryx jingweni, K. primitivus, and the herein described specimens differ from Lophiomeryx by the presence of a developed external postmetacristid and by a distinct entoconidian groove on the anterior side of the compressed entoconid14,17. In Lophiomeryx, the back fossa of m3 is widely open due to the strong reduction of the posthypoconulidcristid34,37. In contrast to this, Krabimeryx primitivus possesses a clearly developed posthypoconulidcristid forming a buckle on the m3 back basin17, similarly to the specimens from Shinao described here.Summing up, the general morphology of the teeth in the herein described specimens is most similar to the one observed in K. primitivus. They both share a similar huge notch in lateral view between the third lobe of m3 and the entoconid and the entoconidian groove, features that clearly distinguishing them both from Lophiomeryx and Zhailimeryx. Thus, we attribute the specimens IVPP V 6546-1, IVPP V 6546-2, IVPP V 6549, and IVPP V 6550 to the genus Krabimeryx. However, significant differences occur with the type species, ruling out the synonymisation of K. gracilis nov. comb. and Krabimeryx primitivus. While both species are very similar in size, K. primitivus has an m3 wider than m2, while it is the converse for K. gracilis nov. comb. Moreover, the entoconid is less shifted to the anterior with respect to the hypoconid in K. gracilis nov. comb. than in K primitivus. There is no ectostylid in K. primitivus, while it is large in K. gracilis nov. comb., forming a transverse cristid between the protoconid and the hypoconid. The cingulum on the upper molars is more developed in K. gracilis nov. comb. than in K. primitivus.Due to these differences we decided to create the new combination Krabimeryx gracilis nov. comb.Chiyoumeryx nov. gen.ZooBank LSIDurn:lsid:zoobank.org:act:464C46E0-5A69-4AC1-A9DD-8A7DF76D5CC0.EtymologyChiyou is a tribe leader of the ancient China, about 5–4 k years ago. Chiyou’s tribe was believed to be in relation with the peoples in southern China; -meryx means ruminant in Greek.DiagnosisChiyoumeryx nov. gen. differs from Zhailimeryx and Krabimeryx notably by the absence of the entoconidian groove. The lower teeth are more laterally compressed in Chiyoumeryx nov. gen. and the metaconid is linguo-labiallly more central than in the two other genera. The posthypoconulidcristid in the lower molars of Chiyoumeryx nov. gen. is longer than in Krabimeryx and its p4 is posteriorly extended, while this part is reduced in Krabimeryx. Chiyoumeryx nov. gen. differs from Lophiomeryx by the shape of the mandible. In Chiyoumeryx nov. gen. there is no diastema between p1 and p2 and the diastema between c and p1 is extremely reduced. The outline of the mandible in occlusal view is relatively straight in this species. Lophiomeryx possesses a long diastema between c and p1 and a small one between p1 and p2, as well as a regularly curved occlusal outline of the corpus. The lower premolars of Chiyoumeryx nov. gen. are laterally compressed giving a more elongated aspect to these teeth than in Lophiomeryx. The trigonid is smaller than the talonid in m1 and m2 in Chiyoumeryx nov. gen. and the preprotocristid terminates centrally and does not reach the lingual side. In Lophiomeryx the trigonid and talonid are of similar size and the preprotocristid is longer and reaches the lingual side. Moreover, in Chiyoumeryx nov. gen., the posthypoconulidcristid is longer than in Lophiomeryx. The shape of the P4 in Chiyoumeryx nov. gen. differs from the one in Lophiomeryx: the posterolingual crista does not meet the posterolabial crista.Type speciesChiyoumeryx nov. gen. shinaoensis (Miao, 198220).Included speciesChiyoumeryx nov. gen. flavimperatoris nov. sp.; ?Chiyoumeryx nov. gen. turgaicus (Flerow 193838).Chiyoumeryx nov. gen. shinaoensis (Miao, 198220).Figure 1B and Figure S2.*v1982 Lophiomeryx shinaoensis—Miao: 530, Table 3, Figs. 3–520.v1987 Lophiomeryx shinaoensis—Janis: 203, 204, 211, 212, Fig. 8B33.v1997 Lophiomeryx shinaoensis—Vislobokova: Fig. 321.v2000 L. shinaoensis—Guo, Dawson, and Beard: 247, Table 214.v2001 L. shinaoensis—Métais, Chaimanee, Jaeger, and Ducroq: 239–241, 24117.v2012 L. shinaoensis—Mennecart: 6234.NeodiagnosisChiyoumeryx nov. gen. shinaoensis is bigger than Chiyoumeryx nov. gen. flavimperatoris nov. sp. but smaller than ?Chiyoumeryx turagicus. The transversely oriented anterior conid in the p4 in Chiyoumeryx nov. gen. shinaoensis differs from the obliquely oriented one in Chiyoumeryx nov. gen. flavimperatoris nov. sp. In Chiyoumeryx nov. gen. shinaoensis, the posterolingual conid is vestigial on p4. Chiyoumeryx nov. gen. shinaoensis has no anterior cingulid, while in Chiyoumeryx nov. gen. flavimperatoris nov. sp. there is a tiny anterior cingulid. Chiyoumeryx nov. gen. shinaoensis possesses lower crowns than ?Chiyoumeryx nov. gen. turgaicus. Chiyoumeryx nov. gen. flavimperatoris nov. sp. possesses an ectostylid, which is absent in ?Chiyoumeryx nov. gen. turgaicus.HolotypeIVPP V 6531, right mandible with p2–m3 and tooth socket of p1.ParatypeIVPP V 6532, right fragmented maxillary with P4–M3.Additional materialIVPP V 6533, right mandible with p2–m3 and tooth socket of i1–p1; IVPP V 6534, left fragments mandible with m1–m3; IVPP V 6535, right fragmented mandible with m1–m3; IVPP V 6536, left fragmented mandible with p4–m3; IVPP V 6537, right fragmented mandible with p4–m2; IVPP V 6538, left p4; IVPP V 6539, right maxillary with P3–M3; IVPP V 6540, right maxillary with P4–M2; IVPP V 6541, right maxillary with M2–M3; IVPP V 6542, left maxillary with P3–M1; IVPP V 6543, right maxillary with M1–M3; IVPP V 6544, Left M3; IVPP V 6545, left maxillary with P4–M3. Measurements are given in Table S1.LocalityShinao Basin, Panxian County, Southwestern Guizhou, China. Late Eocene.Taxonomical attributionMiao20 attributed the here described specimens to the genus Lophiomeryx assuming that these fossils belong to a traguloid. “Lophiomeryx” shinaoensis clearly is a Lophiomerycidae: anterior and posterior fossae are open on the lower molars due to the absence of a premetacristid and the extreme reduction or absence of a postentocristid, there is no external postprotocristid, there is a mesolingual conid on the p4, the symphysis of the mandible extends backward up to the p12,36. It also shares with undisputable Lophiomerycidae a reduced posthypoconulidcristid that does not enclose the third lobe lingually.“Lophiomeryx” shinaoensis differs from Zhailimeryx and Krabimeryx in the absence of the entoconidian groove14,17. Moreover, the teeth are more laterally compressed in “Lophiomeryx” shinaoensis and the metaconid is linguo-labially more centeral14,17. The posthypoconulidcristid in “Lophiomeryx” shinaoensis is more elongated than in Krabimeryx and its p4 has an extended posterior part, while it is reduced in Krabimeryx17.Contrary to what was suggested by Métais and Vislobokova2, Miomeryx altaicus24 is currently known only by its holotype, which is an upper tooth row (AMNH 20383, see Matthew and Granger24). Comparable to M. altaicus, the postprotocrista reaches the premetaconulecrista on the M2 in “Lophiomeryx” shinaoensis. These two cristae fuse totally on the M3 in the here described specimens. However, even if both genera also bear a very strong cingulum, “Lophiomeryx” shinaoensis clearly differs from M. altaicus in having broader and squarer molars and straighter lingual cristae in the P4.Miao20 compared the here revised fossils with the seven Lophiomeryx species considered valid at that time. Unfortunately, very few specimens document most of these species and there is considerable doubt considering the genus attribution of most of them34,36,37,38,39. In any case, we agree with Miao20 (p. 535) that “L. [= Praetragulus] gobiae is readily distinguished from other known Lophiomeryx species as well as from L. shinaoensis by the absence of p1, the anterior flange of metaconid joining protoconid crescent.”. Miao20 (p. 535) already noticed that “Lophiomeryx chalaniati, Lophiomeryx gaudry [= Iberomeryx minor], and Lophiomeryx benarensis are radically different from the present specimens in the anterior branches of the protoconid crescent [= preprotocristid], of m1 and m2 not reaching the lingual border while the posterior branches of hypoconid crescent [= posthypocristid], doing so”. “Lophiomeryx” shinaoensis shares this condition with the Mongolian Lophiomeryx angarae24. However, the trigonid is smaller than the talonid on m1 and m2 in “Lophiomeryx” shinaoensis and the preprotocristid ends in the labio-lingual axis of the molars, while trigonid and talonid are of more similar width combined with a longer preprotocristid in the European Lophiomeryx species and L. angarae16,34,37. The shape of the P4 in “Lophiomeryx” shinaoensis is very different from Lophiomeryx (see Brunet and Sudre37, Figs. 4 and 6). In Lophiomeryx, the posterolingual crista fuses with the posterolabial crista. In “Lophiomeryx” shinaoensis, the curved posterolingual crista does not join the distal end of the posterolabial crista but reaches the labial side. Furthermore, “Lophiomeryx” shinaoensis clearly differs from L. angarae L. mouchelini, and L. chalaniati in the shape of the mandible. These three species of Lophiomeryx possess a very elongated diastema between c and p1 and a small one between p1 and p224,36,37. As part of the genus diagnosis, Mennecart34 (p. 62 and p. 67), adapted from Brunet and Sudre37 and Métais and Vislobokova2, noticed that “the corpus mandibulae presents [in Lophiomeryx: L. angarae, L. mouchelini, and L. chalaniati24,34,37] a concave ventral profile just behind the mandible symphysis, then it becomes regularly convex until the beginning of the ramus, where there is a rounded incisura vasorum. […] On the anterior part of the mandible there are two foramen mentale.” Moreover he wrote that the “p1 is always reduced and leaf-like, separated from c and p2 by diastemata.” (Mennecart34, p. 67). In “Lophiomeryx” shinaoensis there is no diastema between p1 and p2 and the diastema between c and p1 is extremely reduced. The p1 is relatively big considering the root size. The lower outline of the mandible in lateral view is relatively straight. “Lophiomeryx” shinaoensis shares these characteristics with “Lophiomeryx” turgaicus40. Miao20 (p. 535) already noticed strong similarities between “Lophiomeryx” turgaicus and “Lophiomeryx” shinaoensis. The lower premolars of “Lophiomeryx” turgaicus and “Lophiomeryx” shinaoensis are strongly laterally compressed and the p4 is rectangular, giving the lower premolar toothrow an more elongated aspect than in L. angarae, L. mouchelini, and L. chalaniati20,24,30,38,40. Moreover, in these two species, the posthypoconulidcristid is of similar length, longer than in L. angarae, L. mouchelini, and L. chalaniati.Based on these observations, we can assume that “Lophiomeryx” shinaoensis and “Lophiomeryx” turagicus cannot be assigned to the genus Lophiomeryx and may both belong to the same new Lophiomerycidae genus that we here name Chiyoumeryx nov. gen. Chiyoumeryx nov. gen. shinaoensis differs from ?Chiyoumeryx nov. gen. turgaicus nov. comb. in being lower crowned, smaller, possessing an ectostylid, having the symphysis starting under p1, and a shorter diastema.Chiyoumeryx nov. gen. flavimperatoris nov. sp.Figure 1C and Figure S3.v1961 cf. Miomeryx sp.—Xu: 316, 323, 32426.v pars1982 Lophiomeryx gracilis—Miao: 532, Table 3, Fig. 9a,b20.v non1982 Lophiomeryx gracilis?—Miao: 536, Fig. 820.1983 Lophiomeryx sp.—Wang & Zhang: 122, 12741.v1983 cf. Miomeryx sp.—Wang & Zhang: 12341.v1997 Miomeryx sp.—Vislobokova: Fig. 321.v pars1997 L. gracilis—Vislobokova: Fig. 321.v1999 cf. Miomeryx sp.—Zhang, Long, Ji, & Ding: 7, Table 527.v pars2000 L. gracilis—Guo, Dawson, and Beard: 247, Table 214.v pars2001 L. gracilis—Métais, Chaimanee, Jaeger, and Ducrocq: 239, 24117.v2007 Miomeryx sp.—Métais and Vislobokova: 1942.v pars2012 L. gracilis—Mennecart: 6234.ZooBank LSIDurn:lsid:zoobank.org:act:1DF6F58C-F08B-4657-BD4A-7C597653926F.Etymologymeaning yellow (flavor-) emperor (imperatoris) in latin. Chiyou fought with the Yellow Emperor, the ancestor of Chinese, but was defeated.DiagnosisChiyoumeryx nov. gen. flavimperatoris nov. sp. shows the above-mentioned characteristics of the genus. Chiyoumeryx nov. gen. flavimperatoris nov. sp. is smaller than Chiyoumeryx nov. gen. shinaoensis and ?Chiyoumeryx nov. gen. turgaicus. The p4 of Chiyoumeryx nov. gen. flavimperatoris nov. sp. differs from Chiyoumeryx nov. gen. shinaoensis by an oblique anterior conid, which is labio-lingually oriented in the larger species. A very short posterolingual conid is located between the posterolabial cristid and the transverse cristid in the p4 of Chiyoumeryx nov. gen. flavimperatoris nov. sp., while it is absent on Chiyoumeryx nov. gen. shinaoensis. In Chiyoumeryx nov. gen. flavimperatoris nov. sp., there is a tiny anterior cingulid, while it is absent in Chiyoumeryx nov. gen. shinaoensis.HolotypeIVPP V 6547, right mandible with p4–m3 (previously attributed to Lophiomeryx gracilis20).ParatypeIVPP V 6548, left mandible with p4–m3 (previously attributed to Lophiomeryx gracilis20).Additional materialIVPP V 2600, left p4–m2 (previously attributed to cf. Miomeryx sp.26). Measurements are given in Table S1.LocalitiesYangjiachong locality lying in the Caijiachong marls, Qujing, Yunnan, China; Shinao Basin, Panxian County, Southwestern Guizhou, China. Late Eocene.Taxonomical attributionIVPP V 6547 and IVPP V 6548 from Shinao were previously attributed to Lophiomeryx gracilis20, while IVPP V 2600 from Caijiachong marls was first described as cf. Miomeryx sp.26. All these specimens share the same size and dental morphology, and originate from a similar stratigraphic position. That is why we attribute them to the same species.None of these specimens can be attributed to Krabimeryx or Zhailymeryx, as the entoconidian groove is absent14,17. Furthermore, the external postmetacristid is more marked in the considered specimens than in Krabimeryx and Zhailymeryx, forming a deep groove. The third basin is also very different in the here-described specimens from Krabimeryx and Zhailymeryx: the third lobe is a little tilted parallel with the prehypoconulidcristid and posthypoconulidcristid. The back fossa of m3 is very narrow.Furthermore, the here-described specimens can be distinguished from K. gracilis (previously attributed to the same species), by a smaller size and a slenderer shape. The ectostylid is smaller than in K. gracilis. The anterior cingulid in the lower molars is stronger in K. gracilis than in the here-considered specimens. The small postentocristid (especially on m3) of the here-described specimens is absent in K. gracilis.The here-described specimens possess all characteristics in the lower molars that are typical for Chiyoumeryx nov. gen. and distinguish this genus from Lophiomeryx24,34,37. Furthermore, as in Chiyoumeryx nov. gen. shinaoensis, the p4 is laterally compressed giving it a more elongated aspect than in Lophiomeryx24,34,37. Therefore, we consider it justified assigning the here-described specimens to Chiyoumeryx nov. gen. However, they differ from Chiyoumeryx nov. gen. shinaoensis in as smaller size and the morphology of the p4: (1) the anterior conid is oblique while it is labio-lingually oriented in Chiyoumeryx nov. gen. shinaoensis. (2) There is a tiny anterior cingulid that is absent in Chiyoumeryx nov. gen. shinaoensis. (3) There is no additional cristid on the mesolingual conid, which is a well-rounded conid, while in Chiyoumeryx nov. gen. shinaoensis, there is a short posterolingual cristid. (4) The posterolingual conid stands between the posterolabial cristid and the transverse cristid, while in Chiyoumeryx nov. gen. shinaoensis, the posterolingual conid is very small and oblique between the transverse cristid and the posterior stylid and does not join the posterolabial cristid. Due to these distinct differences we erect a new species: Chiyoumeryx nov. gen. flavimperatoris nov. sp.Family Tragulidae Milne-Edwards, 186442.Genus Iberomeryx Gabunia, 196443.Diagnosis (modified from Mennecart et al.36)Small-sized ruminant with upper molars possessing the following combination of characters: well-marked parastyle and mesostyle in small-column shape; strong paracone rib; metacone rib absent; metastyle absent; unaligned external walls of metacone and paracone; strong postprotocrista stopping against the anterior side of the premetaconulecrista; continuous lingual cingulum, stronger under the protocone. Lower dental formula is primitive (3–1–4–3) with non-molarized premolars. Tooth c is adjacent to i3. Tooth p1 is single-rooted, reduced and separated from c and p2 by a short diastema. The premolars have a well-developed anterior conid. Teeth p2–p3 display a distally bifurcated mesolabial conid. Tooth p3 is the largest premolar. Tooth p4 displays no mesolingual conid and a large posterior valley. Regarding the lower molars, the trigonid and talonid are lingually open with a trigonid more tapered than the talonid. The anterior fossa is open, due to a forward orientation of the preprotocristid and the presence of a paraconid. The internal postprotocristid is oblique and the external postprotocristid reaches the prehypocristid. The internal postprotocristid, postmetacristid and preentocristid are fused and Y-shaped. Protoconid and metaconid display a weak Tragulus fold and a well-developed Dorcatherium fold, respectively. The mandible displays a regularly concave ventral profile in lateral view, a marked incisura vasorum, a strong mandibular angular process, a vertical ramus, and a stout condylar process.Type speciesIberomeryx parvus Gabunia, 196443 from Benara (Georgia), late Oligocene44.Included speciesI. minor45, Iberomeryx miaoi nov. sp.Iberomeryx miaoi nov. sp.Figure 1D and Figure S4.v 1982 Lophiomeryx gracilis?—Miao: 536, Fig. 820.ZooBank LSIDurn:lsid:zoobank.org:act:EE3F88E9-0EAF-4EC6-A46F-8623241E614B.DiagnosisIberomeryx with a very large paraconid, which is smaller in Iberomeryx minor and Iberomeryx parvus. The metastylid is not strong but is more developed than in the other species. The ectostylid is big on m1, smaller on m2 and absent on m3, while I. minor displays an ectostylid on all molars and I. parvus none at all. Iberomeryx miaoi nov. sp. is of similar size to I. minor and its m2 is smaller than the one of I. parvus. It differs from I. minor by a thin anterior cingulid. Moreover, its protoconid is positioned slightly more anterior than in I. parvus. The molars appear to be more massive and bulkier in this species than in I. minor and I. parvus.HolotypeIVPP V 6551, left mandible with m1–m3 (only specimen known). m1 5.1 × 3.5, m2 5.2 × 4.1, m3 8.0 × 4.0.EtymologyWe dedicate this species to Prof. Miao Desui who was the first to describe the Shinao fauna.Locality and horizonShinao Basin, Panxian County, Southwestern Guizhou, China. Late Eocene.Taxonomical attributionThis minute ruminant was referred to Lophiomeryx gracilis? by Miao20. However, he already noticed that the size of this individual was smaller than in the other specimens attributed to Lophiomeryx gracilis. Miao20 excluded an attribution of IVPP V 6551 to “Lophiomeryx” gaudryi due to a closed posterior section of the posterior fossa on the m3. However, in both teeth, the posterior fossa is still open by the reduction of the postentocristid.The here-described specimen clearly differs from Lophiomeryx by the presence of an external postmetacristid forming a slight Dorcatherium fold, a developed external postprotocristid (clearly visible at least on m2), and a large paraconid36. Furthermore the external postprotocristid and prehypocristid are connected on their distal ends and the third basin of m3 forms a well-formed buckle, unlike the condition in Lophiomerycidae14,16,33,36,37. The combination of these characters is typical for Tragulidae36.Very few taxa are so far known in the early evolution of the Tragulidae. Only Archaeotragulus, Iberomeryx, and Nalameryx are recognized as potential Paleogene Tragulidae17,36,46, of which Archaeotragulus is currently the oldest representative described17,47. Archaeotragulus possesses lower molars with a broadened talonid in comparison to the trigonid and displays an entoconidian groove36. In the case of IVPP V 6551, the trigonid and talonid are of similar size and no specific entoconidian groove can be observed. Mennecart et al.36 considered Nalameryx a Tragulidae notably based on the presence of the M structure (the external postmetacristid, the internal postmetacristid, the internal postprotocristid, and the external postprotocristid are interconnected forming a M in occlusal view), including the Tragulus fold and Dorcatherium fold, and the absence of a rounded mesolingual conid in the p435. IVPP V 6551 differs from Nalameryx in having an m3 wider than m1 and similar m1 and m2 widths17. In size proportions and molar morphology, IVPP V 6551 resembles the genus Iberomeryx. In IVPP V 6551, the relative size of the m2 is more similar to I. minor. In Iberomeryx minor, the anterior cingulid is big36,46, while in Iberomeryx parvus the cingulid is thin48 like in IVPP V 6551. The teeth of IVPP V 6551 appear to be more massive and bulkier than in I. minor and I. parvus36,48. Similarly to I. minor, the protoconid of IVPP V 6551 is a little more anterior than in I. parvus36,48. IVPP V 6551 clearly differs from I. parvus and I. minor by the presence of a very large paraconid, which is smaller in the two other species36,48. Moreover, the metastylid in IVPP V 6551 is slightly more developed than in I. minor and not present in I. parvus43,48. Iberomeryx minor displays an ectostylid on all molars36, while this structure is absent from I. parvus48. The ectostylid in IVPP V 6551 is large on m1 to absent on m3. Based on these differences we decided to erect the new species Iberomeryx miaoi nov. sp.Origin of crown Ruminantia and dispersal pattern of Paleogene Eurasian ruminantsSo far five families and 13 genera of Ruminantia are known during the middle and late Eocene in Eurasia2,18,19. Based on molecular data, the origin of crown ruminants should be searched for between the latest late Paleocene (56.5 Ma) and the latest early Oligocene (29 Ma)49,50. With the description of stem Tragulidae from the early Oligocene of Western Europe (Iberomeryx) and the late Eocene from Southern Thailand (Archaeotragulus)17, Mennecart et al.26 and Mennecart and Métais51 verified that the oldest crown ruminants date back at least to the latest Eocene (34 Mya). The presence of the tragulid genus Iberomeryx in Shinao, Southern China, further confirms this and may actually represent the oldest fossil of a Tragulidae known and thus of a crown Ruminantia (37–35 Mya, Fig. 1), since no Pecora is known during the Eocene so far51.The here presented reassesment of the Shinao ruminants in combination with literature data reveals a clear pattern in the distribution of Eocene ruminants. Among Archaeomerycidae, Archaeomeryx and Miomeryx are found in Northern and Central Asia [Kazhakstan, Mongolia, and northern part of China2,21,53 (see Fig. 2)]. The lophiomerycid Lophiomeryx (as Lophiomeryx angarae) as well as the Asiatic Praetragulidae (Praetragulus) occupy the same area2. The Mongolian Lophiomeryx angarae is most likely closely related to the European species Lophiomeryx mouchelini. Due to the strong morphological similarities, some specimens of L. mouchelini were actually first described as Lophiomeryx cf. angarae54. Lophiomeryx mouchelini or its ancestors arrived in Europe with the Grande-Coupure dispersal event at the Eocene–Oligocene transition ca. 34 Mya ago (oldest European records: Calaf, Spain, MP22; Möhren 9, Germany, MP21-22; age comprised between the German localities Haag2 MP21 and Möhren 13 MP2234,37,53). The close relationship of these European and the Mongolian species confirms that the origin of the Grande-Coupure cohort may be deeply anchored in the Eocene of Central-Northern Asia (Fig. 2).Figure 2Paleobiogeography of the Eurasiatic ruminants during the Eocene at the genus level. The localities are from the synthesis of data2,17,18,22,49. The palinspastic map is modified from Scotese52.Full size imageThe Southern part of Asia presents a totally different ruminant community at the genus level and includes the Archaeomerycidae Indomeryx and Notomeryx, the Lophiomerycidae Krabimeryx and Chiyoumeryx nov. gen., the Bachitheriidae Bachitherium and the Tragulidae Archaetrogulus and Iberomeryx2,17,18,19,21,53 (see Fig. 2). The oldest Bachitherium is currently known from the Balkan area during the Eocene18,19. The Tethys Ocean separated this area from Western Europe until its progressive disappearance during the Oligocene, ca. 31 Mya55,56. Bachitherium and a cohort of rodents (Pseudocricetodon, Paracricetodon, and the Melissodontinae)19 did not reach Western Europe prior to the opening of this passage. Similarly to the genus Bachitherium, Iberomeryx arrived in Western Europe after the drying out of the Tethys Ocean ca. 31 Mya, during the Bachitherium dispersal event18,19,36. Iberomeryx is mainly known from the middle early Oligocene of Western Europe34,36,57 and the late Oligocene of Anatolia and Georgia43,48,58. Discovering Iberomeryx in the Eocene of Eastern Asia confirms an Asiatic origin of this genus. The close relationship between South-eastern Europe and South-eastern Asia is furthermore supported by anthracotheriids (extinct artiodactyls related to hippopotamids) and rhinocerotoids59,60.Mennecart et al.18,19 proposed that mammals originating from Asia arrived in Western Europe during the early Oligocene in two faunal events: the Grande-Coupure, ca. 33.9 Mya and the Bachitherium dispersal event, ca. 31 Mya. These two faunal events imply two different and diachronous ways of dispersal. The fact that Eocene taxa from South-eastern Asia did not arrive in Western Europe prior to 31 Mya indicates that the Bachitherium dispersal Event cohort might be deeply anchored in the Eocene of Southern Asia (Fig. 2), while genera recorded from the Eocene of Central Asia are known to have arrived already during the Grande-Coupure and thus originated from a different palaeobiogeographic province. The Grande-Coupure was a dispersal event using a Northern way over the closed Turgai Strait and probably originating from Central Asia (Fig. 2). The Bachitherium dispersal event is a stepwise story with a first dispersion from Southern Asia to South-eastern Europe along the Southern path (Fig. 2) and then the dispersal throughout Europe thanks to the closure of the Tethyian Ocean18,19.The south-eastern part of Asia has shown very few changes from a warm and humid climate and environment since the Eocene4, while Northern Asia underwent a transition from warm and humid subtropical environments during the Eocene to steppe environments in the Pliocene, e.g.3,4,5. In this light it is not surprising that an increasing number of paleontological and geological studies indicate that Asia had already experienced a strong latitudinal environmental zonation during the middle and the late Eocene, e.g.6,13.These different climatic and environmental conditions in Central and South Asia led to two distinct palaeobiogeographical provinces clearly traceable in assemblages of herbivores like ruminants that was already apparent during the Eocene. The Central Asian ruminants were living in a more arid environment than the ones from South-eastern Asia (see Fig. 2). The tropical and wet environments from the South-eastern Asia led to the emergence of the Tragulidae (Iberomeryx and Archaeotragulus) and of the anthracotheriids. More

Download PDF

As a wildlife-conservation biologist studying climate change, I want to understand the evolving environment through the eyes of large animals. My work — usually in cold, remote places — involves finding animals, and ways to eat, sleep and be warm. I might be miserable, but I get insights that others cannot into what animals are doing.For about 15 years I’ve been interested in musk oxen (Ovibos moschatus), social herd animals that roamed with woolly mammoths. This picture was taken on Wrangel Island, off the northeast coast of Russia, when I was studying how musk oxen react to polar bears. Because polar ice is melting, more polar bears are hunting on land, and they’re known to have killed musk oxen. These herd animals typically don’t flee from predators such as grizzly bears. They tend to form huddles instead, and male musk oxen have killed grizzlies. Would they try to kill polar bears, too?To find out, I dressed as a polar bear, pulling a bear head on and placing a cape over a range finder, camera and data books. I was cold and nervous. I didn’t want to be killed by a charging musk ox — or by anything else. If some oxen charged, I’d throw off my costume and stand up straight, as I’m doing here; so far, that had stopped them. I’d also encountered a female polar bear with newborn cubs, but she’d left me alone. This picture is from the end of a session, and I’d lived another day. Whew!I learnt that musk oxen are more likely to flee from polar bears than from grizzlies. But during this trip to Russia, I was arrested — over a date error on my permits. In court, the only word I understood was ‘CIA’. I was let go, but banned from returning for three years, so I’m now studying the huemul (Hippocamelus bisulcus), an endangered species of deer that lives in the shadows of glaciers at the tip of South America. As glaciers recede, how will huemul populations respond?

Nature 597, 296 (2021)
doi: https://doi.org/10.1038/d41586-021-02429-2

Related Articles

Tracking Chernobyl’s effects on wildlife

Chasing bats at dawn

To look after these birds is to ‘fall in love’ with them

Subjects

Careers

Climate change

Conservation biology

Latest on:

Careers

Diversity in science workforce an ‘economic imperative’
Career News 02 SEP 21

What makes us tick: lab leaders describe their research philosophies
Career Guide 01 SEP 21

In memory of a game-changing haematologist
Correspondence 31 AUG 21

Climate change

Freak US winters linked to Arctic warming
News 03 SEP 21

Policy, drought and fires combine to affect biodiversity in the Amazon basin
News & Views 01 SEP 21

The contribution of insects to global forest deadwood decomposition
Article 01 SEP 21

Jobs

PhD Project – High recovery and chemical-free desalination using advanced electrodialysis schemes

Wetsus Centre of Excellence for Sustainable Water Technology
Leeuwarden, Netherlands

PhD Project – Beyond chlorine: alternative sustainable compounds to remove biofilms in drinking water environments

Wetsus Centre of Excellence for Sustainable Water Technology
Leeuwarden, Netherlands

PhD Project – Sensing strategies for real-time, early warning monitoring of biofilm formation parameters in drinking water distribution systems

Wetsus Centre of Excellence for Sustainable Water Technology
Leeuwarden, Netherlands

Enhancing the local water cycle via evaporation for a sustainable water supply

Wetsus Centre of Excellence for Sustainable Water Technology
Leeuwarden, Netherlands

Nature Briefing
An essential round-up of science news, opinion and analysis, delivered to your inbox every weekday.

Email address

Yes! Sign me up to receive the daily Nature Briefing email. I agree my information will be processed in accordance with the Nature and Springer Nature Limited Privacy Policy.

Sign up More

We focused our research on countries/territories in Asia (specifically, Hong Kong SAR, Japan, Myanmar, Thailand and Vietnam) because COVID-19 had not spread much outside Asia at the time of data collection and the global effects were predominantly concentrated in East and Southeast Asia. Our five survey countries/territories were chosen because they all have relatively high levels of wildlife trade but also represent very different forms of trade (for example, the pet trade in Japan versus the wild-meat trade in Vietnam). Surveying respondents from markets with these different forms of trade thus allowed an examination of how the full variety of wildlife consumption types may be impacted by perceived disease risk. Budgetary constraints precluded the inclusion of further countries, although we believe those that were surveyed provide a valid snapshot of the main regional issues and patterns. The exception to this may be the exclusion of China, a key global player in the wildlife trade and the possible origin of the COVID-19 virus. Conducting research in China requires an extensive process to obtain permission that was not consistent with the opportunistic nature of our survey, which was mobilized quickly to target opinions from a snapshot view of an (at that time) emerging disease. Given the time-sensitive nature of the research, we were therefore unable to wait for the necessary permissions to include China in this survey.Our online survey was conducted between March 3–11, 2020 and surveyed 1,000 respondents in each of the five target countries/territories. We designed and translated our questionnaires with local experts to ensure questions were culturally appropriate, understandable and relevant. The survey was a quantitative data collection instrument that comprised 32 questions, lasted on average 8 minutes, and respondents were offered an incentive for participating. Respondents aged 18+ were invited via email from an online panel of over 2.5 million people in the target countries/territories, and could answer on any internet-capable device (for example smartphone, tablet, laptop) at their convenience. Only respondents aged 18 and over were eligible to take the survey, which was entirely voluntary. Any respondents working in advertising, public relations, marketing, market research or media industries were screened out to prevent possible bias. The email invite that was sent to participants did not specify the exact nature of the survey to avoid skewing the participants towards those that believed they know about the topic. Instead, the invite indicated that the questions would be about ‘consumption and shopping habits’. The panel is maintained by Toluna (https://tolunacorporate.com/), an online data collection group focused on providing high-quality market research data to clients in various business and non-business sectors. Toluna builds and maintains large online consumer panels to collect these data while adhering to stringent global and local guidelines for panel management and data quality, and is a member of the European Society for Opinion and Market Research (https://www.esomar.org).Toluna respects privacy and is committed to protecting personal data. Their privacy policy (https://tolunacorporate.com/legal/privacy-policy/) provides information on how Toluna collects and processes personal data, explains privacy rights and gives an overview of applicable legislation protecting the handling of personal information. Toluna only uses personal data when the law allows the data to be used.Respondents were asked demographic questions, and quotas based on the most recent census data for each country/territory were used to ensure the final sample profile was nationally representative of age and gender, except in Myanmar where internet access skewed online panel members to a younger male demographic. Specifically, participants were excluded once quotas on age and gender were filled, and again, participants working in advertising/public relations, marketing research or media were excluded from the survey as we believed these jobs could influence responses. Respondents were asked about societal, economic and environmental concerns, their perception of COVID-19 and their attitudes towards wildlife and wildlife consumption (Supplementary Methods). We also excluded respondents who stated that they were unsure whether they or anyone in their social circle had recently purchased wildlife products (n = 421), as well as an additional n = 39 respondents who were unable to answer survey questions that were later included as covariates in our models.Because of the potentially sensitive nature of wildlife consumption, we asked about past wildlife purchases indirectly, questioning respondents on whether anyone within their social circle, including themselves, had recently purchased wildlife products. Indirect questions can improve answer rates for questions that people may feel uncomfortable about answering honestly27. During the pandemic, respondents may have felt uncomfortable about revealing wildlife purchases, given links between wildlife consumption and COVID-19. Additionally, although most wildlife consumption is legal (with restrictions) in the markets surveyed, some is not, and researchers can be perceived as having interests contrary to that of the respondent. For less-sensitive questions on future wildlife consumption and changes in consumption resulting from COVID-19, we asked respondents for their own response rather than that of their social group.Previous studies have found a high correlation between an individual’s admission of using a wildlife product and their likelihood of being within a network of individuals who buy such products28, and suggested that this is linked to homophily in social networks, especially in Southeast Asia. The homophily principle states that people’s personal networks are homogeneous with regard to many socio-demographic, behavioural and intrapersonal characteristics29. Research on wildlife consumption in other Southeast Asian contexts suggests that social groups can be a motivator to begin or maintain consumption of wildlife products28,30. Our own previous research supports this, indicating a strong correlation between one’s own tiger and ivory purchases and knowing someone within one’s social circle who has purchased such products. Additionally and recognizing the homophily principle, behaviour change campaigns targeted at social networks rather than individuals per se are likely to achieve better results than non-targeted campaigns. Changing perceptions of acceptability is a key aspect of social marketing and is used in the social mobilization domain of social and behaviour change communications, which has become a popular framework for reducing demand for illegally traded wildlife products31. Influencing people within a wildlife consumer’s social network may therefore have a higher rate of efficacy than attempting to influence the perceptions of individuals who do not know any consumers of wildlife.We used hierarchical Bayesian regression models to assess relationships between socio-demographic explanators and our three response variables: (1) self-reported recent wildlife consumption, (2) change in wildlife consumption as a result of COVID-19 and (3) anticipated future wildlife consumption. Explanatory variables included 22 non-collinear variables in six categories: basic demographics, awareness and level of worry of COVID-19, COVID-19 personal impacts, support for and effectiveness of wildlife market closures, international travel habits and general attitudes towards global issues (Supplementary Table 1). Aside from household income (measured in US dollars per year), age (midpoint of year categories from the survey question) and education (ordinal, reflecting increasing level of schooling), all other variables were categorical; those with more than two categories were collapsed into dummy variables. Income, age and education were standardized and included to investigate whether a person’s general socio-economic status affects wildlife consumption. General attitudes towards global issues were expected to reflect aspects of respondents’ political tendencies, while travel habits were included to test the hypothesis that those who travel internationally more habitually are, and will be, more frequent consumers of wildlife. Questions regarding awareness and impacts of COVID-19, and concern about future disease epidemics, were asked to determine how the pandemic may be shaping wildlife consumption. Finally, support and perceived effectiveness of wildlife market closures were included as predictor variables since this measure has been suggested as a strong policy lever to reduce wildlife consumption.The general structure of all three models was as follows:$$y_{ij}sim {{{mathrm{Bernoulli}}}}left( {theta _{ij}} right)$$
(1)
$${mathrm{logit}}left( theta right) = alpha + {{u}_1} + {beta} {mathbf{X}} + {{u}_2}{mathbf{Z}}$$
(2)
This model allowed both coefficients and intercepts to vary across countries (that is, a ‘random-slope random-intercept’ model). In equation (1), yij is whether or not individual i in country j reported wildlife consumption, modelled as a Bernoulli trial with probability θij. The logit transformation of θ (equation 2) is a linear function of parameters α and u1 (the fixed intercept term and a vector of the country-specific intercept terms, respectively), as well as a vector of fixed regression coefficients β and a vector of country-specific regression coefficients u2, with X and Z being the corresponding design matrices32. For α and β, we used an improper flat prior over the real numbers, while the group level parameters u1 and u2 were assumed to arise from a multivariate normal distribution with mean 0 and unknown covariance matrix. The covariance matrix was parameterized by a correlation matrix having a Lewandowski–Kurowicka–Joe prior, and a standard deviation with half-Student t prior with three degrees of freedom32.For the three dependent variables, we evaluated the predictive power of a model containing all 22 variables, as well as six subset models, using Watanabe–Akaike Information Criterion and leave-one-out cross-validation33. Each of these six subset models contained all explanatory variables except for those within one of the six categories described above (for example, all explanatory variables except those relating to international travel habits, all explanatory variables except those relating to support for wildlife market closures). We used this model-comparison approach to test whether any of these categories of explanatory variable were more or less important in explaining wildlife consumption; if particular categories of variable are stronger predictors of wildlife consumption, this could help inform where future conservation interventions should focus on. Watanabe–Akaike Information Criterion and leave-one-out cross-validation are both measures of model predictive accuracy (both use log predictive density as the utility function or comparison metric) and have been suggested as useful metrics for Bayesian model selection33. We interpreted variable coefficients whose 95% Bayesian credible intervals did not contain 0 as providing strong evidence for the impact of that variable on the outcome in each of the three models for self-reported wildlife consumption (that is, recent, future and changes due to COVID-19). Models were estimated using the R statistical computing software34, in particular the package brms32, with four chains of 1,000 iterations each, a 500-iteration warm-up period, and with successful convergence verified by confirming that R-hat statistical values were less than or equal to 1.01 (ref. 22).We used the Bayesian hierarchical model of anticipated future wildlife consumption and generated predicted probabilities of future consumption for our sample population (Fig. 2, grey bars). We then predicted future consumption probabilities for a hypothetical behaviour-change intervention (Fig. 2, coloured bars). This intervention was simulated by setting the ‘medical impact’ variable to zero for all individuals, and by assigning all individuals into the ‘aware lots’ and ‘support very likely’ categories for questions related to level of awareness of COVID-19 and level of support for government closure of domestic wildlife markets, respectively. All other variables for individuals were held at the levels recorded in the surveys. We considered the difference between these two predicted probabilities as the impact of the hypothetical behaviour-change intervention, which we examined at the level of the country/territory and within education, age, income and gender demographic classes. Strong evidence for the effectiveness of this hypothetical intervention among countries and demographic classes was suggested where Bayesian credible intervals around the mean predicted difference were less than zero (Supplementary Table 3).Reporting SummaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article. More

Field samplingWe collected samples once per week over four weeks at the Antarctic sea ice edge, in McMurdo Sound, Antarctica (December 28, 2014 “GOS-927”; January 6 “GOS-930”, 15 “GOS-933”, and 22 “GOS-935”, 2015; as previously described in [27]). Sea water (150–250 l) was pumped sequentially through three filters of decreasing size (3.0, 0.8, and 0.1 μm, 293 mm Supor filters). Separate filter sets were acquired for metagenomic, metatranscriptomic, and metaproteomic analyses, over the course of ∼3 h, each week (36 filters in total). Filters for nucleic acid analyses were preserved with a sucrose-based buffer (20 mM EDTA, 400 mM NaCl, 0.75 M sucrose, 50 mM Tris-HCl, pH 8.0) with RNAlater (Life Technologies, Inc.). Filters for protein analysis were preserved in the same sucrose-based buffer but without RNAlater. Filters were flash frozen in liquid nitrogen in the field and subsequently stored at −80 °C until processed in the laboratory.Metagenomic and metatranscriptomic sequencingWe used metagenomics and metatranscriptomics to obtain reference databases of potential proteins for metaproteomics. We additionally used a database assembled from a similarly processed metatranscriptomic incubation experiment [28], conducted with source water from the January 15, 2015 time point (these samples were collected on a 0.2 μm Sterivex filter and processed as previously described).For samples from the GOS-927, GOS-930, GOS-933, and GOS-935 filters, RNA was purified from a DNA and RNA mixture [29]. In total, 2 µg of the DNA and RNA mixture was treated with 1 µl of DNase (2 U/µl; Turbo DNase, TURBO DNase, Thermo Fisher Scientific), followed by processing with an RNA Clean and Concentrator kit (Zymo Research). An Agilent TapeStation 2200 was used to observe and verify the quality of RNA. In total, 200 ng of total RNA was used as input for rRNA removal using Ribo-Zero (Illumina) with a mixture of plant, bacterial, and human/mouse/rat Removal Solution in a ratio of 2:1:1. An Agilent TapeStation 2200 was used to subsequently observe and verify the quality of rRNA removal from total RNA. rRNA-deplete total RNA was used for cDNA synthesis with the Ovation RNA-Seq System V2 (TECAN, Redwood City, USA). DNA was extracted for metagenomics from the field samples (GOS-927, GOS-930, GOS-933, and GOS-935) according to [29]. RNase digestion was performed with 10 µl of RNase A (20 mg/ml) and 6.8 μl of RNase T1 (1000 U/µl), which were added to 2 µg of genomic DNA and RNA mixture in a total volume of 100 µl, followed by 1 h incubation at 37 °C and subsequent ethanol precipitation in −20 °C overnight.Samples of double stranded cDNA and DNA were fragmented using a Covaries E210 system with the target size of 400 bp. In total, 100 ng of fragmented cDNA or DNA was used as input into the Ovation Ultralow System V2 (TECAN, Redwood City, USA), following the manufacturer’s protocol. Ampure XP beads (Beckman Coulter) were used for final library purification. Library quality was analyzed on a 2200 TapeStation System with Agilent High Sensitivity DNA 1000 ScreenTape System (Agilent Technologies, Santa Clara, CA, USA). Twelve DNA and 18 cDNA libraries were combined into two pools with concentration 4.93 and 4.85 ng/µl, respectively. Resulting library pools were subjected to one lane of 150 bp paired-end HiSeq 4000 sequencing (Illumina). Prior to sequencing, each library was spiked with 1% PhiX (Illumina) control library. Each lane of sequencing resulted in between 106,000 and 111,000 Mbp total and 6900–12,000 Mbp and 4800–6900 Mbp for individual DNA or cDNA libraries, respectively.Metagenomic and metatranscriptomic bioinformaticsMetagenomic and metatranscriptomic data were annotated with the same pipelines. Briefly, adapter and primer sequences were filtered out from the paired reads, and then reads were quality trimmed to Phred33. rRNA reads were identified and removed with riboPicker [30]. We then assembled reads into transcript contigs using CLC Assembly Cell, and then we used FragGeneScan to predict open reading frames (ORFs) [31]. ORFs were functionally annotated using Hidden Markov models and blastp against PhyloDB [32]. Annotations which had low mapping coverage were filtered out (less than 50 reads total over all samples), as were proteins with no blastp hits and no known domains. For each ORF, we assigned a taxonomic affiliation based on Lineage Probability Index taxonomy [32, 33]. Taxa were assigned using two different reference databases: NCBI nt and PhyloDB [32]. Unless otherwise specified, we used taxonomic assignments from PhyloDB, because of the good representation of diverse marine microbial taxa.ORFs were clustered by sequence similarity using Markov clustering (MCL) [34]. Sequences were assigned MCL clusters by first running blastp for all sequences against each other, where the query was the same as the database. The MCL algorithm was subsequently used with the input as the matrix of E-values from the blastp output, with default parameters for the MCL clustering. MCL clusters were then assigned consensus annotations based on KEGG, KO, KOG, KOG class, Pfam, TIGRFAM, EC, GO, annotation enrichment [28, 32, 35,36,37,38,39]. Proteins were assigned to coarse-grained protein pools (ribosomal and photosynthetic proteins) based on these annotations. For assignment, we used a greedy approach, such that a protein was assigned a coarse-grained pool if at least one of these annotation descriptions matched our search strings (we also manually examined the coarse grains to ensure there were no peptides that mapped to multiple coarse-grained pools). For photosynthetic proteins, we included light harvesting proteins, chlorophyll a-b binding proteins, photosystems, plastocyanin, and flavodoxin. For ribosomal proteins, we just included the term “ribosom*” (where the * represents a wildcard character), and excluded proteins responsible for ribosomal synthesis.Sample preparation and LC-MS/MSWe extracted proteins from the samples by first performing a buffer exchange from the sucrose-buffer to an SDS-based extraction buffer, after which proteins were extracted from each filter individually (as previously described) [27]. After extraction and acetone-based precipitation, we prepared samples for liquid chromatography tandem mass spectrometry (LC-MS/MS). Precipitated protein was first resuspended in urea (100 µl, 8 M), after which we measured the protein concentration in each sample (Pierce BCA Protein Assay Kit). We then reduced, alkylated, and enzymatically digested the proteins: first with 10 µl of 0.5 M dithiothreitol for reduction (incubated at 60 °C for 30 min), then with 20 µl of 0.7 M iodoacetamide (in the dark for 30 min), diluted with ammonium bicarbonate (50 mM), and finally digested with trypsin (1:50 trypsin:sample protein). Samples were then acidified and desalted using C-18 columns (described in detail in ref. [40]).To characterize each metaproteomic sample, we employed one-dimensional liquid chromatography coupled to the mass spectrometer (VelosPRO Orbitrap, Thermo Fisher Scientific, San Jose, California, USA; detailed in [40]). For each injection, protein concentrations were equivalent across sample weeks, but different across filter sizes. We had higher amounts of protein on the largest filter size (3.0 μm) and less on the smaller filters, so we performed three replicate injections per 3.0 µm filter sample, and two replicate filter injections for 0.8 and 0.1 µm filters. We used a non-linear LC gradient totaling 125 min. For separation, peptides eluted through a 75 µm by 30 cm column (New Objective, Woburn, MA), which was self-packed with 4 µm, 90 A, Proteo C18 material (Phenomenex, Torrance, CA), and the LC separation was conducted with a Dionex Ultimate 3000 UHPLC (Thermo Scientific, San Jose, CA).LC-MS/MS bioinformatics—database searching, configuration, and quantificationMetaproteomics requires a database of potential protein sequences to match observed mass spectra with known peptides. Because we had sample-specific metagenome and metatranscriptome sequencing for each metaproteomic sample, we assessed various database configurations, including those that we predict would be suboptimal, to examine potential options for future metaproteomics researchers. We used five different configurations, described below. In each case, we appended a database of common contaminants (Global Proteome Machine Organization common Repository of Adventitious Proteins). We evaluated the performance of different database configurations based on the number of peptides identified (using a peptide false discovery rate of 1%).In order to make these databases (Table 1), we performed three separate assemblies on (1) the metagenomic reads (from samples GOS-927, GOS-930, GOS-933, and GOS-935), (2) metatranscriptomic reads (from samples GOS-927, GOS-930, GOS-933, and GOS-935), and (3) metatranscriptomic reads from a concurrent metatranscriptomic experiment, started at the location where GOS-933 was taken [28]. Database configurations were created by subsetting from these assemblies. The first configuration was “one-sample database”, constructed to represent the scenario where only one sample was used for metagenomic and metatranscriptomic sequencing (we chose the first sampling week). Specifically, this was done by subsetting and including ORFs from the metagenomic and metatranscriptomic assemblies if reads from this time point were present in that sample (reads mapped as in [28]), and then removing redundant protein sequences (P. Wilmarth, fasta utilities). The second configuration was the “sample-specific database”, where each metaproteomic sample had one corresponding database (prepared from both metagenome and metatranscriptome sequencing completed at the same sampling site), also done by subsetting ORFs from the metagenomic and metatranscriptomic assemblies as described above. The third configuration was pooling databases across size fractions—such that all metagenomic and metatranscriptomic sequences across the same filter sizes (e.g., 3.0 µm) were combined. ORFs were subsetted from the metagenomic and metatranscriptomic assemblies as above. The fourth and fifth configurations are from the concurrent metatranscriptomic experiment [28]. The fourth configuration (“metatranscriptome experiment (T0)”) was the metatranscriptome of the in situ microbial community (i.e., at the beginning of the experiment). This database was created by subsetting from the “metatranscriptome experiment (all)” assembly. Finally, the fifth configuration was the metatranscriptome of all experimental treatments pooled together (two iron levels, three temperatures; “metatranscriptome experiment (all)”). The overlap between databases (potential tryptic peptides) in different samples is presented graphically in Supplementary Figs. S1–S3.Table 1 Characteristics of the five different database configurations we used for metaproteomic database searches.Full size tableAfter matching mass spectra with peptide sequences for each database configuration (MSGF + with OpenMS, with a 1% false discovery rate at the peptide level; [41, 42]), we used MS1 ion intensities to quantify peptides. Specifically, we used the FeatureFinderIdentification approach, which cross-maps identified peptides from one mass spectrometry experiment to unidentified features in another experiment—increasing the number of peptide quantifications [43]. This approach requires a set of experiments to be grouped together (i.e., which samples should use this cross-mapping?). We grouped samples based on their filter sizes (including those samples that are replicate injections). First, mass spectrometry runs within each group were aligned using MapAlignerIdentification [44], and then FeatureFinderIdentification was used for obtaining peptide quantities.After peptides have been identified and quantified, we mapped them to proteins or MCL clusters of proteins, which have corresponding functional annotations (KEGG, KO, KOG, Pfams, TIGRFAM; [28, 32, 35,36,37,38,39]). Functional annotations were used in three separate analyses. (1) Exploring the overall functional changes in microbial community metabolism, we mapped peptides to MCL clusters—groups of proteins with similar sequences. These clusters have consensus annotations based on the annotations of proteins found within the clusters (described in detail in [28]). For this section, we only used peptides that uniquely map to MCL clusters. (2) We restricted the second analysis to two protein groups: ribosomal and photosynthetic proteins. For this analysis, we mapped peptides to one of these protein groups if at least one annotation mapped to the protein group (via string matching with keywords). This approach is “greedy” because does not exclude peptides if they also correspond with other functional groupings, but this is necessary because of the difficulties in comparing various annotation formats. (3) The last analysis for functional annotations was for targeted proteins, and we only mapped functions to peptides where the peptides uniquely identify a specific protein (e.g., plastocyanin).Code for the database setup and configuration, database searching, and peptide quantification is open source (https://github.com/bertrand-lab/ross-sea-meta-omics).LC-MS/MS bioinformatics—normalizationNormalization is an important aspect of metaproteomics: it influences all inferred peptide abundances. Typically, the abundance of a peptide is normalized by the sum of all identified peptide abundances. We use the term normalization factor for the inferred sum of peptide abundances. Note that the apparent abundance of observed peptides is dependent on the database chosen. In theory, if fewer peptides are observed because of a poorly matching database, this will decrease the normalization factor, and those peptides that are observed will appear to increase in abundance. It is not known how much this influences peptide quantification in metaproteomics.For each database configuration, we separately calculated normalization factors. We then correlated the sum of observed peptide abundances with each other. To get a database-independent normalization factor, we used the sum of total ion current (TIC) for each mass spectrometry experiment (using pyopenms; [45]), and also examined the correlation with database-dependent normalization factors. If normalization factors are highly correlated with each other, that would indicate database choice does not impact peptide quantification. Using TIC for normalization may have drawbacks, particularly if there are differences in contamination, or amounts of non-peptide ions across samples.Defining proteomic mass fractionProtein abundance can be calculated in two ways: (1) the number of copies of a protein (independent of a proteins’ mass), or (2) the total mass of the protein copies (the sum of peptides). We refer to the latter as a proteomic mass fraction. For example, to calculate a diatom-specific, ribosomal mass fraction, we sum all peptide abundances that are diatom- and ribosome-specific, and divide by the sum of peptide abundances that are diatom-specific. Note that this is slightly different to other methods, like the normalized spectral abundance factor, which normalizes for total protein mass (via protein length; [46]).Combining estimates across filter sizesOrganisms should separate according to their sizes when using sequential filtration with decreasing filter pore sizes. In practise, however, organisms can break because of pressure during filtration, and protein is typically present for large phytoplankton on the smallest filter size and vice versa. We used a simple method for combining observations across filter sizes, weighted by the number of observations per filter. We begin with the abundance of a given peptide, which was only considered present if it was observed across all injections of the same sample. We calculated the sum of observed peptide intensities (i.e., the normalization factor), and divided all peptide abundances by this normalization factor. Normalized peptide abundances are then averaged across replicate injections. If we are estimating the ribosomal mass fraction of the diatom proteome, we first normalize the diatom-specific peptide intensities as a proportion of diatom biomass (i.e., divide all diatom-specific peptides by the sum of all diatom-specific peptides). We then summed all diatom-normalized peptides intensities that are unique to both diatoms and ribosomal proteins, which would give us the ribosomal proportion of the diatom proteome. Yet, we typically would obtain multiple estimates of, for example, ribosomal mass fraction of diatoms, on different filters. We combined the three values by multiplying each by a coefficient that represents a weight for each observation (specific to a filter size). These coefficients sum to one, and are calculated by summing the total number of peptides observed at a time point for a filter, and dividing by the total number of peptides observed across filters (but within each time point). For example, if we observed 100 peptides that are diatom- and ribosome-specific, and 90 of these peptides were on the 3.0 µm filter and only ten were on the 0.8 μm filter, we would multiply the 3.0 µm filter estimate by 0.9 and the 0.8 µm filter by 0.1. This method uses all available information about proteome composition across different filter sizes (similar to [47]).When we estimate the proteomic mass fraction of a given protein pool, we do not need to adjust for the total protein on each filter. This is because this measurement is independent of total protein. However, for merging estimates of total relative abundance of different organisms across filters, we needed to additionally weight the abundance estimate by the amount of protein on each filter. Therefore, in addition to the weighting scheme described above, we multiplied taxon abundance estimates by the total protein on each filter divided by the total protein across filters on a given day.LC-MS/MS simulationWe used simulations of metaproteomes and LC-MS/MS to (1) quantify biases associated with inferring coarse-grained proteomes from metaproteomes, and (2) to mitigate these biases in our inferences. Specifically, we asked the question: how does sequence diversity impact quantification of coarse-grained proteomes from metaproteomes? Consider a three organism microbial community. If two organisms are extremely similar, there will be very few peptides that can uniquely map to those organisms, resulting in underestimated abundance. The third organism would also be underestimated, but to a lesser degree, unless it had a completely unique set of peptides. A similar outcome is anticipated with differences in sequence diversity across protein groups, such that highly conserved protein groups will be underestimated.Our mass spectrometry simulations offer a unique perspective on this issue: we know the “true” metaproteome, and we can compare this with an “inferred” metaproteome. We simulated variable numbers of taxonomic groups, each with different protein pools of variable sequence diversity. From this simulated metaproteome, we then simulated LC-MS/MS-like sampling of peptides. Complete details of the mass spectrometry simulation are available in [48] and the Supplementary materials. The only difference between this model and that presented in [48] is here we include dynamic exclusion. The ultimate outcomes from these simulations were (1) identifying which circumstances lead to biased inferences about proteomic composition, and (2) determining the underpinnings of these biases.Cofragmentation bias scores for peptidesWe recently developed a computational model (“cobia”) that predicts a peptides’ risk for interference by sample complexity (more specifically, by cofragmentation of multiple peptides; [48]). This study showed that coarse-grained taxonomic and functional groupings are more robust to bias, and that this model can also be used to estimate bias. We ran cobia with the sample-specific databases, which produces a “cofragmentation score”—a measure of risk of being subject to cofragmentation bias. Specifically, the retention time prediction method used was RTPredict [49] with an “OLIGO” kernel for the support vector machine. The parameters for the model were: 0.008333 (maximum injection time); 3 (precursor selection window); 1.44 (ion peak width); and 5 (degree of sparse sampling). Code for running this analysis, as well as the corresponding input parameter file, is found at https://github.com/bertrand-lab/ross-sea-meta-omics.Description of previously published datasets analyzedWe leveraged several previously published datasets to compare our metaproteomic results. Specifically, we used proteomic data of phytoplankton cultures of Phaeocystis antarctica and Thalassiosira pseudonana [27, 50], and of cultures of Escherichia coli under 22 different culture conditions [51]. Coarse-grained proteomic estimates were also compared with previously published targeted metaproteomic data [27]. More

1.Lalli, C. & Parsons, T. R. Biological Oceanography: An Introduction (Elsevier, 1997).
Google Scholar 
2.Mackas, D. L., Denman, K. L. & Abbott, M. R. Plankton patchiness: Biology in the physical vernacular. Bull. Mar. Sci. 37, 652–674 (1985).
Google Scholar 
3.Kjerfve, B. & Magill, K. E. Geographic and hydrodynamic characteristics of shallow coastal lagoons. Mar. Geol. 88, 187–199 (1989).ADS 
Article 

Google Scholar 
4.Kjerfve, B. Chapter 1, Coastal lagoons. In Elsevier Oceanography Series Vol. 60 (ed. Kjerfve, B.) 1–8 (Elsevier, 1994).
Google Scholar 
5.McManus, M. A. & Woodson, C. B. Plankton distribution and ocean dispersal. J. Exp. Biol. 215, 1008–1016 (2012).PubMed 
Article 
PubMed Central 

Google Scholar 
6.Carr, M.-E. Estimation of potential productivity in Eastern Boundary Currents using remote sensing. Deep Sea Res. Part II 49, 59–80 (2001).ADS 
Article 

Google Scholar 
7.Chavez, F. P. & Messié, M. A comparison of Eastern Boundary Upwelling Ecosystems. Prog. Oceanogr. 83, 80–96 (2009).ADS 
Article 

Google Scholar 
8.Schubert, H. & Telesh, I. Estuaries and coastal lagoons. In Biological Oceanography of the Baltic Sea (eds Snoeijs-Leijonmalm, P. et al.) 483–509 (Springer Netherlands, 2017). https://doi.org/10.1007/978-94-007-0668-2_13.Chapter 

Google Scholar 
9.Pecqueur, D. et al. Dynamics of microbial planktonic food web components during a river flash flood in a Mediterranean coastal lagoon. Hydrobiologia 673, 13–27 (2011).CAS 
Article 

Google Scholar 
10.Deininger, A. et al. Simulated terrestrial runoff triggered a phytoplankton succession and changed seston stoichiometry in coastal lagoon mesocosms. Mar. Environ. Res. 119, 40–50 (2016).CAS 
PubMed 
Article 

Google Scholar 
11.Newton, A. & Mudge, S. M. Temperature and salinity regimes in a shallow, mesotidal lagoon, the Ria Formosa, Portugal. Estuar. Coast. Shelf Sci. 57, 73–85 (2003).ADS 
CAS 
Article 

Google Scholar 
12.Huang, J., Gao, J. & Hörmann, G. Hydrodynamic-phytoplankton model for short-term forecasts of phytoplankton in Lake Taihu, China. Limnologica 42, 7–18 (2012).CAS 
Article 

Google Scholar 
13.Pérez-Ruzafa, A. et al. Connectivity between coastal lagoons and sea: Asymmetrical effects on assemblages’ and populations’ structure. Estuar. Coast. Shelf Sci. 216, 171–186 (2019).ADS 
Article 

Google Scholar 
14.Dube, A., Jayaraman, G. & Rani, R. Modelling the effects of variable salinity on the temporal distribution of plankton in shallow coastal lagoons. J. Hydro-environ. Res. 4, 199–209 (2010).Article 

Google Scholar 
15.Pulina, S., Satta, C. T., Padedda, B. M., Sechi, N. & Lugliè, A. Seasonal variations of phytoplankton size structure in relation to environmental variables in three Mediterranean shallow coastal lagoons. Estuar. Coast. Shelf Sci. 212, 95–104 (2018).ADS 
Article 

Google Scholar 
16.Millet, B. & Cecchi, P. Wind-induced hydrodynamic control of the phytoplankton biomass in a lagoon ecosystem. Limnol. Oceanogr. 37, 140–146 (1992).ADS 
Article 

Google Scholar 
17.Souchu, P. et al. Influence of shellfish farming activities on the biogeochemical composition of the water column in Thau lagoon. Mar. Ecol. Prog. Ser. 218, 141–152 (2001).ADS 
CAS 
Article 

Google Scholar 
18.Paphitis, D. & Collins, M. B. Sediment resuspension events within the (microtidal) coastal waters of Thermaikos Gulf, northern Greece. Cont. Shelf Res. 25, 2350–2365 (2005).ADS 
Article 

Google Scholar 
19.Trombetta, T. et al. Water temperature drives phytoplankton blooms in coastal waters. PLoS One 14, e0214933 (2019).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
20.Faust, K. & Raes, J. Microbial interactions: From networks to models. Nat. Rev. Microbiol. 10, 538–550 (2012).CAS 
PubMed 
Article 

Google Scholar 
21.Fuhrman, J. A., Cram, J. A. & Needham, D. M. Marine microbial community dynamics and their ecological interpretation. Nat. Rev. Microbiol. 13, 133–146 (2015).CAS 
PubMed 
Article 

Google Scholar 
22.Trombetta, T., Vidussi, F., Roques, C., Scotti, M. & Mostajir, B. Marine microbial food web networks during phytoplankton bloom and non-bloom periods: Warming favors smaller organism interactions and intensifies trophic cascade. Front. Microbiol. 11, 502336 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
23.Needham, D. M. & Fuhrman, J. A. Pronounced daily succession of phytoplankton, archaea and bacteria following a spring bloom. Nat. Microbiol. 1, 16005 (2016).CAS 
PubMed 
Article 

Google Scholar 
24.Hemraj, D. A., Hossain, A., Ye, Q., Qin, J. G. & Leterme, S. C. Anthropogenic shift of planktonic food web structure in a coastal lagoon by freshwater flow regulation. Sci. Rep. 7, 44441 (2017).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
25.Bec, B., Husseini Ratrema, J., Collos, Y., Souchu, P. & Vaquer, A. Phytoplankton seasonal dynamics in a Mediterranean coastal lagoon: Emphasis on the picoeukaryote community. J. Plankton Res. 27, 881–894 (2005).CAS 
Article 

Google Scholar 
26.Collos, Y. et al. Oligotrophication and emergence of picocyanobacteria and a toxic dinoflagellate in Thau lagoon, southern France. J. Sea Res. 61, 68–75 (2009).ADS 
Article 

Google Scholar 
27.Collos, Y. et al. Pheopigment dynamics, zooplankton grazing rates and the autumnal ammonium peak in a Mediterranean lagoon. Hydrobiologia 550(1), 83–93 (2005).CAS 
Article 

Google Scholar 
28.Gangnery, A. et al. Growth model of the Pacific oyster, Crassostrea gigas, cultured in Thau Lagoon (Méditerranée, France). Aquaculture 215, 267–290 (2003).Article 

Google Scholar 
29.Pernet, F. et al. Marine diatoms sustain growth of bivalves in a Mediterranean lagoon. J. Sea Res. 68, 20–32 (2012).ADS 
CAS 
Article 

Google Scholar 
30.Rose, J. M. & Caron, D. A. Does low temperature constrain the growth rates of heterotrophic protists? Evidence and implications for algal blooms in cold waters. Limnol. Oceanogr. 52, 886–895 (2007).ADS 
Article 

Google Scholar 
31.Kordas, R. L., Harley, C. D. G. & O’Connor, M. I. Community ecology in a warming world: The influence of temperature on interspecific interactions in marine systems. J. Exp. Mar. Biol. Ecol. 400, 218–226 (2011).Article 

Google Scholar 
32.Jones, K. J. & Gowen, R. J. Influence of stratification and irradiance regime on summer phytoplankton composition in coastal and shelf seas of the British Isles. Estuar. Coast. Shelf Sci. 30, 557–567 (1990).ADS 
Article 

Google Scholar 
33.Bosak, S., Godrijan, J. & Šilović, T. Dynamics of the marine planktonic diatom family Chaetocerotaceae in a Mediterranean coastal zone. Estuar. Coast. Shelf Sci. 180, 69–81 (2016).ADS 
Article 

Google Scholar 
34.Wagner, C. & Adrian, R. Consequences of changes in thermal regime for plankton diversity and trait composition in a polymictic lake: A matter of temporal scale. Freshw. Biol. 56, 1949–1961 (2011).Article 

Google Scholar 
35.Rynearson, T. A., Flickinger, S. A. & Fontaine, D. N. Metabarcoding reveals temporal patterns of community composition and realized thermal niches of Thalassiosira spp. (Bacillariophyceae) from the Narragansett Bay long-term plankton time series. Biology 9, 19 (2020).CAS 
PubMed Central 
Article 

Google Scholar 
36.Delaney, M. P. Effects of temperature and turbulence on the predator–prey interactions between a heterotrophic flagellate and a marine bacterium. Microb. Ecol. 45, 218–225 (2003).CAS 
PubMed 
Article 

Google Scholar 
37.Peter, K. H. & Sommer, U. Phytoplankton cell size: Intra- and interspecific effects of warming and grazing. PLoS One 7, e49632 (2012).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
38.van Donk, E. & Kilham, S. S. Temperature effects on silicon- and phosphorus-limited growth and competitive interactions among three diatoms. J. Phycol. 26, 40–50 (1990).Article 

Google Scholar 
39.Stelzer, C.-P. Population growth in planktonic rotifers. Does temperature shift the competitive advantage for different species? In Rotifera VIII: A Comparative Approach (eds Wurdak, E. et al.) 349–353 (Springer Netherlands, 1998).Chapter 

Google Scholar 
40.Arandia-Gorostidi, N., Weber, P. K., Alonso-Sáez, L., Morán, X. A. G. & Mayali, X. Elevated temperature increases carbon and nitrogen fluxes between phytoplankton and heterotrophic bacteria through physical attachment. ISME 11, 641–650 (2017).CAS 
Article 

Google Scholar 
41.Peacock, E. E., Olson, R. J. & Sosik, H. M. Parasitic infection of the diatom Guinardia delicatula, a recurrent and ecologically important phenomenon on the New England Shelf. Mar. Ecol. Prog. Sci. 503, 1–10 (2014).ADS 
Article 

Google Scholar 
42.Käse, L. et al. Host-parasitoid associations in marine planktonic time series: Can metabarcoding help reveal them?. PLoS One 16, e0244817 (2021).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
43.Caron, D. A., Dennett, M. R., Lonsdale, D. J., Moran, D. M. & Shalapyonok, L. Microzooplankton herbivory in the Ross Sea, Antarctica. Deep Sea Res. Part II 47, 3249–3272 (2000).ADS 
Article 

Google Scholar 
44.Vidussi, F. et al. Effects of experimental warming and increased ultraviolet B radiation on the Mediterranean plankton food web. Limnol. Oceanogr. 56, 206–218 (2011).ADS 
CAS 
Article 

Google Scholar 
45.Balas, L. & Özhan, E. Three-dimensional modelling of stratified coastal waters. Estuar. Coast. Shelf Sci. 54, 75–87 (2002).ADS 
Article 

Google Scholar 
46.Dalu, T., Richoux, N. B. & Froneman, P. W. Distribution of benthic diatom communities in a permanently open temperate estuary in relation to physico-chemical variables. S. Afr. J. Bot. 107, 31–38 (2016).Article 

Google Scholar 
47.Sommer, U., Peter, K. H., Genitsaris, S. & Moustaka-Gouni, M. Do marine phytoplankton follow Bergmann’s rule sensu lato?. Biol. Rev. 92, 1011–1026 (2017).PubMed 
Article 
PubMed Central 

Google Scholar 
48.Kirk, J. T. O. Light and Photosynthesis in Aquatic Ecosystems (Cambridge University Press, 1994).Book 

Google Scholar 
49.Litchman, E., de TezanosPinto, P., Klausmeier, C. A., Thomas, M. K. & Yoshiyama, K. Linking traits to species diversity and community structure in phytoplankton. In Fifty years after the “Homage to Santa Rosalia”: Old and new paradigms on biodiversity in aquatic ecosystems (eds Naselli-Flores, L. & Rossetti, G.) 15–28 (Springer Netherlands, 2010).Chapter 

Google Scholar 
50.Unrein, F., Gasol, J. M., Not, F., Forn, I. & Massana, R. Mixotrophic haptophytes are key bacterial grazers in oligotrophic coastal waters. ISME 8, 164–176 (2014).CAS 
Article 

Google Scholar 
51.Polimene, L. et al. Modelling a light-driven phytoplankton succession. J. Plankton Res. 36, 214–229 (2014).CAS 
Article 

Google Scholar 
52.Hatzaki, M. et al. Seasonal aspects of an objective climatology of anticyclones affecting the Mediterranean. J. Clim. 27, 9272–9289 (2014).ADS 
Article 

Google Scholar 
53.Mostajir, B. et al. Experimental test of the effect of ultraviolet-B radiation in a planktonic community. Limnol. Oceanogr. 44, 586–596 (1999).ADS 
Article 

Google Scholar 
54.Lacuna, D. G. & Uye, S.-I. Influence of mid-ultraviolet (UVB) radiation on the physiology of the marine planktonic copepod Acartia omorii and the potential role of photoreactivation. J. Plankton Res. 23, 143–156 (2001).Article 

Google Scholar 
55.Halac, S. et al. An in situ enclosure experiment to test the solar UVB impact on plankton in a high-altitude mountain lake. I. Lack of effect on phytoplankton species composition and growth. J. Plankton Res. 19, 1671–1686 (1997).CAS 
Article 

Google Scholar 
56.Souchu, P. et al. Patterns in nutrient limitation and chlorophyll a along an anthropogenic eutrophication gradient in French Mediterranean coastal lagoons. Can. J. Fish. Aquat. Sci. 67, 743–753 (2010).CAS 
Article 

Google Scholar 
57.Litchman, E. & Klausmeier, C. A. Trait-based community ecology of phytoplankton. Annu. Rev. Ecol. Evol. Syst. 39, 615–639 (2008).Article 

Google Scholar 
58.Reid, P. C., Lancelot, C., Gieskes, W. W. C., Hagmeier, E. & Weichart, G. Phytoplankton of the North Sea and its dynamics: A review. Neth. J. Sea Res. 26, 295–331 (1990).Article 

Google Scholar 
59.Derolez, V. et al. Two decades of oligotrophication: Evidence for a phytoplankton community shift in the coastal lagoon of Thau (Mediterranean Sea, France). Estuar. Coast. Shelf Sci. 241, 106810 (2020).Article 

Google Scholar 
60.Yool, A., Martin, A. P., Fernández, C. & Clark, D. R. The significance of nitrification for oceanic new production. Nature 447, 999–1002 (2007).ADS 
CAS 
PubMed 
Article 

Google Scholar 
61.Constantin, S., Constantinescu, Ș & Doxaran, D. Long-term analysis of turbidity patterns in Danube Delta coastal area based on MODIS satellite data. J. Mar. Syst. 170, 10–21 (2017).Article 

Google Scholar 
62.de Jorge, V. N. & van Beusekom, J. E. E. Wind- and tide-induced resuspension of sediment and microphytobenthos from tidal flats in the Ems estuary. Limnol. Oceanogr. 40, 776–778 (1995).ADS 
Article 

Google Scholar 
63.Ubertini, M. et al. Spatial variability of benthic-pelagic coupling in an estuary ecosystem: Consequences for microphytobenthos resuspension phenomenon. PLoS One 7, e44155 (2012).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
64.Madoni, P. Benthic ciliates in Adriatic Sea lagoons. Eur. J. Protistol. 42, 165–173 (2006).PubMed 
Article 

Google Scholar 
65.Cruz, J. et al. Plankton community and copepod production in a temperate coastal lagoon: What is changing in a short temporal scale?. J. Sea Res. 157, 101858 (2020).Article 

Google Scholar 
66.Audouin, J. Hydrologie de l’étang de Thau. Rev. Trav. Inst. Pêches Marit. 26, 5–104 (1962).
Google Scholar 
67.Byun, D. S., Wang, X. H. & Holloway, P. E. Tidal characteristic adjustment due to dyke and seawall construction in the Mokpo Coastal Zone, Korea. Estuar. Coast. Shelf Sci. 59, 185–196 (2004).ADS 
Article 

Google Scholar 
68.Stefanidou, N., Genitsaris, S., Lopez-Bautista, J., Sommer, U. & Moustaka-Gouni, M. Unicellular eukaryotic community response to temperature and salinity variation in mesocosm experiments. Front. Microbiol. 9, 2444 (2018).PubMed 
PubMed Central 
Article 

Google Scholar 
69.Xu, N. et al. Effects of temperature, salinity and irradiance on the growth of the harmful dinoflagellate Prorocentrum donghaiense Lu. Harmful Algae 9, 13–17 (2010).Article 

Google Scholar 
70.Greenwald, G. M. & Hurlbert, S. H. Microcosm analysis of salinity effects on coastal lagoon plankton assemblages. In Saline Lakes V (ed. Hurlbert, S. H.) 307–335 (Springer Netherlands, 1993).Chapter 

Google Scholar 
71.Fiandrino, A., Giraud, A., Robin, S. & Pinatel, C. Validation d’une méthode d’estimation des volumes d’eau échangés entre la mer et les lagunes et définition d’indicateurs hydrodynamiques associés (2012).72.Mostajir, B., Mas, S., Parin, D. & Vidussi, F. High-Frequency physical, biogeochemical and meteorological data of Coastal Mediterranean Thau Lagoon Observatory. SEANOE (2018).73.Données Publiques de Météo-France—Accueil. https://donneespubliques.meteofrance.fr/.74.Kraberg, A., Baumann, M. & Dürselen, C.-D. Coastal phytoplankton: Photo guide for Northern European seas (Univerza v Ljubljani, 2010).
Google Scholar 
75.Bérard-Therriault, L., Poulin, M. & Bossé, L. Guide d’identification du phytoplancton marin de l’estuaire et du golfe du Saint-Laurent incluant également certains protozoaires Canadian Special Publication of Fisheries and Aquatic Sciences No. 128 (NRC Research Press, 1999).
Google Scholar  More

1.Clarke, A. R., Powell, K. S., Weldon, C. W. & Taylor, P. W. The ecology of Bactrocera tryoni (Diptera: Tephritidae): What do we know to assist pest management?. Ann. Appl. Biol. 158, 26–54. https://doi.org/10.1111/j.1744-7348.2010.00448.x (2011).Article 

Google Scholar 
2.Jessup, A. J. et al. in Area-Wide Control of Insect Pests: From Research to Field Implementation (eds M. J. B. Vreysen, A. S. Robinson, & J. Hendrichs) 685–697 (Springer Netherlands, 2007).3.HIA. Australian Horticulture Statistics Handbook 2019/20. Horticulture Innovation Australia Limited (2020).4.Fanson, B. G., Sundaralingam, S., Jiang, L., Dominiak, B. C. & D’Arcy, G. A review of 16 years of quality control parameters at a mass-rearing facility producing Queensland fruit fly, Bactrocera tryoni. Entomol. Exp. Appl. 151, 152–159. https://doi.org/10.1111/eea.12180 (2014)5.Knipling, E. F. Possibilities of Insect Control or Eradication Through the Use of Sexually Sterile Males. J. Econ. Entomol. 48, 459–462. https://doi.org/10.1093/jee/48.4.459 (1955).Article 

Google Scholar 
6.Shelly, T. & McInnis, D. Sterile Insect Technique and Control of Tephritid Fruit Flies: Do Species With Complex Courtship Require Higher Overflooding Ratios?. Ann. Entomol. Soc. Am. 109, 1–11. https://doi.org/10.1093/aesa/sav101 (2015).CAS 
Article 

Google Scholar 
7.Hendrichs, J. & Robinson, A. in Encyclopedia of Insects (Second Edition) (eds Vincent H. Resh & Ring T. Cardé) 953–957 (Academic Press, 2009).8.Dominiak, B., Clifford, C. S. & Nielsen, S. G. Queensland fruit fly (Bactrocera tryoni Froggatt) attraction to and chemical analysis of male annihilation blocks using three concentratios of cuelure at Dubbo, NSW, Australia. Plant Prot. Q. 24, 157–160 (2009).9.Dominiak, B. C., Ekman, J. & Broughton, S. Mass trapping and other management option for mediterranean fruit fly and Queensland fruit fly in Australia. Gen. Appl. Entomol. 44, 1–8 (2016).
Google Scholar 
10.Khan, M. A. M. et al. Semiochemical mediated enhancement of males to complement sterile insect technique in management of the tephritid pest Bactrocera tryoni (Froggatt). Sci. Rep. 7, 13366. https://doi.org/10.1038/s41598-017-13843-w (2017).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
11.Vargas, R. I., Shelly, T. E., Leblanc, L. & Piñero, J. C. in Vitamins & Hormones Vol. 83 (ed Gerald Litwack) Ch. 23, 575–595 (Academic Press, 2010).12.Benelli, G. et al. Sexual communication and related behaviours in Tephritidae: current knowledge and potential applications for Integrated Pest Management. J. Pest. Sci. 87, 385–405. https://doi.org/10.1007/s10340-014-0577-3 (2014).Article 

Google Scholar 
13.Shelly, T. E. Consumption of methyl eugenol by male Bactrocera dorsalis (Diptera: Tephritidae): low incidence of repeat feeding. Florida Entomologist 77, 201–208 (1994).CAS 
Article 

Google Scholar 
14.Shelly, T. E. Trapping male oriental fruit flies (Diptera: Tephritidae): does feeding on a natural source of methyl eugenol reduce capture probability?. Florida Entomologist 83, 109–111 (2000).Article 

Google Scholar 
15.Tan, K. H., & Toong, Y. C. Floral synomone of a wild orchid, Bulbophyllum cheiri, lures Bactrocera fruit flies for pollination. J. Chem. Ecol. 28, 1161–1172 (2002).16.Shelly, T. E. Evaluation of a Genetic Sexing Strain of the Oriental Fruit Fly as a Candidate for Simultaneous Application of Male Annihilation and Sterile Insect Techniques (Diptera: Tephritidae). J. Econ. Entomol. 113, 1913–1921. https://doi.org/10.1093/jee/toaa099 (2020).CAS 
Article 
PubMed 

Google Scholar 
17.Joseph, R. M. & Carlson, J. R. Drosophila Chemoreceptors: A Molecular Interface Between the Chemical World and the Brain. Trends in genetics : TIG 31, 683–695. https://doi.org/10.1016/j.tig.2015.09.005 (2015).CAS 
Article 
PubMed 

Google Scholar 
18.Venthur, H., & Zhou, J.-J. Odorant receptors and odorant-binding proteins as insect pest control targets: A comparative analysis. Front. Physiol. 9, 1. https://doi.org/10.3389/fphys.2018.01163 (2018).19.Leal, W. S. Odorant reception in insects: roles of receptors, binding proteins, and degrading enzymes. Annu Rev. Entomol. 58, 373–391. https://doi.org/10.1146/annurev-ento-120811-153635 (2013).CAS 
Article 
PubMed 

Google Scholar 
20.Zheng, W. et al. Identification and Expression Profile Analysis of Odorant Binding Proteins in the Oriental Fruit Fly Bactrocera dorsalis. Int. J. Mol. Sci. 14, 14936 (2013).Article 

Google Scholar 
21.Siciliano, P. et al. Identification of pheromone components and their binding affinity to the odorant binding protein CcapOBP83a-2 of the Mediterranean fruit fly, Ceratitis capitata. Insect Biochem. Mol. Biol. 48, 51–62. https://doi.org/10.1016/j.ibmb.2014.02.005 (2014).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
22.Siciliano, P. et al. Sniffing Out Chemosensory Genes from the Mediterranean Fruit Fly, Ceratitis capitata. Plos One 9, e85523https://doi.org/10.1371/journal.pone.0085523 (2014)23.Campanini, E. B. & de Brito, R. A. Molecular evolution of Odorant-binding proteins gene family in two closely related Anastrepha fruit flies. BMC Evol. Biol. 16, 198–198. https://doi.org/10.1186/s12862-016-0775-0 (2016).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
24.Park, K., et al. Expression patterns of two putative odorant-binding proteins in the olfactory organs of Drosophila have different implications for their functions. Vol. 300 (2000).25.Shanbhag, S. R. et al. Expression mosaic of odorant-binding proteins in Drosophila olfactory organs. Microsc. Res. Tech. 55, 297–306. https://doi.org/10.1002/jemt.1179 (2001).CAS 
Article 
PubMed 

Google Scholar 
26.Wu, Z. et al. Discovery of chemosensory genes in the oriental fruit fly, Bactrocera dorsalis. PloS One 10, e0129794-e0129794.https://doi.org/10.1371/journal.pone.0129794 (2015)27.Liu, Z., Smagghe, G., Lei, Z. & Wang, J.-J. Identification of male- and female-specific olfaction genes in antennae of the Oriental Fruit Fly (Bactrocera dorsalis). PLoS ONE 11, e0147783–e0147783. https://doi.org/10.1371/journal.pone.0147783 (2016).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
28.Zhang, J. et al. Identification and expression profiles of novel odorant binding proteins and functional analysis of OBP99a in Bactrocera dorsalis. Arch. Insect Biochem. Physiol. 98, e21452. https://doi.org/10.1002/arch.21452 (2018).ADS 
CAS 
Article 
PubMed 

Google Scholar 
29.Cheng, J. et al. Identification and expression analysis of chemosensory genes in the citrus fruit fly Bactrocera (Tetradacus) minax. PeerJ Preprints 6, e27297v1. https://doi.org/10.7287/peerj.preprints.27297v1 (2018).30.Kumaran, N., Prentis, P. J., Mangalam, K. P., Schutze, M. K. & Clarke, A. R. Sexual selection in true fruit flies (Diptera: Tephritidae): transcriptome and experimental evidences for phytochemicals increasing male competitive ability. Mol. Ecol. 23, 4645–4657. https://doi.org/10.1111/mec.12880 (2014).CAS 
Article 
PubMed 

Google Scholar 
31.Liu, H. et al. BdorOBP2 plays an indispensable role in the perception of methyl eugenol by mature males of Bactrocera dorsalis (Hendel). Sci. Rep. 7, 15894. https://doi.org/10.1038/s41598-017-15893-6 (2017).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
32.Kumaran, N. et al. Plant-Mediated Female Transcriptomic Changes Post-Mating in a Tephritid Fruit Fly, Bactrocera tryoni. Genome biology and evolution 10, 94–107.https://doi.org/10.1093/gbe/evx257 (2017)33.Idrees, A. et al. Protein baits, volatile compounds and irradiation influence the expression profiles of odorant-binding protein genes in Bactrocera dorsalis (Diptera: Tephritidae). Appl. Ecol. Environ. Res. 15, 1883–1899 (2017).Article 

Google Scholar 
34.Pavlidi, N. et al. Transcriptomic responses of the olive fruit fly Bactrocera oleae and its symbiont Candidatus Erwinia dacicola to olive feeding. Sci. Rep. 7, 42633. https://doi.org/10.1038/srep42633 (2017).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
35.Arya, G. H. et al. The genetic basis for variation in olfactory behavior in Drosophila melanogaster. Chem Senses 40, 233–243. https://doi.org/10.1093/chemse/bjv001 (2015).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
36.Akter, H., Adnan, S., Morelli, R., Rempoulakis, P. & Taylor, P. W. Suppression of cuelure attraction in male Queensland fruit flies provided raspberry ketone supplements as immature adults. PLoS ONE 12, e0184086. https://doi.org/10.1371/journal.pone.0184086 (2017).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
37.Gilchrist, A. S. et al. The draft genome of the pest tephritid fruit fly Bactrocera tryoni: resources for the genomic analysis of hybridising species. BMC Genomics 15, 1153. https://doi.org/10.1186/1471-2164-15-1153 (2014).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
38.Fleischer, J. & Krieger, J. Insect pheromone receptors—Key elements in sensing intraspecific chemical signals. Front. Cell. Neurosci. 12. https://doi.org/10.3389/fncel.2018.00425 (2018).39.Liu, Z. et al. An antennae-specific odorant-binding protein is involved in Bactrocera dorsalis Olfaction. Front. Ecol. Evol. 8.https://doi.org/10.3389/fevo.2020.00063 (2020).40.Tan, K. H. Recaptures of feral Bactrocera dorsalis and B. umbrosa (Diptera: Tephritidae) males after feeding on methyl eugenol. Bull. Entomol. Res. 110, 15–21, https://doi.org/10.1017/S0007485319000208 (2019).41.Liu, L. et al. Contribution of Drosophila DEG/ENaC genes to salt taste. Neuron 39, 133–146. https://doi.org/10.1016/s0896-6273(03)00394-5 (2003).CAS 
Article 
PubMed 

Google Scholar 
42.Swarup, S., Huang, W., Mackay, T. F. C. & Anholt, R. R. H. Analysis of natural variation reveals neurogenetic networks for Drosophila olfactory behavior. Proc Natl Acad Sci USA 110, 1017–1022. https://doi.org/10.1073/pnas.1220168110 (2013).ADS 
Article 
PubMed 

Google Scholar 
43.Vijayan, V., Thistle, R., Liu, T., Starostina, E. & Pikielny, C. W. Drosophila pheromone-sensing neurons expressing the ppk25 Ion channel subunit stimulate male courtship and female receptivity. PLoS Genet. 10, e1004238. https://doi.org/10.1371/journal.pgen.1004238 (2014).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
44.Thistle, R., Cameron, P., Ghorayshi, A., Dennison, L. & Scott, K. Contact chemoreceptors mediate male-male repulsion and male-female attraction during Drosophila courtship. Cell 149, 1140–1151. https://doi.org/10.1016/j.cell.2012.03.045 (2012).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
45.Khan, M. A. M. et al. Raspberry ketone accelerates sexual maturation and improves mating performance of sterile male Queensland fruit fly, Bactrocera tryoni (Froggatt). Pest Manag. Sci. 75, 1942–1950. https://doi.org/10.1002/ps.5307 (2019).CAS 
Article 
PubMed 

Google Scholar 
46.Weldon, C. W., Perez-Staples, D. & Taylor, P. W. Feeding on yeast hydrolysate enhances attraction to cue-lure in Queensland fruit flies, Bactrocera tryoni. Entomol. Experim. et Applicata 129, 200–209. https://doi.org/10.1111/j.1570-7458.2008.00768.x (2008)47.Najar-Rodriguez, A. J., Galizia, C. G., Stierle, J. & Dorn, S. Behavioral and neurophysiological responses of an insect to changing ratios of constituents in host plant-derived volatile mixtures. J. Exp. Biol. 213, 3388 (2010).CAS 
Article 

Google Scholar 
48.Bertschy, C., Turlings, T. C., Bellotti, A. C. & Dorn, S. Chemically-mediated attraction of three parasitoid species to mealybug-infested cassava leaves. Florida Entomologist 80, 383–395 (1997).Article 

Google Scholar 
49.Butler, D. G., Cullis, B. R., Gilmour, A. R. & Gogel, B. J. (Queensland Department of Primary Industries and Fisheries, Australia, 2009).50.Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120. https://doi.org/10.1093/bioinformatics/btu170 (2014).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
51.Zhao, Q.-Y. et al. Optimizing de novo transcriptome assembly from short-read RNA-Seq data: a comparative study. BMC Bioinf. 12, S2. https://doi.org/10.1186/1471-2105-12-S14-S2 (2011).CAS 
Article 

Google Scholar 
52.Fu, L., Niu, B., Zhu, Z., Wu, S. & Li, W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics 28, 3150–3152. https://doi.org/10.1093/bioinformatics/bts565 (2012).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
53.Simão, F. A., Waterhouse, R. M., Ioannidis, P., Kriventseva, E. V. & Zdobnov, E. M. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics 31, 3210–3212. https://doi.org/10.1093/bioinformatics/btv351 (2015).CAS 
Article 
PubMed 

Google Scholar 
54.Conesa, A. et al. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21, 3674–3676. https://doi.org/10.1093/bioinformatics/bti610 (2005).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
55.Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760. https://doi.org/10.1093/bioinformatics/btp324 (2009).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
56.Patro, R., Duggal, G., Love, M. I., Irizarry, R. A. & Kingsford, C. Salmon provides fast and bias-aware quantification of transcript expression. Nat. Methods 14, 417–419. https://doi.org/10.1038/nmeth.4197 (2017).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
57.Risso, D., Ngai, J., Speed, T. P. & Dudoit, S. Normalization of RNA-seq data using factor analysis of control genes or samples. Nat. Biotechnol. 32, 896–902. https://doi.org/10.1038/nbt.2931 (2014).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
58.Al-Shahrour, F., Díaz-Uriarte, R. & Dopazo, J. FatiGO: a web tool for finding significant associations of Gene Ontology terms with groups of genes. Bioinformatics 20, 578–580. https://doi.org/10.1093/bioinformatics/btg455 (2004).CAS 
Article 
PubMed 

Google Scholar 
59.Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucl. Acids Res. 43, e47–e47. https://doi.org/10.1093/nar/gkv007 (2015).CAS 
Article 
PubMed 
PubMed Central 

Google Scholar  More

Data collectionData on the birds’ reproductive success and the number of humans present at nest sites were collected as part of a long-term, ongoing monitoring project in Hungary, in which we investigate the impacts of urbanization on populations of great tits. The great tit is an insectivorous passerine bird that is widespread across the Western Palearctic, occupies both urban and forest habitats, readily accepts nestboxes, and shares many important ecological traits with other tit or chickadee species also occurring in urban habitats27. These traits make this species an ideal model organism for studying the effects of the anthropause on wildlife in different environments.Study sitesWe monitored breeding great tit populations and also collected human presence data in two urban areas and at one forest study site. In one of the urban sites, Veszprém (47°05′17.29″N, 17°54′29.66″E; human population: c. 56,000; the monitoring scheme started in 2013), the nestboxes were placed in public green spaces (public parks, university campuses, a bus station, and a cemetery) that are surrounded by built-up areas and roads, and experience frequent anthropogenic disturbance. At the other urban size, Budapest (47°30′27.4″N, 19°01′03.4″E; the capital city of Hungary, human population: c. 1.75 million; the monitoring scheme started in 2019), the nestboxes were placed in two public urban parks, located c. 400 m from each other in the city core area and separated by high-traffic roads. The parks are freely accessible to residents and are heavily embedded within the urban matrix. At both urban sites, most of the nestboxes are distributed along paths or walking trails. Even though the two cities greatly differ in their size and human population, our urban study plots in both cities have similar general characteristics: these are surrounded by built-up areas, are at a similar distance (c. 3–4 km) from the nearest forested areas (for Veszprém, this is the forest at Vilma-puszta: 47°05′06.7″N, 17°51′51.4″E; for Budapest, this is the forest at Normafa: 47°30′27.7″N 18°57′51.1″E), and nests also experienced a similar level of human disturbance in the pre-COVID reference period (Fig. 1b). The forest site, Szentgál (47°06′39.75″N, 17°41′17.94″E; the monitoring scheme started in 2013), is a mature woodland, dominated by beech (Fagus sylvatica) and hornbeam (Carpinus betulus), located 3 km away from the nearest human habitation (Szentgál, human population: c. 2.800), c. 20 km and 110 km away from Veszprém and Budapest, respectively. There are no paved roads in the forest, and the area is relatively free from human disturbance although it experiences occasional hunting and logging activity.Human presence around nestsTo quantify human presence at our study sites for 2020 and the reference years we counted the number of humans (motorized vehicles excluded) during each nest check, for 30 s, in the proximity of the nestboxes (for similar approach see Corsini et al. 2019). The number of humans was recorded within a 50-m radius of the nestboxes between 2013 and 2018 (Veszprém, Szentgál), and within a 15-m radius distance in 2019–2020 (all sites). We changed the counting distance in 2019 due to methodological reasons following28. However, to be able to compare the human presence data of 2020 in Veszprém and Szentgál to that recorded in earlier years, in 2020 we performed the counts with both the 15-m and the 50-m radius distances at these two sites. Thus, for 2020 in Veszprém, we have human presence data both for the 50-m and the 15-m radius areas that were used in the forest-city and the between-cities comparisons, respectively (see below). For each year and study site, we used human presence data only from seasonally first broods (defined below), and only from nests where there were either already eggs or nestlings in the nest, resulting in 9.4 ± 3.6 (mean ± SD) observations per brood which is a reliable indicator of human presence28.Birds’ reproductive successWe monitored nestboxes each year at least twice a week from mid-March to early June to record laying and hatching dates, clutch size, hatching success, and the number of nestlings in active great tit nests. We ringed nestlings at day 14–16 post-hatch (i.e. a few days before fledging; hatching day of the first chick = day 1) with a numbered metal ring and also recorded their body mass (to the nearest 0.1 g), tarsus length (to the nearest 0.1 mm and following Svensson’s ‘alternative’ method29) and wing length (from the bend of the wing to the longest primary; to nearest 1 mm). Shortly after the expected date of fledging we carefully examined the nest material to identify and count the number of chicks that died after ringing (due to e.g. starvation, predation) that we included in the calculation of nestling survival (detailed below). The aim of this is to get a more accurate estimate for the number of offspring that could indeed fledge from the nest. The number of broods (nestlings) that suffered partial or complete mortality between ringing and fledging were: n = 6 (13) in Budapest (2019–2020), n = 70 (152) in Veszprém (2013–2020), and n = 25 (83) in the Szentgál forest.From these data we determined clutch size (the maximum number of eggs observed in a brood), hatching success (the proportion of chicks hatched / eggs laid), the number of fledglings, and nestling survival (the proportion of fledged young / hatched chicks). The number of fledglings (i.e. the number of young fledged successfully) was calculated as the number of chicks ringed minus the number of chicks found dead in the nest after the ringing. We involved only seasonally first breeding attempts (as this period overlapped with the lockdown period; detailed at the Statistical analyses), and defined first broods as follows. In our study system breeding great tits are captured on their nests and receive a unique combination of colour rings. Active nests are also routinely equipped with a small, concealed video camera enabling us to reliably identify over 80% of breeding individuals each year30. Thus, relying on this setup, we considered a clutch as a first breeding attempt of a pair if it was initiated before the date of the first egg laid in the earliest second clutch at that site by an individually identifiable (i.e. colour-ringed) female that successfully raised her first clutch (i.e. fledged at least one young) in that year.Air pollution and meteorological conditionsTo describe the levels of traffic-related air pollution (nitrogen dioxide [NO2], nitrogen oxides [NOX] and ozone [O3]) and the meteorological conditions (temperature and precipitation) at the two urban study sites (Veszprém and Budapest), we used data provided by the Hungarian Air Quality Monitoring Network and the Hungarian Meteorological Service, respectively. To better understand which aspect of the anthropause might have affected great tits’ breeding success we thus assessed if the lockdown affected air pollution levels differently at the two urban study sites (compared to 2019), or if weather conditions showed different fluctuations between 2019 and 2020 at the two cities. For more details on the statistical analyses and results, see ESM: Sect. 1.Statistical analysesThe duration of the official restrictions on human mobility (lockdown) spanned between 28 March–4 May in Veszprém (calendar date: 88–125; 01 January = 1) and 28 March–18 May (88–139) in Budapest. During this period people were allowed to leave their homes e.g. to run essential errands including individual sport and recreational activities in public green spaces, although with keeping at least 1.5 m from each other (social distancing). Very importantly, from the point of view of our study, the period of movement restrictions had almost completely overlapped with the seasonally first breeding attempts (from egg-laying to fledging) of great tits at both urban sites. The date of laying the 1st egg (calendar date, mean ± SD) in Veszprém was 94.2 ± 6.4, while in Budapest 97 ± 7.8; the date of chick ringing and measuring in Veszprém was 128 ± 5.3, while in Budapest: 133 ± 9.1. Thus, we decided not to exclude any first broods based on the date in order to maximize our sample size. Similarly, the period from which we involved human presence data was also strongly overlapped with the duration of the movement restrictions in both cities. Therefore, in Veszprém, the calendar dates of the first and the last human count at each nest were 87–108 (median: 100) and 121–142 (median: 132), respectively, while in Budapest 87–128 (median: 98) and 118–155 (median: 128).Human presence around nestsIn accordance with our first objective (forest-city comparisons), we explored if the lockdown in 2020 caused any changes in human disturbance around the great tit nests. To do so, we compared the number of humans (50-m radius of the nests) between 2020 and the 2013–2018 reference period, separately for the forest (Szentgál) and urban (Veszprém) study sites. Note that in 2019, we did not collect data on human presence within a 50-m radius at Veszprém and Szentgál (see above: Data collection), therefore 2019 was not included in the reference period of this analysis. We, however, also compared human presence in Veszprém between 2019 and 2020 using the 15-m radius data which indicates a change that is consistent with the differences found using the 50-m radius data (detailed below).First, we built generalized linear mixed-effects (GLM, lme4 R package) models with Poisson error distribution with the number of humans as the response variable, including year as a fixed factor and nestbox ID as random factor to control for non-independence of the data. Next, we extracted the mean values (least-squares means; package emmeans31) and associated standard errors for each year as estimated by the model. We computed the mean of these yearly mean estimates for the 2013–2018 reference period (i.e. calculated a single overall mean describing the whole reference period) and compared this long-term mean to the mean estimate of 2020 by calculating the linear contrast between them (with the ‘contrast’ function of the emmeans package), and expressed linear contrasts as 2020 minus the reference period.For our second objective (between-cities comparisons), we compared the changes in human disturbance around the nestboxes at the two urban study sites, Veszprém and Budapest, using the number of humans recorded within the 15-m radius of the active nests in 2019 and 2020. We analysed the data from Budapest and Veszprém separately and built generalized linear mixed-effects models with Poisson error distribution with the number of humans (15-m radius of the nests) as the response variable, including year as a fixed factor and nestbox ID as random factor to control for non-independence.Birds’ reproductive successWe used data from 2019 (reference; for justification see below in this section) and 2020 (lockdown). First, we constructed separate linear models to analyse each component of reproductive success (response variables), and for the forest-city and the between-cities comparisons. We used linear models (LM) for clutch size and the number of fledglings, generalized linear models (GLM, with quasi-binomial error distribution) for hatching success and nestling survival, and linear-mixed effects models (LME) for nestling body size traits (body mass, tarsus length, and wing length). Models on nestling body size traits contained nestlings’ age at ringing as a confounding variable (three-level factor: 14, 15, or 16 d of age) and brood ID as a random factor to control for the non-independence of chicks raised in the same brood. Finally, these models always contained a habitat (Veszprém or Szentgál) × year (2019 or 2020) interaction term for forest-city comparisons and a city (Budapest or Veszprém) × year (2019 or 2020) interaction term for between-cities comparisons. We checked assumptions of residuals’ normality and homogeneity of variance by inspecting the residuals plots which were respected for all models.Next, to test the prediction for our first objective (forest-city comparisons), we extracted the mean values (least-squares means) and associated standard errors of each response variable for each habitat × year combination as estimated by the linear model’s interaction. Then, from these estimates, we calculated habitat contrasts, i.e. the mean forest-city difference (forest minus urban) for each year (i.e. for 2019 and 2020), and compared the mean habitat contrast for the 2019 reference year to the mean habitat contrast of 2020; for similar approach see14,32,33.For our second objective (between-cities comparisons), we followed the same procedure as for the forest-city comparisons (detailed above) except that here we compared the differences between cities (Budapest minus Veszprém) in 2020 and 2019. These full models (i.e. for the forest-city and between-cities comparisons) are presented in Table S1–S2 (ESM: Sect. 2).In our study, we chose 2019 as a reference year for multiple reasons. First, because this was temporally the closest year without a lockdown. Second, because for Budapest we have monitoring data only from 2019 to 2020, using 2019 and 2020 in all analyses makes the results more comparable. Finally, although we have monitoring data from a total of eight years (2013–2020) for Szentgál (forest site) and Veszprém (urban site), for the forest-city comparisons we did not include years before 2019 in the reference because we noticed a negative trend in birds’ reproductive success throughout the study years (Fig. S3). This trend was especially apparent in the forest population, and may have reduced the forest-city difference by the end of the study period. Indeed, 2019 and 2020 were amongst the poorest years and resulted in a very similar reproductive success between both years within both habitats (Fig. S3). Because such temporal trend may have confounded the comparisons of 2020 with earlier years, to take account for its effect, and to further justify our approach of using 2019 as the reference year, we conducted additional analyses on the birds’ reproductive success by comparing both 2019 and 2020 (separately) to the 2013–2018 long-term reference period. We predict that if 2019 and 2020 are similarly affected by the decreasing trend in reproductive success than then the differences between the long-term reference period and 2019 and 2020, respectively, should be similar. For the details of these long-term forest-city comparisons see ESM: Sect. 3 and Table S3).Finally, we did not conduct the forest-city comparisons (first objective) between the forest site (Szentgál) and the other urban site (Budapest) for two reasons. First, because unlike to the Szentgál vs. Veszprém setup, we did not have an appropriate forest (control) location which is close to Budapest. Second, because conducting comparisons between the long-term data and 2019 and 2020, respectively (see: ESM Sect. 3) was not possible for Budapest because we do not have similar long-term data for the latter site.Clutches that failed before reaching the incubation stage (due to predation or desertion; i.e. final clutch size was uncertain), suffered complete mortality due to weather (e.g. nestbox fall from the tree due to strong wind), and cases when complete or partial clutch or brood loss may have occurred due to the monitoring process (e.g. when a nestbox was dropped or when complete brood failure occurred soon after capturing a parent on the nest) were excluded from all analyses. In the analyses investigating the number of fledglings, fledging success, and nestling body size traits we involved nests only where at least one nestling hatched, and excluded broods that were involved in a food-supplementation experiment (as treatment group) during the nestling rearing period in 201714. We used the R 4.0.5 software environment for statistical analysis and creating figures34.Ethical statementAll procedures were in accordance with Hungarian laws, and adhered to the ASAB/ABS guidelines for the use of animals in behavioural research and teaching. Permit to the use of animals in this study was provided by the National Scientific Ethical Committee on Animal Experimentation (permit number: PE-06/KTF/997–8/2018, FPH061/1329–5/2018, PE-06/KTF/06,543–7/2020 and FPH061/3036–4/2020). Permits to study protected species and access to protected areas were provided by the Middle Transdanubian Inspectorate for Environmental Protection, Natural Protection and Water Management (permit numbers: 31559/2011, 24,861/2014 and VE-09Z/03,454–8/2018, for working in Veszprém and Szentgál) and the Environment Protection and Nature Conservation Department of the Pest County Bureau of the Hungarian Government and the Mayor’s Office of Budapest (permit numbers: PE-06/KTF/997–8/2018, FPH061/1329–5/2018, PE-06/KTF/06,543–7/2020 and FPH061/3036–4/2020, for working in Budapest). More