Philippa Kaur

Wahl, M., Hay, M. E. & Enderlein, P. Effects of epibiosis on consumer–prey interactions. Hydrobiologia 355, 49–59 (1997).Article 

Google Scholar 
Fernandez-Leborans, G., Zitzler, K. & Gabilondo, R. Protozoan ciliate epibionts on the freshwater shrimp Caridina (Crustacea, Decapoda, Atyidae) from the Malili lake system on Sulawesi (Indonesia). J. Nat. Hist. 40, 1983–2000 (2006).Article 

Google Scholar 
Puckett, G. L. & Carman, K. R. Ciliate epibiont effects on feeding, energy reserves, and sensitivity to hydrocarbon contaminants in an estuarine harpactacoid copepod. Estuaries 25, 372–381 (2002).Article 

Google Scholar 
Fernandez-Leborans, G. Epibiosis in Crustacea: an overview. Crustaceana 83, 549–640 (2010).Article 

Google Scholar 
Regali-Seleghim, M. H. & Godinho, M. J. Peritrich epibiont protozoans in the zooplankton of a subtropical shallow aquatic ecosystem (Monjolinho Reservoir, São Carlos, Brazil). J Plankton Res. 26, 501–508 (2004).Article 

Google Scholar 
Bickel, S. L., Tang, K. W. & Grossart, H. P. Ciliate epibionts associated with crustacean zooplankton in German lakes: distribution, motility, and bacterivory. Front. Microbiol. 3, 1–11 (2012).Article 

Google Scholar 
Souissi, A., Souissi, S. & Hwang, J. S. The effect of epibiont ciliates on the behavior and mating success of the copepod Eurytemora affinis. J. Exp. Mar. Biol. Ecol. 445, 38–43. https://doi.org/10.1016/j.jembe.2013.04.002 (2013).Article 

Google Scholar 
Willey, R. L., Cantrell, P. A. & Threlkeld, S. T. Epibiotic euglenoid flagellates increase the susceptibility of some zooplankton to fish predation. Limnol. Oceanogr. 35, 952–959 (1990).Article 
ADS 

Google Scholar 
Ólafsdóttir, S. H. & Svavarsson, J. Ciliate (Protozoa) epibionts of deep-water asellote isopods (Crustacea): pattern and diversity. J. Crust. Biol. 22, 607–618 (2002).Article 

Google Scholar 
Kumari, S., Kumar, R., Sarkar, U. K. & Das, B. S. Record of epibiont ciliates (Ciliophora: Peritrichia) living on freshwater invertebrates in a floodplain wetland. J. Inland Fish. Soc. India. 53, 210–214. https://doi.org/10.47780/jifsi.52.3.2021 (2021).Article 

Google Scholar 
Utz, L. R. P. & Coats, D. W. Spatial and temporal patterns in the occurrence of peritrich ciliates as epibionts on calanoid copepods in the Chesapeake Bay, USA. J. Eukaryot. Microbiol. 52, 236–244 (2005).Article 

Google Scholar 
Utz, L. R. P. & Coats, D. W. Telotroch formation, survival and attachment in the epibiotic peritrich Zoothamnium intermedium (Ciliophora, Oligohymenophorea). Invert. Biol. 127, 237–248 (2008).Article 

Google Scholar 
Ohtsuka, S., et al. Symbiosis of planktonic copepods and mysids with epibionts and parasites in the Northpacific: diversity and interactions. In New Frontiers in Crustacean Biology, 1–14, Brill (2011).Sługocki, Ł et al. Passenger for millenniums: association between stenothermic microcrustacean and suctorian epibiont – the case of Eurytemora lacustris and Tokophyra sp. Sci. Rep. 10, 1–10. https://doi.org/10.1038/s41598-020-66730-2 (2020).Article 

Google Scholar 
Fernandez-Leborans, G. A review of the species of protozoan epibionts on crustaceans. III. Chonotrich ciliates. Crustaceana 74, 581–607. https://doi.org/10.1163/156854001300228852 (2001).Article 

Google Scholar 
Fernandez-Leborans, G. & Tato- Porto, M. L. A review of the species of the protozoan epibionts on crustaceans. I. Peritrich ciliates. Crustaceana 73, 643–683. https://doi.org/10.1163/156854000504705 (2000).Article 

Google Scholar 
Utz, L. R. P. & Coats, D. W. The role of motion in the formation of free living stages and attachment of the peritrich epibiont Zoothamnium intermedium (Ciliophora, Peritrichia). Biosciências 13, 69–74 (2005).
Google Scholar 
Pan, Y. et al. Effects of epibiotic diatoms on the productivity of the Calanoid Copepod Acartia tonsa (Dana) in intensive aquaculture systems. Front. Mar. Sci. 8, 2296–7745. https://doi.org/10.3389/fmars.2021.728779 (2021).Article 

Google Scholar 
Bickel, S. L., Tang, K. W. & Grossart, H. P. Ciliate epibionts associated with crustacean zooplankton in German lakes: distribution, motility, and bacterivory. Front. Microbiol. 3, 243. https://doi.org/10.3389/fmicb.2012.00243 (2012).Article 

Google Scholar 
De Domitrovic, Y. Z. et al. Epibiont algae on planktic micro-crustaceans from a subtropical shallow lake (Argentina). Algol. Stud. 127, 29–38 (2008).Article 

Google Scholar 
Ohman, M. D. Behavioral responses of zooplankton to predation. Bull. Mar. Sci. 43(3), 530–550 (1988).
Google Scholar 
Acevedo-Trejos, E., Marañón, E. & Merico, A. Phytoplankton size diversity and ecosystem function relationships across oceanic regions. Proc. Roy. Soc. B Biol. Sci. 285, 2180621. https://doi.org/10.1098/rspb.2018.0621 (2018).Article 

Google Scholar 
Francesco, P. et al. Interacting temperature, nutrients and Zooplankton Grazing Control Phytoplankton size-abundance relationships in eight Swiss Lakes. Front. Microbiol. 10, 1664–2302. https://doi.org/10.3389/fmicb.2019.03155 (2020).Article 

Google Scholar 
Carman, K. & Dobbs, F. C. Epibiotic microorganisms on copepods and other marine crustaceans. Microsci. Res. Tech. 37, 116–135 (1997).Article 

Google Scholar 
Cabral, A. F. et al. Spatial and temporal occurrence of Rhabdostyla cf. chronomi Kahl, 1933 (Ciliophora, Peritrichia) as an epibiont on chironomid larvae in a lotic system in the neotropics. Hydrobiologia 644, 351–359 (2010).Article 

Google Scholar 
Burris, Z. & Dam, H. G. Deleterious effects of the ciliate epibiont Zoothamnium sp. On fitness of the copepod Acartia tonsa. J. Plankton Res. 36, 788–799. https://doi.org/10.1093/plankt/fbt137 (2014).Article 

Google Scholar 
Yin, Y. et al. Hidden defensive morphology in rotifers: benefits, costs, and fitness consequences. Sci. Rep. 7, 4488. https://doi.org/10.1038/s41598-017-04809-z (2017).Article 
ADS 

Google Scholar 
Gilbert, J. J. & Shröder, T. The ciliate epibiont Epistylis Pygmaeum: selection for zooplankton hosts, reproduction and effect on two rotifers. Freshw. Biol. 48, 878–893 (2003).Article 

Google Scholar 
Gilbert, J. J. Morphological and behavioural responses of a rotifer to the predator Asplanchna. J. Plankton Res. 36, 1576–1584. https://doi.org/10.1093/plankt/fbu075 (2014).Article 

Google Scholar 
Fernandez-Leborans, G. Epibiosis in crustacea: an overview. Crustaceana 83(5), 549–640. https://doi.org/10.1163/001121610X491059 (2010).Article 

Google Scholar 
Iyer, N. & Rao, T. R. Epizoic mode of life in Brachionus rubens Ehrenberg as a deterrent against predation by Asplanchna intermedia Hudson. Hydrobiologia 313, 377–380 (1995).Article 

Google Scholar 
Boyan, B. D., Lotz, E. M. & Schwartz, Z. Roughness and hydrophilicity as osteogenic biomimetic surface properties. Tissue Eng. 23, 1479–1489. https://doi.org/10.1089/ten.TEA.2017.0048 (2017).Article 

Google Scholar 
Ubuo, E. E. et al. the direct cause of amplified wettability: roughness or surface chemistry?. J. Compos. Sci. 5, 213. https://doi.org/10.3390/jcs5080213 (2021).Article 

Google Scholar 
Gilbert, J. J. Attachment behavior in the rotifer Brachionus rubens: induction by Asplanchna and effect on sexual reproduction. Hydrobiologia 844, 9–20. https://doi.org/10.1007/s10750-018-3805-7 (2019).Article 

Google Scholar 
Kumar, R. Effect of Mesocyclops thermocyclopoides (Copepoda, Cyclopoida) predation on population dynamics of different prey: a laboratory study. J. Freshwater Ecol. 18, 383–393. https://doi.org/10.1080/02705060.2003.966397 (2003).Article 

Google Scholar 
Bulut, H. & Saler, S. Presence of an epibiont Epistylis sp. (Protozoa, Ciliophora) on some zooplankton. Fresenius Environ. Bull. 26(11), 6334–6339 (2017).
Google Scholar 
Threlkeld, S. T., Chiavelli, D. A. & Willey, R. L. The organization of zooplankton epibiont communities. Trends Ecol Evol. 8, 317–321 (1993).Article 

Google Scholar 
Iyer, N. & Rao, T. R. Effect of epizoic rotifer Brachionus rubens on the population growth of three cladoceran species. Hydrobiologia 255(256), 325–332 (1993).Article 

Google Scholar 
Ramírez-Ballesteros, M., Fernandez-Leborans, G., Mayén-Estrada, R. New record of Epistylis hentscheli (Ciliophora, Peritrichia) as an epibiont of Procambarus (Austrocambarus) sp. (Crustacea, Decapoda) in Chiapas, Mexico. ZooKeys. 782, 1–9. https://doi.org/10.3897/zookeys.782.26417 (2018).Wu, H.X., Feng, M.G. Mass mortality of larval Eriocheir sinensis (Decapoda: Grapsidae) population bred under facility conditions: possible role of Zoothamnium sp. (Peritrichida: Vorticellidae) epiphyte. J. Invertebr. Pathol. 86, 59–60 (2004).Kumar, R. et al. Potential of three aquatic predators to control mosquitoes in the presence of alternative prey: a comparative experimental assessment. Mar. Freshw. Res. 59, 817–835 (2008).Article 

Google Scholar 
Kumar, R., Sami Souissi, S. & Hwang, J. S. Vulnerability of carp larvae to copepod predation as a function of larval age and body length. Aquaculture. 338, 274–283 (2012).Rao, T. R. & Kumar, R. Patterns of prey selectivity in the cyclopoid copepod Mesocyclops thermocyclopoides. Aquat. Ecol. 36, 411–424 (2002).Article 

Google Scholar 
Kumar, R. & Rao, T. R. Predation on Mosquito Larvae by Mesocyclops thermocyclopoides (Copepoda: Cyclopoida) in the Presence of Alternate Prey. Int Rev Hydrobiol. 88, 570–581 (2003).Article 

Google Scholar 
Baldrighi, E. et al. The cost for biodiversity: records of ciliate-nematode epibiosis with the description of three new Suctorian species. Diversity 12, 224. https://doi.org/10.3390/d12060224 (2020).Article 

Google Scholar 
Morado, J. F. & Small, E. B. Ciliate parasites and related diseases of Crustacea: a review. Rev. Fish. Sci. 3, 275–354 (1995).Article 

Google Scholar 
Lúcia, S. L., Safi Kam, W., Tang, Ryan, B. Carnegie. Investigating the epibiotic peritrich Zoothamnium intermedium Precht, 1935: Seasonality and distribution of its relationships with copepods in Chesapeake Bay (USA), Eur. J. Protistol. 84, 125880, https://doi.org/10.1016/j.ejop.2022.125880 (2022).Coats, D. W. & Heinbokel, J. F. A study of reproduction and other life cycle phenomena in planktonic protists using an acridine orange fluorescence technique. Mar. Biol. 67, 71–79. https://doi.org/10.1007/BF00397096 (1982).Article 

Google Scholar 
Montagnes, D. J. S. A Quantitative Protargol Stain (QPS) for Ciliates: method description and test of its quantitative nature. Mar. Microb. Food Webs. 2, 83–93 (1987).
Google Scholar 
Montagnes, D. J. S. & Lynn, D. H. A Quantitative Protargol stain (QPS) for ciliates and other protists. In Handbook of methods in aquatic microbial ecology (eds Kemp, P. et al.) 229–240 (Lewis Publishers, 1993).
Google Scholar 
Warren, A. Revision of the genus Vorticella (Ciliophora: Peritrichida). Bull. Br. Museum Nat. History 50, 48–52 (1986).
Google Scholar 
Foissner, W. et al. Intraclass evolution and classification of the Colpodea (Ciliophora). J. Eukaryot. Microbiol. 58, 397–415 (2011).Article 

Google Scholar 
Foissner, W., Berger, H. & Kohmann, F. Taxonomische und oekologische Revision der Ciliaten des Saprobiensystems—Band II: Peritrichia, Heterotrichida, Odontostomatida – Informationsberichte des Bayr. Landesamtes fuer Wasserwirtschaft. Heft 5(92), 1–502 (1992).
Google Scholar 
Santoferrara, L. F., Alder, V. V. & McManus, G. B. Phylogeny, classification and diversity of Choreotrichia and Oligotrichia (Ciliophora, Spirotrichea). Mol. Phylogenet. Evol. 112, 12–22 (2017).Article 

Google Scholar 
Hudson, P. L. et al. Cyclopoid and Harpacticoid Copepods of the Laurentian Great Lakes. Ohio Biol. Survey Bull. New Series. 12, 50 (1998).
Google Scholar 
Hudson, P. L et al. Cyclopoid copepods of the Laurentian Great Lakes US Geological Survey, Great Lakes Science Center, Ann Arbor, Michigan. Available: www.glsc.usgs.gov/greatlakescopepods/Key.asp (2003).Kumar, R., Muhid, P., Dahms, H. U., Sharma, J. & Hwang, J.-S. Biological mosquito control is affected by alternative prey. Zool. Stud. 54, 55. https://doi.org/10.1186/s40555-015-0132-9 (2015).Article 

Google Scholar 
Chesson, J. The estimation and analysis of preference and its relationship to foraging models. Ecology 64, 1297–1304 (1983).Article 

Google Scholar  More

Sets of interaction-associated mutants change across interactive conditionsTo investigate how microbial interactions are reorganized in a microbial community with increasing complexity, we reconstructed in vitro a modified bloomy rind cheese-associated microbiome on Cheese Curd Agar plates (CCA plates) as described in our previous work14 Growth as a biofilm on agar plates models the surface-associated growth of these communities, and allows inclusion of the filamentous fungus, P. camemberti, which grows poorly in shaken liquid culture. The original community is composed of the gamma-proteobacterium H. alvei, the yeast G. candidum and the mold P. camemberti. Using a barcoded transposon library of the model bacterium E. coli as a probe to identify interactions, we investigated microbial interactions in 2-species cultures (E. coli + 1 community member), in 3-species cultures (E. coli + 2 community members) and in 4-species cultures (or whole community: E. coli + 3 community members) (Fig. 1a).Figure 1Changes of E. coli’s genes associated with interaction-associated mutants in 2-species, 3-species and 4-species cultures. (a) Experimental design for the identification of interaction-associated mutants in 7 interactive conditions from the Brie community. The E. coli RB-TnSeq Keio_ML9 (Wetmore et al. 2015) is either grown alone or in 2, 3 or 4 species cultures to calculate E. coli gene fitness in each condition (in triplicate). Interaction fitness effect (IFE) is calculated for each gene in each interactive culture as the difference of the gene fitness in the interactive condition and in growth alone. IFE that are significantly different from 0 (two-sided t-test, Benjamini–Hochberg correction for multiple comparisons) highlight interaction-associated mutants in an interactive condition. (b) Volcanoplots of IFEs calculated for each interactive condition. Adjusted p-values lower than 0.1 highlight significant IFEs. Negative IFEs (blue) identify negative interactions and positive IFE (red) identify positive interactions. Numbers on each plot indicate the number of negative (blue) or positive (red) IFEs. (c) Functional analysis of the interaction-associated genes (significant IFEs). Genes of interaction-associated mutants have been separated into two groups: negative IFE and positive IFE. For each group, we represent the STRING network of the genes associated with interaction-associated mutants (Nodes). Edges connecting the genes represent both functional and physical protein association and the thickness of the edges indicates the strength of data support (minimum required interaction score: 0.4—medium confidence). Nodes are colored based on their COG annotation and the size of each node is proportional to the number of interactive conditions in which that given gene has been found associated with a significant IFE. Higher resolution of the networks with apparent gene names are found in Supplementary Figs. 2, 3.Full size imageQuantification of species’ final CFUs after 3 days of growth highlighted consistent growth for H. alvei and G. candidum independent of the culture condition and slightly reduced growth for E. coli in interactive conditions compared to growth alone (Dunnett’s test against growth alone; adjusted-p value ≤ 5%) except for the 2-species growth with P. camemberti (Supplementary Fig. 1). Although we were unable to quantify spores of P. camemberti after three days, growth of P. camemberti was visually evident in all of the expected samples. Quantitative analysis of E. coli’s library final growth using an epistatic model highlighted that the growth of E. coli in the 3-species and 4-species condition can be predicted from the corresponding 2-species growths (Supplementary Fig. 1).Previously, we developed an assay and a pipeline to identify microbial genes associated with interactions by adapting the original RB-TnSeq approach19 to allow for consistent implementation of biological replicates as well as for direct quantitative comparison of fitness values between different culture conditions15. More specifically, the original RB-TnSeq assay relies on the use of a dense pooled library of randomly barcoded transposon mutants of a given microorganism (RB-TnSeq library)19 containing multiple insertion mutants for each gene as well as intergenic insertion mutants. Measuring the variation of the abundance of each transposon mutant before and after growth, the pipeline allows the calculation of a fitness value for each insertion-mutant as well as a fitness value for each gene corresponding to the average of the insertion-mutants’ fitness of the associated genes across biological replicates. A negative fitness indicates that disruption of this gene decreases growth of the mutant relative to a wild type strain, whereas a positive fitness value indicates increased growth in the studied condition. Then, we infer the interactions based on the effects of insertion-mutants between interactive growth and growth alone. In other words, we measure and compare gene fitness across the different studied conditions. Any significant change in fitness values identifies an interaction-associated mutant. The subsequent analysis of interactions, including the inference of the interaction mechanisms and the comparison of interactions across the different interactive conditions, is mainly based on the nature of the disrupted genes by the transposon and their characterized function. Also, by measuring interactions as the difference of fitness value of a given gene between growth with other species and growth alone, we consider that interactions between insertion-mutants of the RB-TnSeq library are controlled and included in our calculation. Then, any interaction-associated mutant predominantly identifies inter-species interactions.In this work, we used the E. coli RB-TnSeq Keio_ML9 library19 and grew it for 3 days alone or in the seven different interactive conditions studied here (Fig. 1a). This library contains 152,018 pooled insertion mutants with an average of 16 individual insertion mutants per gene and many intergenic insertion mutants. For each interactive condition, we calculated the Interaction Fitness Effect (IFE) associated with 3699 E. coli genes as the difference between the gene fitness in the studied interactive condition and the gene fitness in growth alone (Supplementary Data 1). Negative IFE occurs when gene fitness decreases in the interactive condition, and positive IFE occurs when gene fitness improves in the interactive condition. We then tested for all the IFEs that are significantly different from 0 (adjusted p-value ≤ 0.1; two-sided t-test and Benjamini–Hochberg correction for multiple comparison20) to screen for interactions and to identify, in each condition, the insertion-mutants that are associated with inter-species interactions. Here, we identified between 6 (with P. camemberti) and 71 (with H. alvei + P. camemberti) significant IFEs per condition (Fig. 1b). Both negative IFEs and positive IFEs were found in each interactive condition except for the 2-species culture with P. camemberti, where only negative interactions were identified. A total of 330 significant IFEs associated with 218 unique genes were identified (as the same gene can be associated with a significant IFE in multiple conditions) including 125 genes associated with negative IFE and 120 genes associated with positive IFE (Supplementary Figs. 2, 3). Altogether, we didn’t notice any strong correlation between the number and type of IFE identified by condition and the overall growth impact measured on E. coli.
To gain insight into the interaction mechanisms among microbes, we next analyzed the functions of the genes of the interaction-associated mutants (i.e., genes associated with a significant IFE). Here, the vast majority of the genes associated with interaction-associated mutants are part of an interaction network (Fig. 1c). These STRING networks connect genes that code for proteins that have been shown or are predicted to contribute to a shared function, with or without having to form a complex21. A significant fraction of the interaction-associated mutants associated with a negative IFE are part of amino acid biosynthesis and transport (17%—Fig. 1c and Supplementary Figs. 2, 4), and more specifically with histidine, tryptophan and arginine biosynthesis. This points to competition for these nutrients between E. coli and the other species. Another large set of interaction-associated mutants is related to nucleotide metabolism and transport (14%—Fig. 1c and Supplementary Figs. 2, 5), highlighting competitive interactions for nucleotides and/or their precursors. The majority of the associated genes relate to purine nucleotides and more specifically to the initial steps of their de novo biosynthesis associated with the biosynthesis of 5-aminoimidazole monophosphate (IMP) ribonucleotide. Of the genes associated with interaction-mutants with a positive IFE, 15% are related to amino acid biosynthesis and transport (Fig. 1c and Supplementary Figs. 3, 4), suggesting cross feeding of amino acids between E. coli and the other species. More specifically, this includes phosphoserine, serine, homoserine, threonine, proline and arginine. The presence of amino acid biosynthetic genes among both negative and positive IFEs indicate that trophic interactions (competition versus cross-feeding) depend on the type of amino-acid and/or the species interacting with E. coli. For both negative and positive IFEs, numerous genes of the associated interaction-mutants were annotated as transcriptional regulators (Fig. 1c and Supplementary Figs. 2, 3) emphasizing the importance of transcriptional reprogramming in response to interactions. These transcriptional regulators include metabolism regulators as well as regulators of growth, cell cycle and response to stress. Finally, these interaction-associated mutants and the infered interaction mechanisms are consistent with previous findings in this microbiome14 as well as in a study of bacterial-fungal interactions involving E. coli and cheese rind isolated fungal species15. While this approach allows us to infer the interaction mechanisms that are happening between the transposon library and the other species, further experimental validation would be needed to confirm that these interactions more generally happen between a WT strain and the other species.Introduction of a third interacting species deeply reshapes microbial interactionsThe differences in the number and sign of significant IFEs observed among the different interactive conditions, with different numbers of interaction species, suggest that the number and type of interacting partners influence interaction mechanisms. To characterize how the interactions are reorganized with community complexity, we then investigated if and how the genetic basis of interactions changes when the number of interacting partners increases by comparing the genes associated with interaction-associated mutants with significant IFE in 2-species cultures, in 3-species cultures and then in 4-species cultures.First, we have identified 104 IFEs associated with 98 genes in 2-species cultures as well as 168 IFEs associated with 136 unique genes in 3-species conditions (Supplementary Fig. 6 and Supplementary Data 2). Comparing these gene sets, we can identify how the interaction-associated mutants change when a third-species is added to a 2-species culture. We identified 45 genes associated with 2-species interaction-associated mutants maintained in at least one 3-species condition (maintained interaction-mutants), 55 genes associated with 2-species interaction-associated mutants no longer associated with interaction in any 3-species condition (dropped interaction-mutants) and 100 genes associated with 3-species interaction-associated mutants that aren’t related to any 2-species interaction-associated mutants (emergent interaction-mutants) (Fig. 2a, Supplementary Fig. 6 and Supplementary Data 3). Both dropped and emerging interaction-associated mutants represent 3-species HOIs; the third species either removes an existing interaction or brings about a new one.Figure 2Comparison of the genetic basis of interaction for 2-species and 3-species conditions. (a) Venn Diagram of 2-species and 3-species sets of genes related to interaction-associated mutants. This Venn Diagram identifies 2-species interaction-mutants that are dropped when a third species is introduced (Left side; Dropped interaction-mutants = any 2-species gene that is not found in any 3-species condition), 2-species interaction-mutants that are maintained in at least one associated 3-species condition (Intersection; Maintained interaction-mutants) and interaction-mutants that are specific to 3-species condition (Right side; Emerging interaction-mutants). (b) Functional analysis of the genes associated with dropped, maintained and emerging interaction-mutants from 2-species to 3-species. Each dot represents the fraction of genes of the studied gene set associated with a given COG category (Number of genes found in the category / Total number of genes in the gene set). The color of the dots indicates the general COG group of the COG category: Teal: Metabolism; Blue: Information storage and processing; Orange: Cellular Processes and Signaling; Grey: Unknown or no COG category. (c) Species-level analysis of 3-species HOIs: for each 2-species condition, we measure the fraction of interaction-mutants that are dropped in associated 3-species cultures (Dropped in 3-species) or maintained in at least one of the 3-species cultures (Maintained in 3-species); for each 3-species condition, we measure the fraction of interaction-mutants that have been conserved from at least one associated 2-species condition (Maintained from 2-species) or that are emerging with 3-species (Emerging in 3-species).Full size imageWe further carried out functional analysis of the genes related to maintained, dropped and emerging interaction-mutants to elucidate whether maintained and HOIs interaction-mutants would be associated with specific functions and thus interaction mechanisms (Fig. 2b). For each set of genes, we calculated the fraction of genes of that set associated with a given COG ontology category. Metabolism and transport is the most observed COG group (Fig. 2b—teal dots). For genes related to maintained interaction-mutants, this indicates that some trophic interactions can be maintained from 2-species to 3-species conditions. For instance, serine biosynthetic genes serA, serB and serC as well as threonine biosynthetic genes thrA, thrB and thrC are associated with positive IFEs in the 2-species condition with G. candidum as well as in the 3-species conditions involving G. candidum (Supplementary Fig. 4). This suggests that, (i) G. candidum facilitates serine and threonine cross feeding and (ii) this cross-feeding is still observed when another species is introduced. However, metabolism-related genes identified among the dropped and emerging interaction-mutants indicate that many trophic interactions are also rearranged through HOIs. Genes associated with lactate catabolism (lldP and lldD) and lactate metabolism regulation (lldR) have a negative IFE in the 2-species culture with H. alvei, suggesting competition for lactate between E. coli and H. alvei. Yet, mutants of these genes are no longer associated with a significant IFE when at least another partner is introduced (Supplementary Fig. 7). Histidine biosynthesis genes hisA, hisB, hisD, hisH and hisI are associated with interaction-mutants with negative IFE in the 2-species culture with H. alvei and sometimes in the 3 species culture with H. alvei + P. camemberti. However, the negative IFE is alleviated whenever G. candidum is present, suggesting that potential competition for histidine between E. coli and H. alvei is alleviated by this fungal species (Supplementary Fig. 4). Also, genes related to the COG section “Information storage and processing” are mostly found among genes of HOIs-mutants suggesting a fine-tuning of specific cellular activity depending on the interacting condition. For instance, we identified many transcriptional regulators of central metabolism among the dropped interaction-mutants genes (rbsR and lldR) and the emerging interaction-mutants genes (purR, puuR, gcvR and mngR), highlighting again the reorganization of trophic interactions associated with HOIs. Also, many transcriptional regulators broadly associated with growth control, cell cycle and response to stress were found among the emerging interaction-mutants genes with 3-species (hyfR, chpS, sdiA, slyA and rssB), underlining a noticeable modification of E. coli’s growth environment with 3-species compare to with 2-species.Finally, we further aimed to understand whether HOIs are associated with the introduction of any specific species (Fig. 2c and Supplementary Fig. 8). We observe that interaction-associated mutants with H. alvei are more likely to be dropped, as 65% of them are alleviated by the introduction of a fungal species (Fig. 2c). This can be seen, for instance, with the reorganization of E. coli and H. alvei trophic interactions following the introduction of G. candidum (alleviation of lactate and histidine competition for instance). Also, we observe that 76% of the interactions in the 3-species cultures with H. alvei + P. camemberti and 65% in the 3-species culture with H. alvei + G. candidum are emerging interaction-mutants (compared to 38% of emerging interaction-associated mutants in the 3-species condition with G. candidum + P. camemberti) (Fig. 2c). For the interaction-associated mutans found in the 3-species with H. alvei + P. camemberti, they include for instance the genes associated with purine de novo biosynthesis (purR, purF, purN, purE, purC) and the genes associated with pyrimidine de novo biosynthesis (pyrD, pyrF, pyrC, carA and ulaD), suggesting important trophic HOIs. For the 3-species condition with H. alvei + G. candidum, emerging interaction-mutants include for example the transcriptional regulator genes chpS, sdiA and slyA, indicating the presence of a stress inducing environment. Together, these observations suggest that the introduction of a fungal partner may introduce multiple 3-species HOIs by both canceling existing interactions and introducing new ones.HOIs are prevalent in a 4-species communityTo further decipher whether microbial interactions continue to change with increasing community complexity, we investigated the changes in the genetic basis of interactions going from 3-species to 4-species experiments. We identified 58 interaction-associated mutants in the 4-species condition (E. coli with H. alvei + G. candidum + P. camemberti), compared with 145 interaction-associated mutants in any 3-species condition. Comparing the two sets of interaction-associated mutants and corresponding genes we identify: 26 3-species interaction-mutants that are maintained in the 4-species condition (including 16 directly from 2-species interactions), 115 3-species interaction-mutans that are no longer associated with interactions in the 4-species condition (dropped interaction-mutants) and 32 interaction-mutants that are observed solely in the 4-species condition (emerging interaction-mutants) (Fig. 3a, Supplementary Fig. 6 and Supplementary Data 3). Both dropped and emerging interaction-mutants represent 4-species HOIs. Here, HOIs are remarkably abundant when introducing a single new species and moving up from 3-species interactions to 4-species interactions. Functional analysis of the genes of maintained-mutants and HOI-mutants reveals the presence of many metabolism related genes in every gene set (Fig. 3), suggesting that some trophic interactions can be maintained from 3-species to 4-species interactions while some other trophic interactions are rearranged with HOIs. For instance, most of the genes of the initial steps of de novo purine biosynthesis have been found to be associated with a negative IFE in the 3 species condition with H. alvei + P. camemberti (purC, purE, purF, purL and purN) as well as in the pairwise condition with H. alvei for purH and purK (Supplementary Fig. 5), suggesting competition for purine initial precursor IMP in these conditions. Yet, the introduction of the yeast G. candidum as a fourth species cancels the negative IFE value, suggesting that the competition is no longer happening in its presence. Altogether, the observation of noticeable trophic HOIs moving up from 2 to 3 species and then from 3 to 4-species interaction highlights a consistent reorganization of trophic interactions along with community complexity. Also, genes related to Cell wall/membrane/envelope biogenesis are found abundantly among the 4-species emerging-mutants (Fig. 3b) and they represent the largest functional fraction of this gene set. These genes are associated with a negative IFE and are related to Enterobacterial Common Antigen (ECA) biosynthetic processes (wecG, wecB and wecA) (Supplementary Fig. 9). While the roles of ECA can be multiple but are not well defined22, they have been shown to be important for response to different toxic stress, suggesting the development of a specific stress in the presence of the four species.Figure 3Organization of the interactions in the 4-species community. (a) Venn Diagram of 3-species and 4-species sets of genes related to interaction-associated mutants. This Venn Diagram identifies 3-species interaction-mutants that are dropped when a fourth species is introduced (Left side; Dropped interaction-mutants = any 3-species interaction-associated mutant that is not found in the 4-species condition), 3-species interaction-mutants that are maintained in the 4-species condition (Intersection; Maintained interaction-mutants) and interaction-mutants that are specific to 4-species condition (Right side; Emerging interaction-mutants). (b) Functional analysis of the genes associated with dropped, maintained and emerging interaction-mutants from 3-species to 4-species. Each dot represents the fraction of genes of the studied gene set associated with a given COG category (Number of genes found in the category/Total number of genes in the gene set). The color of the dots indicates the general COG group of the COG category: Teal: Metabolism; Blue: Information storage and processing; Orange: Cellular Processes and Signaling; Grey: Unknown or no COG category. (c) Species-level analysis of 4-species HOIs: for each 3-species cultures we measure the fraction of interaction-genes that is conserved in the 4-species culture (Maintained in 4-species) and the fraction of interaction-genes that has been dropped (Dropped in 4-species). (d) Alluvial plots of the interaction genes across community complexity levels. (e) STRING network of the 4-species interaction genes (Nodes). Edges connecting the genes represent both functional and physical protein association and the thickness of the edges indicates the strength of data support (minimum required interaction score: 0.4—medium confidence). Nodes are colored based on the level of community complexity the genes are conserved from.Full size imageAs for the 2 to 3 species comparison, we investigated whether the introduction of a specific fourth species would be most likely associated with HOIs. The 3-species culture that appears to be the least affected by the introduction of a fourth member is with G. candidum + P. camemberti where 34% of the observed interactions are still conserved in the 4-species condition after the introduction of H. alvei (versus 22% for with H. alvei + G. candidum when P. camemberti is added and 21% for with H. alvei + P. camemberti when G. candidum is added) (Fig. 3c and Supplementary Fig. 10). Together, these observations suggest that, again, the introduction of a fungal partner may introduce multiple 4-species HOIs.Finally, by increasing the number of interacting species in our system and investigating interaction-mutants maintenance and modification with every increment of community complexity, we are able to build our understanding of the architecture of interactions in a microbial community. Altogether, we have observed a total of 218 individual interaction-associated mutants in any experiment. Only 16 of them (7%) were conserved across all levels of community complexity (Fig. 3d). Starting from 2-species interaction-mutants, 48% of them were maintained with 3-species and only 15% (16 out of 104) were still maintained with 4-species. Thus, we demonstrate here a progressive loss and replacement of 2-species interactions as community complexity increases and the prevalent apparition of HOIs. Tracking back the origins of the genetic basis of interactions in the 4-species experiment that represents the full community of our model, we identify that 28% of the full community interactions can be traced back to 2-species interactions, 18% are from 3-species interaction and 54% are specific to the 4-species interaction (Fig. 3d,e). Most of the maintained interaction-mutants from 2-species as well as from 3-species are associated with metabolism (Fig. 3d and Supplementary Fig. 11) while Signal transduction and cell membrane biosynthesis genes are most abundant among the 4-species interaction-mutants as previously mentioned. To conclude, this shows that the genetic basis of interactions and thus the sets of microbial interaction are deeply reprogrammed at every level of community complexity and illustrates the prevalence of higher order interactions (HOIs) even in simple communities.The majority of maintained 2-species interaction-mutants in the 4-species culture follows an additive conservation behaviorWhile HOIs are abundant in the 4-species condition, our data yet suggest that up to 28% of the interactions are maintained from 2-species interactions. However, we don’t know whether and how 2-species interactions are quantitatively affected by the introduction of other species and whether they would follow specific quantitative models of conservation. For instance, we can wonder how the strength of a given 2-species interaction is modified by the introduction of one or two other species, or how two 2-species interactions associated with the same gene will combine when all the species are present. In other words, can we treat species interactions as additive when we add multiple species? Such information would generate a deeper mechanistic understanding of the architecture of microbial interactions while allowing us to potentially predict some whole community interactions from 2-species interactions. Here, two main hypothetical scenarios can be anticipated. First, the conservation of 2-species interactions follows a linear or additive behavior, where the introduction of other species either doesn’t affect the strength of the conserved 2-species interaction or two similar 2-species interactions combine additively. The second scenario identifies non-linear or non-additive conservation of 2-species interactions, where the strength of the conserved 2-species interaction is modified by the introduction of other species or two similar 2-species interactions are not additive. The second scenario would encompass for instance synergistic effects or inhibitory effects following the introduction of more species. We next use an epistasis and quantitative genomics approach to understand whether interactions that are conserved follow a linear, or additive, pattern. For the 16 interaction-associated mutants that are associated with interaction in 2-species cultures, in associated 3-species cultures and in the 4-species condition, we use epistasis analysis to test the linear behavior of their IFE when the number of interacting species increases, as IFEs are quantitative traits related to the interaction strength. In multi-dimensional systems, an epistasis analysis quantifies the additive (or linear) behavior of conserved quantitative traits. In quantitative genetics, for instance, epistasis measures the quantitative difference in the effects of mutations introduced individually versus together18,23,24. Using a similar rationale, we can use IFEs as a quantitative proxy for interaction strength and test whether the IFEs of the maintained interaction genes in 3-species and in 4-species conditions result from the linear combination of associated 2-species IFEs (Fig. 4a). Nonlinear combination, or non-additivity of 2-species IFEs in higher community level also highlights higher-order interactions.Figure 4Quantitative analysis of IFE conservation for the interaction-associated mutants conserved from 2-species to 4-species conditions. (a) Schematized quantitative epistasis/non-linearity measured in 3-species conditions (with partner i and j). Epistasis (εij) is the difference between the individual IFE of partner i and partner j (red and orange bars) versus placing them together (green). Mathematically, we need three terms (IFEi, IFEj, and εij) to reproduce the observed IFE for the 3-species condition. (b) This analysis can be extended to higher levels of community complexity: 4-species (E. coli with 3-partners i, j, and k). The model first accounts for epistasis between i/j, i/k, and j/k. In this example, i and j exhibit epistasis; i/k and j/k are additive (dark blue and purple). The predicted IFE for the 4-species community is the sum of the individual 2-species effects (red, orange, light blue) and the 3-species epistatic terms (green). The 4-species epistatic coefficient is the difference between this low-order prediction and the observed IFE for the i,j,k community (pink). (c) Conservation profiles of the 16 2-species interaction-associated mutants conserved up to 4-species. 2-species conditions: a colored square indicates the 2-species condition(s) in which the interaction-associated mutant was identified; a grey square indicates non-significant 2-species IFEs. 3-species conditions: a teal square indicates that the associated IFE is associated with additive behavior from associated 2-species IFE (no εij epistatic coefficient), a red square indicates that the associated IFE displays non-additivity from 2-species IFE and thus epistasis, a grey square corresponds to a 3-species condition that is not associated with significant 2-species IFE (no epistasis analysis performed); 4-species condition: a teal square indicates that the associated IFE is associated with additive behavior (no εijk epistatic coefficient) , a red square indicates that the associated IFE is associated with non-additivity from lower-order IFE. (d) Comparison of the observed and predicted IFE for the genes and condition associated with 3-species and 4-species non-additive IFE.Full size imageWe adapted the pipeline Epistasis17, originally designed for quantitative genetics investigation. We implemented the linear model with the gene fitness values of the interaction-associated mutants for growth alone, for each of the 2-species conditions, for each of the 3-species cultures and for the 4-species condition. For each gene, the software finds the simplest mathematical model that reproduces the observed IFEs across all levels of community complexity. In the simplest case, the model will have a term describing the effects for adding each species individually to the E. coli alone culture; that term corresponds to the 2-species IFE. Then, if the IFE for two E. coli’s partners combined (3-species IFE) differs from the sum of their individual effects (corresponding 2-species IFE), the software adds a term capturing this epistasis (Fig. 4a). Here, we call that term 3-species epistatic coefficient or εi,j. Finally, if the IFE for the combined community (E. coli plus all three species; 4-species condition) differs from the prediction based on the 2-species and 3-species terms, the software will add a high-order interaction term to the model (Fig. 4b). Here, we name that term 4-species epistatic coefficient or εijk.We performed this analysis on the 16 interaction-associated mutants that are associated with interactions at every level of community complexity. To identify real additive behavior of IFE from non-additivity, we screen for 3-species epistatic coefficients and 4-species epistatic coefficients that are significantly different from 0 (adjusted p-value ≤ 0.01, Benjamini–Hochberg correction for multiple testing). We found that 13 interaction-associated mutants behaved additively from 2-species to 4-species culture, with no epistatic contributions in the 3-species conditions nor in the 4-species condition (Fig. 4c, (i)). One interaction-associated mutant (gene (gadW)) exhibited nonlinear conservation of IFE only in the 4-species condition, but additive IFE conservation from 2-species to 3-species (Fig. 4c, (ii)). Another interaction-associated mutant (gene (lsrG)) showed epistasis in one 3-species condition but no epistasis in the 4-species condition (Fig. 4c, (iii)) Finally, one interaction-associated mutant (gene (gltB)) displayed both non-additivity in 3-species and 4-species conditions (Fig. 4c, (iv)). If we look more closely at the genes related to interaction-associated mutant with an additive behavior, we find genes (betA, betT, purD and purH) that are associated with the conservation of negative IFEs (Supplementary Fig. 12). While betA and betT are associated with choline transport (betT) and glycine betaine biosynthesis from choline (betA)25, purD and purH are associated with de novo purine biosynthesis26. This suggests that requirements for glycine betaine biosynthesis from choline and for purine biosynthesis caused by microbial interactions, possibly due to competition for the nutrients used as precursors, are additively conserved from individual 2-species interactions requirements. Also, 5 genes associated with amino acid biosynthesis (serA, thrC, cysG, argG and proA) are associated with the additive conservation of positive IFE (Supplementary Fig. 12), suggesting that cross feeding can be additive when the community complexity increases. Altogether, this highlights the existence of 2-species interactions, including trophic ones, conserved in an additive fashion in the highest-level of complexity.This leaves 3 interaction-associated mutants (18%) of the maintained 2-species interaction-mutants, that are associated with non-additive behavior, and thus HOIs, at at least one higher level of community complexity (Fig. 4c—(ii), (iii) and (iv)). The interaction-associated mutant for the gene gadW is associated with non-additivity at the 4-species level, suggesting that while IFEs are additive in 3-species cultures, the introduction of a fourth species introduces HOI. Moreover, the observed 4-species IFE is greater than the IFE predicted by a linear model (Fig. 4d), highlighting a potential synergistic effect when the 4 species are together. The interaction-assoacited mutant for the gene lsrg is associated with non-additivity only at the 3-species culture w G.c + P.c. More specifically, this indicates that HOI arise when these 2 fungal species are interacting together with E. coli, but that no more HOI emerge when H. alvei is introduced (i.e., the 4-species IFE can be predicted by the linear combination of the lower levels IFEs). As the observed IFE for the 3-species condition w G.c + P.c is greater than the predicted IFE (Fig. 4c), this suggests a synergistic effect between the 2 fungal species. Finally, the interaction-associated mutants for the gene gltB is associated with non-additivity at both the 3-species and 4-species levels. For this interaction-associated mutant, the conservation of IFE is never associated with an additive model. Here, the observed 4-species IFE is not as negative as it would be as the result of the linear combination of the associated lower IFE (Fig. 4d), suggesting the existence of a possible IFE threshold, or plateau effect. Altogether, this indicates that maintained 2-species-interactions can follow nonlinear behaviors that could involve synergistic effects, inhibitory effects or constraints. More

In this section, we introduce flapping flight dynamics and describe the bird model used in our computational framework. Furthermore, we describe how such a dynamical model is used in order to identify steady and level flapping flight regimes, study their stability, and assess their energetic performance.Equations of motion modelling flapping flightFlight dynamics is restricted to the longitudinal plane and thus the bird main body is captured as a rigid-body with three degrees of freedom, i.e. two in translation and one in rotation. This model preserves symmetry with respect to this plane, without any lateral force and moment. The aerodynamic model of the wing relies on the theory of quasi-steady lifting line23. Additionally, the present work does not account for the inertial forces due to the acceleration of the wing, and thus also neglecting the so-called inertial power. This inertial power was shown to be negligible in fast forward flight conditions, in comparison to the other contributions to actuation power24, and is thus systematically neglected in similar work10,11,25 since wing inertia is neglected.The body is thus modelled with a mass (m_b) and a rotational inertia (I_{yy}) about its center of mass. The equations of motion are expressed in the body frame (G(x’, y’, z’)) with unit vectors ((hat{textbf{e}}_{x’}, hat{textbf{e}}_{y’}, hat{textbf{e}}_{z’})), and an origin located at the center of mass, as pictured in Fig. 1a. The state space vector is thus$$begin{aligned} textbf{x} = {u, w, q, theta } end{aligned}$$where u and w are the body velocities along the (x’-) and (z’-)axis and (theta) and q are the pitch angle and its time derivative about the (y’-)axis, respectively. Consequently, the equations of motion read11,13,26$$begin{aligned} begin{aligned} dot{u}&= -qw – gsin theta + frac{1}{m_b}big ( {F_{x’}(textbf{x}(t), t)} \&quad + {F_{x’, t}(textbf{x}(t), t)} + D (textbf{x}(t), t) big )\ dot{w}&= qu + gcos theta +frac{1}{m_b} big ( F_{z’}(textbf{x}(t), t) + F_{z’, t}(textbf{x}(t), t) big ) \ dot{q}&=frac{1}{I_{yy}} big ( M_{y’}(textbf{x}(t), t) + M_{y’, t}(textbf{x}(t), t) big )\ dot{theta }&= q end{aligned} end{aligned}$$
(1)
Figure 1(a) Bird model for describing the flight dynamics in the longitudinal plane. The state variables are expressed with respect to the moving body-frame located at the flier’s center of mass (G(x’,z’)). These state variables are the component of forward flight velocity, u, the velocity component of local vertical velocity, w, the orientation of this body-centered moving frame with respect to the fixed frame, (theta) and its angular velocity, q. A second frame (O(x’_{w}, z’_{w})) is used to compute the position of the wing, relative to the body. The wings (dark gray) and the tail (red) are the surfaces of application of aerodynamic forces. (b) Top view of the bird model. The left wing emphasizes a cartoon model of the skeleton. The shoulder joint s connects the wing to the body via three rotational degrees of freedom (RDoF), the elbow joint e connects the arm with the forearm via one RDoF and the wrist joint w connects the forearm to the hand via two RDoF. Each feather is attached to a bone via two additional RDoF, except the most distal one ”1” which is rigidly aligned with the hand. The right wing further emphasizes the lifting line (red) which is computed as a function of the wing morphing. The aerodynamic forces generated on the wing are computed on the discretized elements (P_{i}). The tail is modelled as a triangular shape with fixed chord (c_{t}) and maximum width (b_{t}) that can be morphed as a function of its opening angle (beta). (c) Wing element i between two wing profiles, identifying a plane (Sigma) containing the lifting line (red). (d) Cross section of the wing element, containing the chord point (mathbf {P_i}) where the velocities are computed (Color figure online).Full size imageThe forcing terms in Eq. (1) are the aerodynamic forces and moments applied to the wing (namely (F_{x’}), (F_{z’}), and (M_{y’}) ) and to the tail ((F_{x’, t}), (F_{z’, t}), and (M_{y’, t})). The whole drag is captured by an extra force D that sums contributions due to the body (D_{b}), the skin friction of the wing (wing profile) (D_{p,w}), and the skin friction of the tail (tail profile) (D_{p,t}). These terms are described in detail in the next sections. Importantly, we accounted for the drag acting purely along (x’) direction, after proving that the projection of the drag forces along (z’)-axis is between two and three orders of magnitude smaller with respect to the aerodynamic forces produced by two other main lifting surfaces. This assumptions holds for the fast forward flight regime that are subject of our study, but such components of drag along (z’) axis should be accounted for other flight situations.Wing modelThe bird has two wings. Each wing is a rigid poly-articulated body, comprising the bird arm, forearm and hand, as pictured in Fig. 1b. Each segment is actuated by a joint to induce wing morphing. We refer to13,15 for a complete description of this wing kinematic model.Each joint is kinematically driven to follow a sinusoidal trajectory specified as:$$begin{aligned} q_{i}(t) = q_{0,i}(t) + A_{i} sin (omega t + phi _{0,i}) end{aligned}$$
(2)
with (omega = 2 pi f) and f being the flapping frequency which is identical for each joint, (q_{0,i}) being the mean angle over a period (or offset), (A_i) the amplitude, and (phi _{0,i}) the relative phase of joint i. A complete wingbeat cycle is therefore described through a set of 19 kinematic parameters, including the frequency f.We assume that the wing trajectory is rigidly constrained, and therefore we do not need to explicitly solve the wing dynamics. Under this assumption, the motion generation does not require the computation of joint torques. The model further embeds seven feathers of length (l_{ki}) in each wing. The feathers in the model have to be considered a representative sample of the real wing feathers. They thus have a limited biological relevance; their number is chosen so as to interpolate the planform satisfactorily and to smoothly capture the morphing generated by the bone movements. These feathers are attached to their respective wing bones via two rotational degrees of freedom allowing them to pitch and spread in the spanwise direction. These two degrees of freedom are again kinematically driven by relationships that depend on the angle between the wing segments13. This makes the feathers spreading and folding smoothly through the wingbeat cycle. In sum, the kinematic model of the wing yields the position of its bones and feathers at every time step. This provides a certain wing morphing from which the wing envelope (leading edge and trailing edge) can be computed (see Fig. 1b). From the wing envelope, the aerodynamic chord and the lifting line are computed. The lifting line is the line passing through the quarter of chord, which is itself defined as the segment connecting the leading edge to the trailing edge and orthogonal to the lifting line (Fig. 1b). This extraction algorithm is explained in detail in15.In order to calculate the aerodynamic forces, the angle of attack of the wing profile has to be evaluated. Each wing element defines a plane containing the lifting line and the aerodynamic chord as pictured in Fig. 1c. The orientation of the plane is identified by the orthogonal unit vectors ((hat{textbf{e}}_n, hat{textbf{e}}_t, hat{textbf{e}}_b)), where (hat{textbf{e}}_n) is the vector perpendicular to the plane and (hat{textbf{e}}_t) is the tangent to the lifting line. To compute the effective angle of attack, the velocity perceived by the wing profile is computed as the sum of the velocities due to the body and wing motion, and the velocity induced by the wake. The first contribution, (textbf{U}), accounts for$$begin{aligned} textbf{U} = textbf{U}_{infty } – textbf{v}_{kin} – textbf{v}_{q}end{aligned}$$where (textbf{U}_{infty } = u hat{textbf{e}}_{x’} + w hat{textbf{e}}_{z’}) is the actual flight velocity, (textbf{v}_{kin}) is the relative velocity of the wing due to its motion, and (textbf{v}_{q}) is the component induced by the angular velocity of the body q and calculated as$$begin{aligned} textbf{v}_{q} = qhat{textbf{e}}_{y’} wedge (textbf{P}_{i} – textbf{G})end{aligned}$$This velocity vector (textbf{U}) defines the angle (alpha), as pictured in Fig. 1d.The second contribution is due to the induced velocity field by the wake, i.e. the downwash velocity (w_{d}), and acting along the normal unit vector (-w_{d}hat{textbf{e}}_n). The resulting effective angle of attack, (alpha _{r}), is thus$$begin{aligned} alpha _{r} = alpha – frac{w_{d}}{|textbf{U}|}end{aligned}$$The downwash velocity (w_d) is computed according to the Biot-Savart law23, assuming the wake being shed backwards in the form of straight and infinitely long vortex filaments at each time step of the simulation13,15. This quasi-steady approximation is justified a posteriori by ensuring that our reduced frequency, inversely proportional to the unknown airspeed, never exceeds the value of 0.2, below which the effects of time-dependent wake shapes on wing circulation are negligible (e.g. see discussion in27). Once the downwash is evaluated, it is possible to evaluate the circulation, and consequently the aerodynamic force and moment acting at the element (P_i), i.e. (F_{x’, i}(textbf{x}(t), t), F_{z’, i}(textbf{x}(t), t), M_{y’, i}(textbf{x}(t), t)), as explained in detail in13. We use the thin airfoil theory for the estimation of the lift coefficient, with a slope of (2pi) that saturates at an effective angle of attack (alpha _{r}) of (pm 15^{circ }).Drag production by body and wingThe main body and the wings induce drag that should be accounted for in a model aiming at characterizing energetic performance. Body-induced drag is named parasitic because the body itself does not contribute to lift generation, and only induces skin friction and pressure drag around its envelope28. The total body drag is$$begin{aligned} D_{b} = frac{1}{2}rho C_{d, b} S_{b}|textbf{U}_{infty }|^{2} end{aligned}$$
(3)
where (rho) is the air density. We implemented the model described by Maybury28 to compute the body drag coefficient (C_{d, b}). This depends on the morphology of the bird and the Reynolds number Re according to$$begin{aligned} C_{d,b} = 66.6m_{b}^{-0.511}FR_{t}^{0.9015}S_{b}^{1.063}Re^{-0.197} end{aligned}$$
(4)
with (S_{b}) and (FR_{t}) are respectively the frontal area of the body and the fitness ratio of the bird, and both of them can be estimated from other allometric formulas i.e.28,29.$$begin{aligned} S_{b}= & {} 0.00813m_{b}^{2/3} end{aligned}$$
(5)
$$begin{aligned} FR_{t}= & {} 6.0799m_{b}^{0.1523} end{aligned}$$
(6)
The Reynolds number (Re = rho |textbf{U}_{infty }| overline{c} / mu) is calculated with the reference length of the mean aerodynamic chord (overline{c}), with (mu) being the dynamic viscosity. This model is found to be suitable for Reynolds number in the range (10^{4}-10^{5})28. Another source of drag is the profile drag due to friction between the air and the feathers on the wings. It is the sum of the profile drag at each section along the wingspan, i.e.$$begin{aligned} D_{p,w} = frac{1}{2} rho C_{d, pro} sum _{i=1}^{n} c_{i}|textbf{U}_{r,i}|^{2} ds_{i} end{aligned}$$
(7)
with (c_{i}) the chord length, (ds_{i}) the length of the lifting line element along the wingspan, and (textbf{U}_{r,i}) the velocity at the wing section i accounting for the body velocity, the kinematics velocity of the wing and the downwash velocity (Fig. 1c,d). We used a value of profile drag of (C_{d, pro} = 0.02) and this is assumed to be constant over the wingspan and throughout the flapping cycle30. In reality, due to the wing motion, this value should be gait dependent. However, the aforementioned assumption has been largely used in previous works31,32.Tail modelSince the span of the tail is of the same magnitude as its aerodynamic chord, here the lifting line approach cannot be used23. Therefore, the tail is modelled according to the slender delta wing theory, as a triangular planform33. Its morphology is illustrated in Fig. 1b and characterized by the opening angle (beta) and the chord (c_t). This latter parameter is kept constant, thus the tail span is controlled via (beta) from the trigonometrical relationship$$begin{aligned} b_{t} = 2c_{t}tan frac{beta }{2}end{aligned}$$The main limitation of this framework is the low range of angles of attack ((alpha _{tail} More

Study areaThe study was carried out at the University of Dodoma (UDOM) and specifically at College of Natural and Mathematical Sciences (CNMS) and College of Education (COED) (Fig. 1). These two sites were considered because they have both afforested and non-afforested areas. Furthermore, unlike other places where afforestation is uncoordinated, the selected study area has proper management and records for the afforestation program that is taking place. The study area is located at latitude of 6° 57´ and 3° 82´ and longitudinal of 36° 26´ and 35° 26´. Its elevation is estimated to be 1120 m above the sea level. The site is semi-arid area dominated by sandy loam soil classified as Oxisol. The average annual rainfall of the areas is 447 mm. Temperatures vary depending on the season, with average minimum and maximum of (18^circ{rm C}) and 32 (^circ{rm C}) respectively.Figure 1Map showing the study area within the University of Dodoma (Created using QGIS 3.28.0 Firenze version, 2022). Note: CNMS-College of natural and mathematical science, CHS-College of humanities and social science, CIVE- College of informatic and virtual education, COED-College of Education.Full size imageThe bush is leafless and dry during the dry season, but comes to life during the rainy season, when the entire countryside turns a vibrant green19,20 The remaining land is covered in woodlands, with the highest concentrations in hills (URT 2014). The vegetation consists of dry savanna shrub-thicket areas with scattered trees and grassland patches interrupted by trees and shrubs.Study design on abundance and diversity of lizardsData on lizard abundance and diversity were collected at two sites, namely the CNMS and another site located at COED. These areas were purposely selected because the afforestation program is taking place. In the selected areas, trees have been planted for the past three years, which are 2019–2021. More effort is being made to plant more trees. Also, the areas have natural vegetation characterized by thickets, shrubs, and nature trees with species as described above in the study area. This makes the areas ideal for making comparisons between the afforested and un-afforested areas. In each site, two blocks were established, in which one block consisted of an afforested area while the other block was a non-afforested area.Data collectionDocumentation of planted tree speciesThe plants observed in the study areas were recorded. In addition to that, we worked with the restoration team, which provided the list of tree species that are grown in those study areas. Secondary data was collected from the restoration team regarding the tree species and how much has been planted in the last 5 years in the study areas.Sampling of lizard for abundance and diversity determinationPitfall trapsEach block had a size of 60 m by 60 m (2600 m2). In each block, two transects were established, each with a length of 60 m and a spacing of 20 m. In each transect, 4 points were identified, whereby 10 pitfall traps of 5 L each were set at an interval of 12 m. This makes 40 pitfall traps and eight walking transects. Emptying was performed every morning for 10 consecutive days in each pitfall. Thus, a total of 800 samples were collected from pitfall traps, with 400 samples being collected at each site.Direct searchingGeneral direct searching involving time-constrained observation was also used to collect data on the lizards found in the study area. Time constrained searches were conducted as an opportunistic means of finding animals hiding under cover and flushing them as the observer approached. Searching was conducted in an area of 20 m × 20 m at each sampling point where pitfalls were set. Searching was performed by an individual who is an ecologist and is an expert in reptiles for 10 min, 3 times a day for 10 days (n = 240). To ensure consistency, the same individual was employed in searching for each sampling point.At each site, the observed lizards were identified by their numbers and habitats. Photographs of captured or observed animals were taken to aid in identification. In addition, human activities such as cultivation, roads, tree cutting, building, and distance from roads and buildings, were recorded. Furthermore, more physical structures like rocks and distances from rocks were recorded. Identification of species of lizards was performed using a guide book for east African reptiles21.Sampling and interview for the assessment of awareness of the importance of afforestationA cross-sectional survey using a semi-structured questionnaire was used to collect data from undergraduate students in four colleges, which are CNMS, COED, CHSS, and CIVE. The respondents were selected randomly from each college. These students were selected based on their familiarity with the areas that are anticipated to see what is taking place within the University of Dodoma. It was anticipated that awareness would vary by college because the programs offered differed. For example, it was predicted that students from CNMS would be more aware than others because they have programs and courses that teach conservation, restoration, and afforestation knowledge. Both genders were included in the survey. A total of 394 interviewees were recruited; 100 participants were from CHSS, 103 from CIVE, 101 from CNMS, and 90 from COED. The questionnaire consisted of both closed and open-ended questions. The questions consisted of information on the demographic structure of students and their awareness of the afforestation program. Concerning awareness, the questions focused on their understanding of afforestation, their participation, and other stakeholders involved in the program.Some questions had to be ranked from 1 to 5, with the answers classified as very high, high, moderate, low, and very low if they scored 5, 4, 3, 2, and 1, respectively. The questions were designed to elicit responses from respondents regarding their knowledge of the ongoing afforestation program. In addition, information on the program’s participants and their level of involvement was requested.Human ethical guideline statementAll methods were carried out in accordance with relevant guidelines and regulations.Ethical approval and consent to participateThe ethics committee of University of Dodoma granted ethical approval for this study, with reference number MA.84/261/02.Informed consentInformed consent was obtained from all participants included in the study. More

Darwin, C. On the Origin of Species by Means of Natural Selection (Murray, 1859).
Google Scholar 
Mallet, J. Hybridization as an invasion of the genome. Trends Ecol. Evol. 20(5), 229–237. https://doi.org/10.1016/j.tree.2005.02.010 (2005).Article 

Google Scholar 
Ottenburghs, J. Multispecies hybridization in birds. Avian Res. 10(1), 20. https://doi.org/10.1186/s40657-019-0159-4 (2019).Article 

Google Scholar 
Grant, P. R. & Grant, B. R. Hybridization of bird species. Science 256(5054), 193–197. https://doi.org/10.1126/science.256.5054.193 (1992).Article 
ADS 

Google Scholar 
Ottenburghs, J. Exploring the hybrid speciation continuum in birds. Ecol. Evol. 8(24), 13027–13034. https://doi.org/10.1002/ece3.4558 (2018).Article 

Google Scholar 
Stebbins, G. L. Variation and Evolution in Plants in Variation and Evolution in Plants (Columbia University Press, 1950).Book 

Google Scholar 
Justyn, N. M., Callaghan, C. T. & Hill, G. E. Birds rarely hybridize: A citizen science approach to estimating rates of hybridization in the wild. Evolution 74(6), 1216–1223. https://doi.org/10.1111/evo.13943 (2020).Article 

Google Scholar 
Mayr, E. Systematics and the Origin of Species, from the Viewpoint of a Zoologist (Harvard University Press, 1999).
Google Scholar 
Uy, J. A. C., Irwin, D. E. & Webster, M. S. Behavioral isolation and incipient speciation in birds. Annu. Rev. Ecol. Evol. Syst. 49(1), 1–24. https://doi.org/10.1146/annurev-ecolsys-110617-062646 (2018).Article 

Google Scholar 
Leighton, G. M., Lu, L. J., Holop, E., Dobler, J. & Ligon, R. A. Sociality and migration predict hybridization across birds. Proc. Biol. Sci. 288(1947), 20201946. https://doi.org/10.1098/rspb.2020.1946 (2021).Article 

Google Scholar 
Chunco, A. J. Hybridization in a warmer world. Ecol. Evol. 4(10), 2019–2031. https://doi.org/10.1002/ece3.1052 (2014).Article 

Google Scholar 
Grabenstein, K. C. & Taylor, S. A. Breaking barriers: Causes, consequences, and experimental utility of human-mediated hybridization. Trends Ecol. Evol. 33(3), 198–212. https://doi.org/10.1016/j.tree.2017.12.008 (2018).Article 

Google Scholar 
Quilodrán, C. S., Montoya-Burgos, J. I. & Currat, M. Harmonizing hybridization dissonance in conservation. Commun. Biol. 3(1), 391. https://doi.org/10.1038/s42003-020-1116-9 (2020).Article 

Google Scholar 
Rhymer, J. M. & Simberloff, D. Extinction by hybridization and introgression. Annu. Rev. Ecol. Syst. 27(1), 83–109. https://doi.org/10.1146/annurev.ecolsys.27.1.83 (1996).Article 

Google Scholar 
Chan, C. et al. Genetic analysis of interspecific hybridisation in the world’s only Forbes’ parakeet (Cyanoramphus forbesi) natural population. Conserv. Genet. 7(4), 493–506. https://doi.org/10.1007/s10592-005-9060-2 (2006).Article 

Google Scholar 
Huang, L. et al. Molecular evidence of introgressive hybridization between related species Jankowski’s Bunting (Emberiza jankowskii) and Meadow Bunting (Emberiza cioides) (Aves: Passeriformes). Avian Res. 13, 100035. https://doi.org/10.1016/j.avrs.2022.100035 (2022).Article 

Google Scholar 
Reudink, M. W., Mech, S. G., Mullen, S. P. & Curry, R. L. Structure and dynamics of the hybrid zone between Black-capped chickadee (Poecile atricapillus) and Carolina chickadee (P. carolinensis) in southeastern Pennsylvania. Auk 124(2), 463–478. https://doi.org/10.1093/auk/124.2.463 (2007).Article 

Google Scholar 
Negro, J. J., Torres, M. J. & Godoy, J. A. RAPD analysis for detection and eradication of hybrid partridges (Alectoris rufa × A. graeca) in Spain. Biol. Conserv. 98(1), 19–24. https://doi.org/10.1016/S0006-3207(00)00129-4 (2001).Article 

Google Scholar 
Bensch, S., Helbig, A. J., Salomon, M. & Seibold, I. Amplified fragment length polymorphism analysis identifies hybrids between two subspecies of warblers. Mol. Ecol. 11(3), 473–481. https://doi.org/10.1046/j.0962-1083.2001.01455.x (2002).Article 

Google Scholar 
Caballero-López, V., Lundberg, M., Sokolovskis, K. & Bensch, S. Transposable elements mark a repeat-rich region associated with migratory phenotypes of willow warblers (Phylloscopus trochilus). Mol. Ecol. 31(4), 1128–1141. https://doi.org/10.1111/mec.16292 (2022).Article 

Google Scholar 
Haig, S. M., Mullins, T. D., Forsman, E. D., Trail, P. W. & Wennerberg, L. Genetic identification of spotted owls, barred owls, and their hybrids: Legal implications of hybrid identity. Conserv. Biol. 18(5), 1347–1357. https://doi.org/10.1111/j.1523-1739.2004.00206.x (2004).Article 

Google Scholar 
Michalczuk, J. et al. Tests of multiple molecular markers for the identification of Great Spotted and Syrian woodpeckers and their hybrids. J. Ornithol. 155(3), 591–600. https://doi.org/10.1007/s10336-014-1040-1 (2014).Article 

Google Scholar 
Väli, Ü. et al. Microsatellites and single nucleotide polymorphisms in avian hybrid identification: A comparative case study. J. Avian Biol. 41(1), 34–49. https://doi.org/10.1111/j.1600-048X.2009.04730.x (2010).Article 

Google Scholar 
Wiley, C., Qvarnström, A., Andersson, G., Borge, T. & Saetre, G. P. Postzygotic isolation over multiple generations of hybrid descendents in a natural hybrid zone: How well do single-generation estimates reflect reproductive isolation?. Evolution 63(7), 1731–1739. https://doi.org/10.1111/j.1558-5646.2009.00674.x (2009).Article 

Google Scholar 
Adedze, Y. M. N. et al. Agarose-resolvable InDel markers based on whole genome re-sequencing in cucumber. Sci. Rep. 11(1), 3872. https://doi.org/10.1038/s41598-021-83313-x (2021).Article 
ADS 

Google Scholar 
Weinman, L. R., Solomon, J. W. & Rubenstein, D. R. A comparison of single nucleotide polymorphism and microsatellite markers for analysis of parentage and kinship in a cooperatively breeding bird. Mol. Ecol. Resour. 15(3), 502–511. https://doi.org/10.1111/1755-0998.12330 (2015).Article 

Google Scholar 
Noda, T., Daiou, K., Mihara, T. & Nagano, Y. Development of indel markers for the selection of Satsuma mandarin (Citrus unshiu Marc.) hybrids that can be used for low-cost genotyping with agarose gels. Euphytica 216(7), 115. https://doi.org/10.1007/s10681-020-02654-2 (2020).Article 

Google Scholar 
Väli, U., Brandström, M., Johansson, M. & Ellegren, H. Insertion-deletion polymorphisms (indels) as genetic markers in natural populations. BMC Genet. 9(1), 8. https://doi.org/10.1186/1471-2156-9-8 (2008).Article 

Google Scholar 
Shapiro, M. D. et al. Genomic diversity and evolution of the head crest in the rock pigeon. Science 339(6123), 1063–1067. https://doi.org/10.1126/science.1230422 (2013).Article 
ADS 

Google Scholar 
Schwenk, K., Brede, N. & Streit, B. Introduction. Extent, processes and evolutionary impact of interspecific hybridization in animals. Philos. Trans. R. Soc. Lond. B Biol. Sci. 363(1505), 2805–2811. https://doi.org/10.1098/rstb.2008.0055 (2008).Article 

Google Scholar 
Darwin, C. The Variation of Animals and Plants under Domestication 2 (Murray, USA, 1868).
Google Scholar 
Long, J. L. Introduced Birds of the World: The Worlwide History, Distribution and Influence of Birds Introduced to New Environments (David and Abbott, 1981).
Google Scholar 
Shapiro, M. D. & Domyan, E. T. Domestic pigeons. Curr. Biol. 23(8), R302–R303. https://doi.org/10.1016/j.cub.2013.01.063 (2013).Article 

Google Scholar 
Gholamhosseini, A., Vardakis, M., Aliabadian, M., Nijman, V. & Vonk, R. Hybridization between sister taxa versus non-sister taxa: A case study in birds. Bird Study 60(2), 195–201. https://doi.org/10.1080/00063657.2013.770815 (2013).Article 

Google Scholar 
Johnston, R. F., Siegel-Causey, D. & Johnson, S. G. European populations of the rock dove Columba livia and genotypic extinction. Am. Midl. Nat. 120(1), 1–10. https://doi.org/10.2307/2425881 (1988).Article 

Google Scholar 
Kim, J. Y. et al. Population genetic structure and conservation management of hill pigeons (Columba rupestris) recently endangered in South Korea. Genes Genom. https://doi.org/10.1007/s13258-021-01212-x (2022).Article 

Google Scholar 
Baptista, L. F., Trail, P. W., Horblit, H. M. & Kirwan, G. M. In Hill Pigeon (Columba rupestris). version 1.0 in Birds of the World (eds del Hoyo, J. et al.) (Cornell Laboratory of Ornithology, 2020). https://doi.org/10.2173/bow.hilpig1.01.Chapter 

Google Scholar 
McCarthy, E. M. Handbook of Avian Hybrids of the World (Oxford University Press, 2006).
Google Scholar 
Lijtmaer, D. A., Mahler, B. & Tubaro, P. L. Hybridization and postzygotic isolation patterns in pigeons and doves. Evolution 57(6), 1411–1418. https://doi.org/10.1111/j.0014-3820.2003.tb00348.x (2003).Article 

Google Scholar 
Johnston, R. F. & Janiga, M. Feral Pigeons (Oxford University Press, 1995). https://doi.org/10.2307/2425881.Book 

Google Scholar 
Stringham, S. A. et al. Divergence, convergence, and the ancestry of feral populations in the domestic rock pigeon. Curr. Biol. 22(4), 302–308. https://doi.org/10.1016/j.cub.2011.12.045 (2012).Article 

Google Scholar 
Hernández, F., Brown, J. I., Kaminski, M., Harvey, M. G. & Lavretsky, P. Genomic evidence for rare hybridization and large demographic changes in the evolutionary histories of four North American dove species. Animals (Basel) 11(9), 2677. https://doi.org/10.3390/ani11092677 (2021).Article 

Google Scholar 
Séré, M. et al. Null allele, allelic dropouts or rare sex detection in clonal organisms: Simulations and application to real data sets of pathogenic microbes. Parasit. Vectors 7(1), 331. https://doi.org/10.1186/1756-3305-7-331 (2014).Article 

Google Scholar 
Chan, W. Y., Hoffmann, A. A. & Oppen, M. J. H. Hybridization as a conservation management tool. Conserv. Lett. 12, 5. https://doi.org/10.1111/conl.12652 (2019).Article 

Google Scholar 
Randi, E. Detecting hybridization between wild species and their domesticated relatives. Mol. Ecol. 17(1), 285–293. https://doi.org/10.1111/j.1365-294X.2007.03417.x (2008).Article 

Google Scholar 
Wayne, R. K. & Shaffer, H. B. Hybridization and endangered species protection in the molecular era. Mol. Ecol. 25(11), 2680–2689. https://doi.org/10.1111/mec.13642 (2016).Article 

Google Scholar 
Pfennig, K. S., Kelly, A. L. & Pierce, A. A. Hybridization as a facilitator of species range expansion. Proc. Biol. Sci. 283(1839), 20161329. https://doi.org/10.1098/rspb.2016.1329 (2016).Article 

Google Scholar 
Oliveira, R. et al. Toward a genome-wide approach for detecting hybrids: Informative SNPs to detect introgression between domestic cats and European wildcats (Felis silvestris). Heredity 115(3), 195–205. https://doi.org/10.1038/hdy.2015.25 (2015).Article 

Google Scholar 
Austin, O. L. Jr. The Birds of Korea. Bulletin of the Museum of Comparative Zoology (Harvard College, 1948).
Google Scholar 
Kim, J. A. et al. Whole-genome sequencing revealed different demographic histories among the Korean endemic hill pigeon (Columba rupestris), rock pigeon (Columba livia var. domestica) and oriental turtle dove (Streptopelia orientalis). Genes Genom. 44(10), 1231–1242. https://doi.org/10.1007/s13258-022-01288-z (2022).Article 

Google Scholar 
NIBR. Genetic Diversity of Animal Resources IV_1 (National Institute of Biological Resources, 2018).
Google Scholar  More

De Jesús-Crespo, R. & Ramírez, A. Effects of urbanisation on stream physicochemistry and macroinvertebrate assemblages in a tropical urban watershed in Puerto Rico. J. N. Am. Benthol. Soc. 30, 739. https://doi.org/10.1899/10-081.1 (2011).Article 

Google Scholar 
Start, D., Barbour, M. A. & Bonner, C. Urbanisation reshapes a food web. J. Anim. Ecol. 89(3), 808–816. https://doi.org/10.1111/1365-2656.131366 (2020).Article 

Google Scholar 
Liu, Y. et al. Analysis of the influence paths of land use and landscape pattern on organic matter decomposition in river ecosystems: Focusing on microbial groups. Sci. Total Environ. 95(14), 106408. https://doi.org/10.1016/j.scitotenv.2022.152999 (2022).Article 
CAS 

Google Scholar 
Vörösmarty, C. J., Mcintyre, P. B., Gessner, M. O., Dudgeon, D. & Prusevich, A. Global threats to human water security and river biodiversity. Nature 467(7315), 555–561. https://doi.org/10.1038/nature09440 (2010).Article 
ADS 
CAS 

Google Scholar 
Weideman, E. A., Perold, V., Arnold, G. & Ryan, P. G. Quantifying changes in litter loads in urban stormwater runoff from Cape Town, South Africa, over the last two decades. Sci. Total Environ. 724, 138310. https://doi.org/10.1016/j.scitotenv.2020.138310 (2020).Article 
ADS 
CAS 

Google Scholar 
Charters, F. J., Cochrane, T. A. & O’Sullivan, A. D. The influence of urban surface type and characteristics on runoff water quality. Sci. Total Environ. 755, 142470. https://doi.org/10.1016/j.scitotenv.2020.142470 (2021).Article 
ADS 
CAS 

Google Scholar 
Espinoza-Toledo, A., Mendoza-Carranza, M., Castillo, M. M., Barba-Macías, E. & Capps, K. A. Taxonomic and functional responses of macroinvertebrates to riparian forest conversion in tropical streams. Sci. Total Environ. 757, 143972. https://doi.org/10.1016/j.scitotenv.2020.143972 (2021).Article 
ADS 
CAS 

Google Scholar 
Akamagwuna, C. F., Ntloko, P., Edegbene, A. O. & Odume, O. N. Are Ephemeroptera, Plecoptera and Trichoptera traits reliable indicators of semi- urban pollution in the Tsitsa River, Eastern Cape Province of South Africa ? Environ. Monit. Assess. 193, 1–15. https://doi.org/10.1007/s10661-021-09093-z (2021).Article 
CAS 

Google Scholar 
Edegbene, A. O., Odume, O. N., Arimoro, F. O. & Keke, U. N. Identifying and classifying macroinvertebrate indicator signature traits and ecological preferences along urban pollution gradient in the Niger Delta. Environ. Pollut. 281, 117076. https://doi.org/10.1016/j.envpol.2021.117076 (2021).Article 
CAS 

Google Scholar 
Odume, O. N. Searching for urban pollution signature and sensitive macroinvertebrate traits and ecological preferences in a river in the Eastern Cape of South Africa. Ecol. Indic. 108, 105759. https://doi.org/10.1016/j.ecolind.2019.105759 (2020).Article 
CAS 

Google Scholar 
World Bank Group. e-Conomy Africa 2020 Africa’s $180 billion Internet economy future. In E-Conomy Africa 2020 (2020).Petersen, C. R., Jovanovic, N. Z., Grenfell, M. C., Oberholster, P. J. & Cheng, P. Responses of aquatic communities to physical and chemical parameters in agriculturally impacted coastal river systems. Hydrobiologia 813(1), 157–175. https://doi.org/10.1007/s10750-018-3518-y (2018).Article 
CAS 

Google Scholar 
Zhao, Q., Guo, F., Zhang, Y., Yang, Z. & Ma, S. Effects of secondary salinisation on macroinvertebrate functional traits in surface mining-contaminated streams, and recovery potential. Sci. Total Environ. 640–641, 1088–1097. https://doi.org/10.1016/j.scitotenv.2018.05.347 (2018).Article 
ADS 
CAS 

Google Scholar 
Wang, Y. et al. Trophic structure in response to land use in subtropical streams. Ecol. Indic. 127, 107746. https://doi.org/10.1016/j.ecolind.2021.107746 (2021).Article 
CAS 

Google Scholar 
David, A. Landscapes and riverscapes: The influence of land use on stream ecosystems. Annu. Rev. Ecol. Evol. Syst. 35, 257–284. https://doi.org/10.1146/annurev.ecolsys.35.120202.110122 (2004).Article 

Google Scholar 
Monk, W. A., Wood, P. J., Hannah, D. M. & Wilson, D. A. Macroinvertebrate community response to inter-annual and regional river flow regime dynamics. River Res. Appl. 24(70), 988–1001 (2008).Article 

Google Scholar 
Fitzgerald, D. B., Winemiller, K. O., SabajPerez, M. H. & Sousa, L. M. Seasonal changes in the assembly mechanisms structuring tropical fish communities. Ecology 98, 21–31. https://doi.org/10.1002/ecy.1616 (2017).Article 

Google Scholar 
Rosenberg, D. M. & Resh, V. H. Freshwater Biomonitoring and Benthic Macroinvertebrates (Chapman and Hall, 1993).
Google Scholar 
Jiang, X., Xiong, J., Xie, Z. & Chen, Y. Longitudinal patterns of macroinvertebrate functional feeding groups in a Chinese river system: A test for river continuum concept (RCC). Quatern. Int. 244(2), 289–295. https://doi.org/10.1016/j.quaint.2010.08.015 (2011).Article 

Google Scholar 
Sun, L. Q. et al. Food web structure and ecosystem attributes of integrated multi-trophic aquaculture waters in Sanggou Bay. Aquac. Rep. 16, 100279. https://doi.org/10.1016/j.aqrep.2020.100279 (2020).Article 

Google Scholar 
Covich, A. P., Palmer, M. A. & Crowl, T. A. The role of benthic invertebrate species in freshwater ecosystems. Bioscience 49, 119–127 (1999).Article 

Google Scholar 
Hladyz, S., Kajsa, Å., Paul, S. G. & Guy, W. Impacts of an aggressive riparian invader on community structure and ecosystem functioning in stream food webs. J. Appl. Ecol. 48(2), 443–452. https://doi.org/10.1111/j.1365-2664.2010.01924.x (2011).Article 

Google Scholar 
Crowl, T. A. & Covich, A. P. Predator-induced life history shifts in a freshwater snail. Science 247, 949–951 (1990).Article 
ADS 
CAS 

Google Scholar 
Fierro, P. et al. Effects of local land-use on riparian vegetation, water quality, and the functional organisation of macroinvertebrate assemblages. Sci. Total Environ. 609, 724–734. https://doi.org/10.1016/j.scitotenv.2017.07.197 (2017).Article 
ADS 
CAS 

Google Scholar 
Lenat, D. R. & Barbour, M. T. Using benthic macroinvertebrate community structure for rapid, cost-effective, water quality monitoring: Rapid bioassessment. Biol. Monit. Aquat. Syst. 1, 187–215 (1994).
Google Scholar 
Edegbene, A., Arimoro, F. O. & Odume, O. N. How does urban pollution influence macroinvertebrate traits in forested riverine systems? Water 1, 2–17. https://doi.org/10.3390/w12113111 (2020).Article 

Google Scholar 
Walsh, C. J. et al. The urban stream syndrome: Current knowledge and the search for a cure. J. N. Am. Renthol. Soc. 24(3), 706–723. https://doi.org/10.1899/04-028.1 (2005).Article 

Google Scholar 
Bêche, L. A., Eric, P. M. & Vincent, H. R. Long-term seasonal variation in the biological traits of benthic-macroinvertebrates in two mediterranean-climate streams in California, USA. Freshw. Biol. 51(1), 56–75. https://doi.org/10.1111/j.1365-2427.2005.01473.x (2006).Article 

Google Scholar 
Sitati, A., Raburu, P. O., Yegon, M. J. & Masese, F. O. Land-use influence on the functional organization of Afrotropical macroinvertebrate assemblages. Limnologica 88, 125875. https://doi.org/10.1016/j.limno.2021.125875 (2021).Article 
CAS 

Google Scholar 
Álvarez-Cabria, M. J. & José, A. J. Spatial and seasonal variability of macroinvertebrate metrics: Do macroinvertebrate communities track river health? Ecol. Indic. 10(2), 370–379. https://doi.org/10.1016/j.ecolind.2009.06.018 (2010).Article 
CAS 

Google Scholar 
Masese, F. O. et al. Litter processing and shredder distribution as indicators of riparian and catchment influences on ecological health of tropical streams. Ecol. Indic. 46, 23–37. https://doi.org/10.1016/j.ecolind.2014.05.032 (2014).Article 

Google Scholar 
Merritt, R. & Cummins, K. An Introduction to the Aquatic Insects of North America 1996–862 (Kendall Hunt Publishing Co, 1995).
Google Scholar 
Merritt, R. W., Fenoglio, S. & Cummins, K. W. Promoting a functional macroinvertebrate approach in the biomonitoring of Italian lotic systems. J. Limnol. https://doi.org/10.4081/jlimnol.2016.1502 (2016).Article 

Google Scholar 
Edegbene, A.O., Akamagwuna, F.C., Arimoro, F.O., Akumabor, E.C. & Kaine, E.A. Effects of urban-agricultural land-use on Afrotropical macroinvertebrate functional feeding groups in selected rivers in the Niger Delta Region, Nigeria. Hydrobiologia 849, 4857–4869. https://doi.org/10.1007/s10750-022-05034-0 (2022).Article 
CAS 

Google Scholar 
Masese, F. O. et al. Macroinvertebrate functional feeding groups in Kenyan highland streams: Evidence for a diverse shredder guild. Freshw. Sci. 33(2), 435–450. https://doi.org/10.1086/675681 (2014).Article 

Google Scholar 
Moyo, S. & Richoux, N. B. Macroinvertebrate functional organisation along the longitudinal Gradient of an austral temperate river. Afr. Zool. 52(3), 125–136. https://doi.org/10.1080/15627020.2017.1354721 (2017).Article 

Google Scholar 
Vannote, R. L. et al. The river continuum concept. Can. J. Fish. Aquat. Sci. 37, 130–137 (1980).Article 

Google Scholar 
Ding, J. et al. Impacts of land use on surface water quality in a subtropical river basin: A case study of the Dongjiang River Basin, Southeastern China. Water 7(8), 4427–4445. https://doi.org/10.3390/w7084427 (2015).Article 
ADS 
CAS 

Google Scholar 
Miserendino, M. L. & Masi, C. I. The effects of land use on environmental features and functional organisation of macroinvertebrate communities in Patagonian low order streams. Ecol. Indic. 10(2), 311–319. https://doi.org/10.1016/j.ecolind.2009.06.008 (2010).Article 
CAS 

Google Scholar 
Solis, M., Arias, M., Fanelli, S., Bonetto, C. & Mugni, H. Agrochemicals’ effects on functional feeding groups of macroinvertebrates in Pampas streams. Ecol. Indic. 101, 373–379. https://doi.org/10.1016/j.ecolind.2019.01.036 (2019).Article 
CAS 

Google Scholar 
Mangadze, T., Wasserman, R., Froneman, W. & Dalu, T. Macroinvertebrate functional feeding group alterations in response to habitat degradation of headwater Austral streams. Sci. Total Environ. 695, 133910. https://doi.org/10.1016/j.scitotenv.2019.133910 (2019).Article 
ADS 
CAS 

Google Scholar 
Akamagwuna, F. C. & Odume, O. N. Ephemeroptera, Plecoptera and Trichoptera (EPT) functional feeding group responses to fine grain sediment stress in a river in the Eastern Cape, South Africa. Environ. Monit. Assess. 2, 1–11 (2020).
Google Scholar 
Iwegbue, C. M. A. et al. Polycyclic aromatic hydrocarbons (PAHs) in surficial sediments from selected rivers in the western Niger Delta of Nigeria: Spatial distribution, sources, and ecological and human health risks. Mar. Pollut. Bull. 167, 112351. https://doi.org/10.1016/j.marpolbul.2021.112351 (2021).Article 
CAS 

Google Scholar 
Arimoro, F. O., Abubakar, M. D., Obi-iyeke, G. E. & Keke, U. N. Environmental and sustainability indicators achieving sustainable river water quality for rural dwellers by prioritising the conservation of macroinvertebrates biodiversity in two Afrotropical streams. Environ. Sustain. Indic. 10, 100103. https://doi.org/10.1016/j.indic.2021.100103 (2021).Article 

Google Scholar 
Zabbey, N., Erondu, E. S. & Hart, A. I. Nigeria and the prospect of shrimp farming: Critical issues. Livestock Res. Rural Dev. 22, 11 (2010).
Google Scholar 
Edegbene, A. O., Arimoro, F. O. & Odume, O. N. Developing and applying a macroinvertebrate-based multimetric index for urban rivers in the Niger Delta, Nigeria. Ecol. Evol. 9(22), 12869–12885. https://doi.org/10.1002/ece3.5769 (2019).Article 

Google Scholar 
Iwegbue, C. M. A. et al. Distribution, sources and ecological risks of metals in surficial sediments of the Forcados River and its Estuary, Niger Delta, Nigeria. Environ. Earth Sci. 77(6), 1–18. https://doi.org/10.1007/s12665-018-7344-3 (2018).Article 
CAS 

Google Scholar 
Stoddard, J. L., Larsen, D. P., Hawkins, C. P., Johnson, R. K. & Norris, R. H. Setting expectations for the ecological condition of running waters: The concept of reference condition. Ecol. Appl. 16, 1267–1276 (2006).Article 

Google Scholar 
Whittier, T. R. et al. A structured approach for developing indices of biotic integrity: Three examples from streams and rivers in the western USA. Trans. Am. Fish. Soc. 136, 718–735 (2007).Article 

Google Scholar 
APHA. Standard Methods for the Examination of Water and Wastewater (American Public Health Association, 1995).
Google Scholar 
Dickens, C. W. & Graham, P. M. The South African scoring system (SASS) Version 5 rapid bioassessment method for Rivers. Afr. J. Aquat. Sci. 27(1), 1–10. https://doi.org/10.2989/16085914.2002.9626569 (2002).Article 

Google Scholar 
Day, J. A., & de Moor, I. J. Guides to the freshwater invertebrates of southern Africa. In Volume 6: Arachnida and Mollusca (Araneae, Water Mites and Mollusca). Water Research Commision, 6. WRC Report No. TT 182/02 (2002).De Moor, I. J., Day, J. A. & De Moor, F. C. Guides to the freshwater invertebrates of southern Africa. In Volume 8: Insecta II: Hemiptera, Megaloptera, Neuroptera, Trichoptera and Lepidoptera. Water Research Commision, 8 (2003).Merrit, R. W. et al. An Introduction to the Aquatic Insects of North America 4th edn. (Kendall Hunt Publishing Company, 2008).
Google Scholar 
Cummins, K. W., Merritt, R. W. & Andrade, P. C. N. The use of invertebrate functional groups to characterize ecosystem attributes in selected streams and rivers in South Brazil. Stud. Neotrop. Fauna Environ. 40(1), 69–89. https://doi.org/10.1080/01650520400025720 (2005).Article 

Google Scholar 
Palmer, C. G. Benthic Assemblage Structure, and the Feeding Biology of Sixteen Macro Invertebrate Taxa from the Buffalo River, Eastern Cape, South Africa. Rhodes University, Ph.D. thesis (1991).Palmer, C. G., Maart, B., Palmer, A. R. & O’keeffe, J. H. An assessment of macroinvertebrate functional feeding groups as water quality indicators in the Buffalo River, eastern Cape Province, South Africa. Hydrobiologia 318, 153–164 (1996).Article 

Google Scholar 
Palmer, C. G. & O’Keeffe, J. H. O. Feeding patterns of four macroinvertebrate taxa in the headwaters of the Buffalo River, Eastern Cape. Hydrobiologia 228, 157–173 (1992).Article 

Google Scholar 
Gayraud, S. & Michel, P. Influence of bed-sediment features on the interstitial habitat available for macroinvertebrates in 15 French streams. Int. Rev. Hydrobiol. 88(1), 77–93. https://doi.org/10.1002/iroh.200390007 (2003).Article 

Google Scholar 
Beketov, M. A. et al. SPEAR indicates pesticide effects in streams—Comparative use of species- and family-level biomonitoring data. Environ. Pollut. 157, 1841–1848. https://doi.org/10.1016/j.envpol.2009.01.021 (2009).Article 
CAS 

Google Scholar 
Anderson, M., Gorley, R. N. & Clarke, K. R. PERMANOVA + for PRIMER user manual. Primer-E Ltd 1(1), 218 (2008).
Google Scholar 
Dolédec, S., Chessel, D., ter Braak, C. J. F. & Champely, S. Matching species traits to environmental variables: A new three-table ordination method. Environ. Ecol. Stat. 3(2), 143–166. https://doi.org/10.1007/BF02427859 (1996).Article 

Google Scholar 
Juvigny-Khenafou, N. P. D. et al. Impacts of multiple anthropogenic stressors on stream macroinvertebrate community composition and functional diversity. Ecol. Evol. 11(1), 133–152. https://doi.org/10.1002/ece3.6979 (2021).Article 

Google Scholar 
Dray, P. S. et al. Combining the fourth-corner and the RLQ methods for assessing trait responses to environmental variation. Ecology 95(1), 14–21 (2014).Article 

Google Scholar 
R Core Team, E. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2020).Lubanga, H. L., Manyala, J. O., Sitati, A., Yegon, M. J. & Masese, F. O. Spatial variability in water quality and macroinvertebrate assemblages across a disturbance gradient in the Mara River. Ecohydrol. Hydrobiol. https://doi.org/10.1016/j.ecohyd.2021.03.001 (2021).Article 

Google Scholar 
Arimoro, F. O. & Ikomi, R. B. Ecological integrity of upper Warri River, Niger Delta using aquatic insects as bioindicators. Ecol. Indic. 9(3), 455–461. https://doi.org/10.1016/j.ecolind.2008.06.006 (2009).Article 
CAS 

Google Scholar 
Arimoro, F. O., Odume, O. N., Uhunoma, S. I. & Edegbene, A. O. Anthropogenic impact on water chemistry and benthic macroinvertebrate associated changes in a southern Nigeria stream. Environ. Monit. Assess. 187(2), 1–14 (2015).Article 
CAS 

Google Scholar 
Keke, U. N. et al. Macroinvertebrate communities and physicochemical characteristics along an anthropogenic stress gradient in a southern Nigeria stream: Implications for ecological restoration. Environ. Sustain. Indic. https://doi.org/10.1016/j.indic.2021.100157 (2021).Article 

Google Scholar 
Edegbene, A. O. et al. A macroinvertebrate-based multimetric index for assessing ecological condition of forested stream sites draining Nigerian urbanizing landscapes. Sustainability 14, 11289. https://doi.org/10.3390/su141811289 (2022).Article 

Google Scholar 
Matemilola, S., Adedeji, O. H. & Enoguanbhor, E. C. Land use/land cover change in petroleum-producing regions of Nigeria. In The Political Ecology of Oil and Gas Activities in the Nigerian Aquatic Ecosystem (eds Matemilola, S. et al.) (Elsevier Inc., 2018).
Google Scholar 
Ukhurebor, K. E. et al. Environmental implications of petroleum spillages in the Niger Delta region of Nigeria: A review. J. Environ. Manag. 293, 112872. https://doi.org/10.1016/j.jenvman.2021.112872 (2021).Article 
CAS 

Google Scholar 
Masese, F. O. & Raburu, P. O. Improving the performance of the EPT Index to accommodate multiple stressors in Afrotropical streams. Afr. J. Aquat. Sci. 42(3), 219–233. https://doi.org/10.2989/16085914.2017.1392282 (2017).Article 

Google Scholar 
Akamagwuna, F. C. Application of Macroinvertebrate-Based Biomonitoring and Stable Isotopes for Assessing the Effects of Agricultural Land-Use on River Ecosystem Structure and Function in the Kat River, Eastern Cape, South Africa, 4 (Rhodes University, 2021).
Google Scholar 
Yang, F. et al. Application of stable isotopes to the bioaccumulation and trophic transfer of arsenic in aquatic organisms around a closed realgar mine. Sci. Total Environ. 726, 138550. https://doi.org/10.1016/j.scitotenv.2020.138550 (2020).Article 
ADS 
CAS 

Google Scholar 
Minaya, V., Mcclain, M. E., Moog, O., Omengo, F. & Singer, G. A. Scale-dependent effects of rural activities on benthic macroinvertebrates and physico-chemical characteristics in headwater streams of the Mara River, Kenya. Ecol. Indic. 32, 116–122. https://doi.org/10.1016/j.ecolind.2013.03.011 (2013).Article 
CAS 

Google Scholar 
Nelson Mwaijengo, G., Msigwa, A., Njau, K. N., Brendonck, L. & Vanschoenwinkel, B. Where does land use matter most? Contrasting land use effects on river quality at different spatial scales. Sci. Total Environ. 715, 134825. https://doi.org/10.1016/j.scitotenv.2019.134825 (2020).Article 
ADS 
CAS 

Google Scholar 
Moyo, S. & Richoux, N. B. The relative importance of autochthony along the longitudinal gradient of a small South African river influenced by agricultural activities. Food Webs 15, 1–12. https://doi.org/10.1016/j.fooweb.2018.e00082 (2018).Article 

Google Scholar 
Jones, I., Growns, I., Arnold, A., McCall, S. & Bowes, M. The effects of increased flow and fine sediment on hyporheic invertebrates and nutrients in stream mesocosms. Freshw. Biol. 60(4), 813–826. https://doi.org/10.1111/fwb.1253664 (2015).Article 
CAS 

Google Scholar 
Krynak, E. M. & Yates, A. G. Benthic invertebrates taxonomic and trait associations with land use intensively managed watershed: Implications for indicator identification. Ecol. Ind. 93, 1050–1059 (2018).Article 

Google Scholar 
Kuzmanovic, M. et al. Environmental stressors as driver of the trait composition of benthic macroinvertebrates assemblages in polluted Iberian rivers. Environ. Res. 156, 485–493 (2017).Article 
CAS 

Google Scholar 
Dalu, T. et al. Benthic diatom-based indices and isotopic biomonitoring of nitrogen pollution in a warm temperate Austral river system. Sci. Total Environ. 748, 142452. https://doi.org/10.1016/j.scitotenv.2020.142452 (2020).Article 
ADS 
CAS 

Google Scholar 
Tomanova, S., Goitia, E. & Helesˇic, J. H. Trophic levels and functional feeding groups of macroinvertebrates in neotropical streams. Hydrobiologia 556(3), 251–264. https://doi.org/10.1007/s10750-005-1255-5 (2006).Article 

Google Scholar 
da Conceição, A. A., Albertoni, E. F., Milesi, S. V. & Hepp, L. U. Influence of anthropic impacts on the functional structure of aquatic invertebrates in subtropical wetlands. Wetlands. https://doi.org/10.1007/s13157-020-01317-1 (2020).Article 

Google Scholar 
De, R. B. Effects of forest conversion on the assemblages’ structure of aquatic insects in subtropical regions. Rev. Bras. Entomol. 59(1), 43–49. https://doi.org/10.1016/j.rbe.2015.02.005 (2015).Article 
ADS 

Google Scholar 
Cheshire, K., Boyero, L. & Pearson, R. G. Food webs in tropical Austrian streams: Shredders are not scarce. Freshw. Biol. https://doi.org/10.1111/j.1365-2427.2005.01355-x (2005).Article 

Google Scholar 
Camacho, R., Boyero, L., Cornejo, A., Ibáñez, A. & Pearson, R. G. Local variation in shredder distribution can explain their oversight in tropical streams. Biotropica 41, 625–632 (2009).Article 

Google Scholar 
Boyero, L., Ramírez, A., Dudgeon, D. & Pearson, R. G. Are tropical streams really different? J. N. Am. Benthol. Soc. 28, 397–403 (2009).Article 

Google Scholar 
Ferreira, V. & Chauvet, E. Synergistic effects of water temperature and dissolved nutrients on litter decomposition and associated fungi. Glob. Change Biol. 17(1), 551–564. https://doi.org/10.1111/j.1365-2486.2010.02185.x (2011).Article 
ADS 

Google Scholar 
Whiles, M. R. & Wallace, J. B. Leaf litter decomposition and macroinvertebrate communities in headwater streams draining pine and hardwood catchments. Hydrobiologia 353, 107–119 (1997).Article 

Google Scholar 
Foucreau, N., Piscart, C., Puijalon, S. & Hervant, F. Effects of rising temperature on a functional process: Consumption and digestion of leaf litter by a freshwater shredder. Fundam. Appl. Limnol. 187(4), 295–306. https://doi.org/10.1127/fal/2016/0841 (2016).Article 

Google Scholar 
Poff, N. L. et al. Functional trait niches of North American lotic insects: Traits-based ecological applications in light of phylogenetic relationships. J. N. Am. Benthol. Soc. 25(4), 730–755 (2006).Article 

Google Scholar 
Menezes, S., Baird, D. J. & Soares, A. M. V. M. Beyond taxonomy: A review of macroinvertebrate trait-based community descriptors as tools for freshwater biomonitoring. J. Appl. Ecol. 47(4), 711–719. https://doi.org/10.1111/j.1365-2664.2010.01819.x (2010).Article 

Google Scholar 
Verberk, W. C. E. P., van Noordwijk, C. G. E. & Hildrew, A. G. Delivering on a promise: integrating species traits to transform descriptive community ecology into a predictive science. Freshw. Sci. 32(2), 531–547. https://doi.org/10.1899/12-092.1 (2013).Article 

Google Scholar 
Usseglio-Polatera, P., Bournaud, M., Richoux, P. & Tachet, H. Biomonitoring through biological traits of benthic macroinvertebrates: How to use species trait databases? Hydrobiologia 423, 153–162 (2000).Article 

Google Scholar 
Kneitel, J. & Chase, J. Trade-offs in community ecology: Linking spatial scales and species coexistence. Ecol. Lett. 7, 69–80 (2004).Article 

Google Scholar 
Pilière, A. F. H. et al. On the importance of trait interrelationships for understanding environmental responses of stream macroinvertebrates. Freshw. Biol. 61, 181–194. https://doi.org/10.1111/fwb.12690 (2016).Article 

Google Scholar  More

Research site descriptionA 3-year stationary field experiment was conducted at the Yuanziqu experimental station of the Bayannur Academy of Agricultural and Animal Husbandry Sciences (40° 90′ N, 107° 17′ E), Linhe, Inner Mongolia, China, from 2019 to 2021. The site has a continental monsoon climate typical of the northern mid-temperate zone, with a mean annual temperature from 3.7  to  7.6 °C, and the potential evaporation is 2200–2400 mm15. The total precipitation during the wheat growth period (March–July) was 66 mm, 110 mm and 47 mm in 2019, 2020, and 2021, respectively. Daily air temperature and precipitation during the field trial period are presented in Fig. 1.Figure 1Daily maximum temperature, daily minimum temperature and precipitation during the growth period (March–July) of spring wheat from 2019 to 2021 in the field experiment at Linhe, Inner Mongolia, China.Full size imageThe soil at the experimental site is a silt loam. The major physical and chemical properties of the 0–20 cm soil layer at the experimental site in 2019 were as follows: a bulk density of 1.48 g cm−3, pH 8.3, organic matter content of 15.49 g kg−1, total N concentration of 1.20 g kg−1, nitrate (NO3− N) concentration of 3.98 mg N kg−1, Olsen-P concentration of 32.3 mg P kg−1 and available K concentration of 180.0 mg K kg−1.Experimental design and field managementA local popular spring wheat (Triticum aestivum L.) cultivar, Yongliang No. 4, was used in the trials. The sowing dates were 20 March, 16 March and 13 March in 2019, 2020 and 2021, respectively; the harvest dates were 15 July, 15 July and 5 July in 2019, 2020 and 2021, respectively. A total of 375 kg ha−1 wheat seeds were sown at a depth of 5 cm.Five fertilization treatments were set in the three consecutive experimental years, including the control (CK), farmer practice (FP) and three balanced fertilization treatments (BF1, BF2 and BF3), as presented in Table 1. The P and K fertilizers as single superphosphate and K2SO4 were basal dressed, respectively. Both basal- and top-dressing of N fertilizer as urea were applied as shown in the research program. The basal applications occurred during sowing, and the remaining N fertilizer was top-dressed at the tillering stage (Table 1). The experimental plot was 10 m × 7 m with 13 cm row spacing and a buffer zone of 1 m between plots. The plots were laid out in a completely randomized block design with three replications.Table 1 Fertilization regimes of the different treatments in the 2019–2021 field trial.Full size tableEvery plot was ridged around its border to ensure the uniformity of irrigation. Flood irrigation from Yellow River water was performed according to the local policy and farmer practices. Irrigation water (30 mm) was applied at the tillering, jointing, heading and grain filling stages of spring wheat in 2019–2021. Disease, weed, and pest control, as well as other management, were performed according to local standard methods.Sampling and sample analysisThree 50 cm-long rows of spring wheat plants were selected randomly and pulled from each plot, from which 10 large, middle and small seedlings were picked out during each sampling effort. Then, the roots were cut off from the junction between the root and the stem, two plant parts (leaves and stems) before the heading stage, three plant parts (leaves, stems and spikes) at the heading stage, and four plant parts (leaves, stems, glumes and grains) at the grain filling stages and maturity were separated and pooled. The samples were dried for 30 min at 105 °C and then at 80 °C in an oven (DHG-9070A, China) until they reached a constant weight; the dry weight was then measured.The N concentrations in leaves, stems, spikes and grains of spring wheat were measured with three replications depending on the crop stage, following the Kjeldahl procedure using an element analyzer (Vario El cube, Elementar, Germany).Three soil cores containing 0–100 cm of soil were taken from each plot using an auger at the harvest of spring wheat each year. The soil samples of each 20 cm layer were collected separately and sealed immediately in a marked plastic bag. The extracts were immediately measured for nitrate-N concentration as described by Dai et al. (2015) with a continuous flow analyzer (SKALAR SAN++, Netherlands)16. The soil nitrate-N concentration was expressed on the basis of the oven-dried soil.Grain yield was evaluated at maturity by selecting two 2 m2 (avoiding border rows) randomly and harvested. A fresh weight of ∼ 1 kg of grain from each plot was weighed in the field, and the water content from each plot was oven dried for the calculation. The actual yield was adjusted by a grain water content of 13%17. Grains per spike, 1000-grain weight and spike number were determined at three 50-cm sites sampled randomly from each plot.Calculation methodsTo clarify the effect of nitrate residue in the soil under balanced fertilization, the amount of soil nitrate-N (AN, kg N ha−1) in each layer was expressed as:$${text{AN}} = left( {{text{Ti}};*;{text{Di}};*;{text{Ci}}} right)/10$$
where Ti is the soil layer thickness (cm), Di is the soil bulk density (g cm−3), Ci is the soil nitrate concentration (mg N kg−1) of the corresponding layer, and 10 is the conversion coefficient16. The AN of 0–20, 20–40, 40–60, 60–80 and 80–100 cm soil layers were recorded and measured, respectively.Nitrogen accumulation in the vegetative organs and their distribution into the grain were investigated. Based on the dry weight and corresponding measured N concentration in the different organs, apparent N translocation (TA, kg ha−1) and apparent N translocation efficiency (TR, %) were calculated as proposed by Cox et al.18 as follows:$$begin{aligned} {text{TA}} & {text{ = H}}_{{text{N}}} – {text{M}}_{{text{N}}} \ {text{TR}} & = {text{TA/H}}_{{text{N}}} *100 \ end{aligned}$$
where HN is the N assimilation in leaves or stems prior to anthesis (kg ha−1), MN is the N assimilation in leaves or stems at maturity (kg ha−1).Two parameters of nitrogen use efficiency of spring wheat, nitrogen fertilizer partial productivity (PFPN, kg/kg) and agronomic nitrogen efficiency (NAEN, kg kg−1) were determined using the following formulas:$$begin{aligned} {text{PFP}}_{{text{N}}} & = {text{ Y}}_{N; , fertilizer} /{text{N}}_{rate} \ {text{NAE}}_{{text{N}}} & = , left( {{text{Y}}_{{N;fertilizer , {-}}} {text{Y}}_{blank} } right)/{text{N}}_{rate} \ end{aligned}$$
where YN fertilizer is the grain yield of the plot with dressed N fertilizer (kg ha−1), Yblank is the grain yield of the plot without dressed N fertilizer (kg ha−1), and Nrate is the N rate of the dressed fertilizer plot (kg ha−1). Three measurements for each treatment was recorded and calculated.Two key indicators were chosen to evaluate the risk of N losses as described by Li et al. (2020)7, including N surplus (kg N per hectare per year, Nsur) and N input (kg N per hectare per year, Ninput). The N surplus was used to evaluate the risk of N losses and the N input to guide farmers’ fertilization practices directly. The detailed calculation is as follows:$$begin{aligned} {text{N}}_{{{text{sur}}}} & = {text{ N}}_{{{text{fer}}}} + {text{ N}}_{{{text{dep}}}} + {text{ N}}_{{{text{fix}}}} – {text{ N}}_{{{text{har}}}} \ {text{N}}_{{{text{input}}}} & = {text{ N}}_{{{text{har}}}} + {text{ N}}_{{{text{sur}}}} + {text{ soil N change in stock }};;;( approx 0{text{ in long run}}) \ end{aligned}$$
where Nfer, Ndep and Nfix represent N from fertilizer, atmospheric deposition and biological fixed N, respectively. Seed N was negligible as it was present in a very small amount compared to the fertilization input19. The total N deposition of spring wheat field and biological N fixation were adopted according to Li et al. (2020). Nhar refers to the harvested N in spring wheat.Economic analysisThe inputs into local spring wheat production included chemical fertilizer, irrigated water, agricultural chemicals, seed, mechanical effort and labor costs, while income was obtained from the grain and wheat straw. The net income was determined from the difference between the total output and total input. The irrigated water, agricultural chemicals, seed, mechanical effort and labor costs were the same for the different treatments. The prices of the input and output materials were determined according to the average local market prices, and fluctuations were not considered among years.Statistical analysesThe results were analyzed using SPSS software (version 19.0; SPSS Inc., Chicago, IL, USA). Analysis of variance (ANOVA) and the least significant difference (LSD) test were used, and a P value of 0.05 was considered significant. More

Smith, K. F., Sax, D. F. & Lafferty, K. D. Evidence for the role of infectious disease in species extinction and endangerment. Conserv. Biol. 20, 1349–1357 (2006).Article 

Google Scholar 
Daszak, P., Cunningham, A. A. & Hyatt, A. D. Emerging infectious diseases of wildlife—Threats to biodiversity and human health. Science 287, 443–449 (2000).Article 
ADS 

Google Scholar 
Pedersen, A. B., Jones, K. E., Nunn, C. L. & Altizer, S. Infectious diseases and extinction risk in wild mammals. Conserv. Biol. 21, 1269–1279 (2007).Article 

Google Scholar 
Kołodziej-Sobocińska, M. Factors affecting the spread of parasites in populations of wild European terrestrial mammals. Mamm. Res. 64, 301–318 (2019).Article 

Google Scholar 
Borowik, T., Ratkiewicz, M., Maślanko, W., Duda, N. & Kowalczyk, R. Too hot to handle: Summer space use shift in a cold-adapted ungulate at the edge of its range. Landsc. Ecol. 35, 1341–1351 (2020).Article 

Google Scholar 
Demiaszkiewicz, A. W. et al. The Nematodes Thelazia gulosa Railiet and Henry, 1910 and Thelazia skrjabini Erschov, 1928 as a cause of blindness in European Bison (Bison bonasus) in Poland. Acta Parasitol. 65, 963–968 (2020).Article 

Google Scholar 
Anderson, R. C. (ed.) Nematode Parasites of Vertebrates. Their Development and Transmission (CAB International, 1992).
Google Scholar 
Otranto, D. & Traversa, D. Thelazia eyeworm: An original endo- and ecto-parasitic nematode. Trends Parasitol. 21, 1–4 (2005).Article 

Google Scholar 
Bradbury, R. S. et al. A second case of human conjunctival infestation with Thelazia gulosa and a review of T. gulosa in North America. Clin. Infect. Dis. 70, 518–520 (2020).
Google Scholar 
Lyons, E. T., Drudge, J. H. & Tolliver, S. C. Thelazia lacrymalis in horses in Kentucky and observations on the face fly (Musca autumnalis) as a probable intermediate host. J. Parasitol. 62, 877–880 (1976).Article 

Google Scholar 
Giangaspero, A., Traversa, D. & Otranto, D. Ecology of Thelazia spp. in cattle and their vectors in Italy. Parasitologia 46, 257–259 (2004).
Google Scholar 
Kostyra, J. The clinical picture and treatment of bovine thelaziosis (Przebieg i leczenie telazjozy u bydła, in Polish). Med. Weter. 16, 584–587 (1960).
Google Scholar 
Rosłan, J. Badania nad telazjozą bydła w Polsce. Wiad. Parazyt. 11, 73–79 (1965).
Google Scholar 
Kozakiewicz, B. Bovine thelaziosis in Żuławy (Telazjoza bydła na Żuławach, in Polish). Med. Weter. 27, 241–243 (1971).
Google Scholar 
Samojlik, T. (ed.) European Bison. The Nature Monograph (Springer, 2013).
Google Scholar 
Raczyński, J. (ed.) European Bison Pedigree Book (Księga rodowodowa żubrów, in Polish) 1–82 (Białowieski Park Narodowy, 2019).
Google Scholar 
Pucek, Z. et al. (eds) European Bison. Status Survey and Conservation Action Plan (IUCN/SSC Action Plans for the Conservation of Biological Diversity, 2004).
Google Scholar 
Buchmann, K., Christiansen, L.-L., Kania, P. W. & Thamsborg, S. M. Introduced European bison (Bison bonasus) in a confined forest district: A ten year parasitological survey. Int. J. Parasitol. Parasites Wildl. 18, 292–299 (2022).Article 

Google Scholar 
Giangaspero, A., Otranto, D., Vovlas, N. & Puccini, V. Thelazia gulosa Railliet & Henry, 1910 and T. skrjabini Erschow, 1928 infection in southern Europe (Italy). Parasite 7, 327–329 (2000).Article 

Google Scholar 
Klësov, M. D. Contribution to the question of the biology of two nematodes of the genus Thelazia Bosc, 1819, parasites of the eyes of cattle. Dokl. Akad. Nauk SSSR 75, 591–594 (1950).
Google Scholar 
Arbuckle, J. B. & Khalil, L. F. A survey of Thelazia worms in the eyelids of British cattle. Vet. Rec. 102, 207–210 (1978).Article 

Google Scholar 
Tweedle, D. M., Fox, M. T., Gibbons, L. M. & Tennant, K. V. Change in the prevalence of Thelazia species in bovine eyes in England. Vet. Rec. 157, 555–556 (2005).Article 

Google Scholar 
Kolstrup, N. Thelazia skrjabini in Danish cattle. Nord. Vet. Med. 26, 459–462 (1974).
Google Scholar 
Deak, G., Ionică, A. M., Oros, N. V., Gherman, C. M. & Mihalca, A. D. Thelazia rhodesi in a dairy farm in Romania and successful treatment using eprinomectin. Parasitol. Int. 80, 102183 (2021).Article 

Google Scholar 
Dróżdż, J. A study on helminth and helminthiases in bison, Bison bonasus (L.) in Poland. Acta Parasitol. Pol. 9, 55–96 (1961).
Google Scholar 
Dróżdż, J., Demiaszkiewicz, A. W. & Lachowicz, J. The helminth fauna of free-ranging European bison, Bison bonasus (L.). Acta Parasitol. Pol. 34, 117–124 (1989).
Google Scholar 
Otranto, D., Tarsitano, E., Traversa, D., De Luca, F. & Giangaspero, A. Molecular epidemiological survey on the vectors of Thelazia gulosa, Thelazia rhodesi and Thelazia skrjabini (Spirurida: Thelaziidae). Parasitology 127, 365–373 (2003).Article 

Google Scholar 
Viney, M. E. & Graham, A. L. Patterns and processes in parasite co-infection. Adv. Parasitol. 82, 321–369 (2013).Article 

Google Scholar 
Demiaszkiewicz, A. W. & Kaczor, S. Thelaziasis in European bison in Bieszczady—A case report. Życie Weter. 90, 108–110 (2015).
Google Scholar 
Thompson, R. C. Parasite zoonoses and wildlife: One health, spillover and human activity. Int. J. Parasitol. 43, 1079–1088 (2013).Article 

Google Scholar 
Naem, S. Thelazia rhodesi (Spirurida, Thelaziidae), bovine eyeworm: Morphological study by scanning electron microscopy. Parasitol. Res. 100, 855–860 (2007).Article 

Google Scholar 
Krasińska, M. & Krasiński, Z. A. Composition, group size, and spatial distribution of European bison bulls in Białowieża Forest. Acta Theriol. 40, 1–21 (1995).Article 

Google Scholar 
Krasińska, M., Krasiński, Z. A. & Bunevich, A. N. Factors affecting the variability in home range size and distribution in European bison in the Polish and Belarussian parts of the Białowieża Forest. Acta Theriol. 45, 321–334 (2000).Article 

Google Scholar 
Lloyd, S. Effect of pregnancy and lactation upon infection. Vet. Immunol. Immunopathol. 4, 153–176 (1983).Article 

Google Scholar 
Martin, L. B., Weil, Z. M. & Nelson, R. J. Seasonal changes in vertebrate immune activity: Mediation by physiological trade-offs. Philos. Trans. R. Soc. Lond. B Biol Sci. 363, 321–339 (2008).Article 

Google Scholar 
Muturi, K. N. et al. The effect of dietary polyunsaturated fatty acids (PUFA) on infection with the nematodes Ostertagia ostertagi and Cooperia oncophora in calves. Vet. Parasitol. 129, 273–283 (2005).Article 

Google Scholar 
Krasnov, B. R. & Matthee, S. Spatial variation in gender-biased parasitism: Host-related, parasite-related and environment-related effects. Parasitology 137, 1527–1536 (2010).Article 

Google Scholar 
Dunn, A. M. (ed.) Veterinary Helminthology (William Heinemann Medical Books, 1978).
Google Scholar 
Khedri, J., Radfar, M. H., Borji, H. & Azizzadeh, M. Epidemiological survey of bovine thelaziosis in southeastern of Iran. Iran. J. Parasitol. 11, 221–225 (2016).
Google Scholar 
Draber-Mońko, A. Synanthropic Calyptrata in selected habitats in Poland. Wiad. Parazytol. 32, 411–418 (1986).
Google Scholar 
Draber-Mońko, A. Dipterans of the families Muscidae (Muscinae) and Scathophagidae (Diptera, Calyptrata) of the Świętokrzyski Region. Fragmenta Faunistica 36, 205–233 (1993).Article 

Google Scholar 
Nosal, P., Kowal, J., Węglarz, A. & Wyrobisz-Papiewska, A. The occurrence and diversity of flies (Diptera) related to ruminant farming in southern Poland. Ann. Parasitol. 65, 357–363 (2019).
Google Scholar 
Cotuțiu, V. D. et al. Thelazia lacrymalis in horses from Romania: Epidemiology, morphology and phylogenetic analysis. Parasites Vectors 15, 425 (2022).Article 

Google Scholar 
Nowosad, M. Outlines of climate of the Bieszczady National Park and its buffer zone in the light of previous studies. Roczniki Bieszczadzkie 4, 163–183 (1995).
Google Scholar 
Guindon, S. & Gascuel, O. A. Simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst. Biol. 52, 696–704 (2003).Article 

Google Scholar 
Darriba, D., Taboada, G. L., Doallo, R. & Posada, D. jModelTest 2: More models, new heuristics and parallel computing. Nat. Methods 9, 77 (2012).Article 

Google Scholar 
Altman, D. G. et al. (eds) Statistics with Confidence 2nd edn, 46–47 (BMJ Books, 2000).
Google Scholar  More