A. Environmental data

Tabulated in the file data_envi.csv

At 34 crater lakes located in western Uganda (map in Fig. 1 of the article, coordinates in Table S1), surface-water conductivity and pH were recorded using a Hydrolab Quanta multimeter, to confirm freshwater conditions and to avoid low-pH habitats that are naturally less suitable for aquatic snails. Additionally, we retrieved high-quality environmental data on each of the sampled lakes from the literature. Estimates of aquatic productivity (and associated lake trophic status) are based on multi-seasonal measurements of the ubiquitous photosynthetic pigment chlorophyll-a (Chl-a) as proxy of phytoplankton biomass in 18 lakes [15], and the documented relationship between water transparency and Chl-a concentration [1] in the other lakes (details in Table S1). Chl-a values were log10-transformed to reduce skewness. Surface water transparency data are provided by [2] or [3] for all lakes except Nyamirima, for which missing data were replaced with the median values of the entire dataset (Table 1). The average slope of terrain within each crater basin was retrived from [4]. Estimates of total sediment influx to the lakes resulting from soil erosion, as proxy for the compound impact of land use on the aquatic ecosystem (through excess nutrient input and siltation), were obtained from [4]. These sediment yields were calculated by applying the Revised Universal Soil Loss Equation [5] and a sediment delivery distributed model [6] to Sentinel-2A satellite images [7]. For lakes with inflowing streams, sediment yield estimates were averaged over the range reported by [4]. Specific data on the distribution of different types of natural and anthropogenic land cover in each catchment, as inferred from Sentinel-2A satellite images [4], were used to calculate the fraction (% area) of each catchment occupied by (semi-) natural vegetation (forest, woodland or grassland), timber plantations (eucalyptus and pine), and crop fields (banana, coffee, manioc, maize and vegetable gardens) as proxy for variation in land use intensity. Finally, we retrieved reanalysis data on mean annual precipitation (MAP) and mean air temperature (MAT) at each study site from WorldClim [8].

B. Snail data

Tabulated in the file data_envi.csv

In February 2019, we collected molluscs (snails and bivalves) from the 34 selected lakes by scooping nearshore aquatic vegetation and unconsolidated sediments for 45 minutes and down to ~1m water depth. All collected specimens were sorted, counted and identified to species or genus level using shell-diagnostic features [9]. They were then sacrificed by heat shock at ~70 °C for ~45 seconds and preserved in 80% analytical-grade ethanol. We obtained in total 8865 molluscs from 32 of the surveyed lakes; none were found in Katinda and Kisibendi. For the purpose of this study we focused on B. tropicus, the most abundant species (3262 specimens; 36.8%) and found at 24 lakes. To ensure that our findings are consistent with the known regional distribution of B. tropicus, we compared our field data to those of two earlier snail surveys [3,10]. Results were highly similar, except that we did not find B. tropicus in four lakes (Kanyamukali, Kayihara, Murabyo and Nyahirya) where it had been recorded before at very low abundance. For analyses of B. tropicus distribution in relation to environmental factors we made our data set representative of all three surveys by assuming that a nominal number of five B. tropicus specimens were found at these four lakes.

C. Trematode data

Processed ASV data in the file trematode_ASV_table.csv / Maximum likelihood tree in the file ML_trematode_tree.nexus / information on species taxonomy and host preference in file trematode_spe_info.csv

Genetic screening was limited to the 16 lakes where >50 B. tropicus specimens were collected to ensure accurate characterisation of local trematode diversity and community composition. In lakes with abundant B. tropicus we limited infection screening to ca 200 randomly selected specimens per lake. In total, we screened 2385 B. tropicus (73% of those collected) for trematode infection with a multiplex polymerase chain reaction [11]. Specifically, all soft tissues were separated from the shell, homogenised with a sterilised scalpel and subjected to genomic DNA extraction using the E.Z.N.A. Mollusc DNA Kit (OMEGA Biotek, Norcross, GA, USA). DNA extracts were diluted in ultrapure water at 1:10 concentration and screened for the presence of trematode DNA using multiplex PCR. Subsequently, we genotyped the trematodes of all infected B. tropicus (n=861, or 36.1% of those screened) with high-throughput amplicon sequencing (HTAS; [12]). The sequencing data were subjected to quality control and processed into Amplicon Sequence Variants (ASVs) using dada2 [13]. Our quality-controlled ASVs are deposited in GenBank under accession numbers OQ548105-OQ549888 (ITS2); OQ543469-OQ543563 (cox1 I); OQ606413-OQ606758 (cox1 II); OQ573734-OQ574607 (NAD1) and OQ574626-OQ575328 (cytb). DNA vouchers are deposited at the Royal Museum for Central Africa (Tervuren, Belgium) under BE_RMCA_MOL_DNA codes numbered 000001 to 002472. Finally, we used these sequences to identify the trematode species present via BLAST and phylogenetic analyses using the methodology detailed in supplementary materials (Text S1).

We characterised trematode infracommunities from all ASVs recovered per infected snail. Using these infracommunities we reconstructed the total trematode communities (i.e., component communities) infecting the B. tropicus population of each studied lake. The composition of these communities was tabulated in the form of lake-specific abundance matrices which we used to estimate local trematode species richness and Hill-Shannon diversity (a version of the Shannon diversity index expressed in units of species) using iNEXT 2.0.20 [14]. To obtain comparable estimates of community diversity across lakes, we standardised sampling by estimating Hill-Shannon diversity at 0.95 coverage of the trematode community.We also characterised the phylogenetic diversity of each lake’s trematode community by computing the mean pairwise phylogenetic distance (MPD; [41]) among all trematodes recovered from a lake, using picante 1.8.2 [15]. Finally, based on the observed number of parasite infections, we also calculated the adjusted trematode abundance per lake, which accounts for the bias of characterising parasite infracommunities from ca 200 snails in lakes where >200 specimens were collected. Per lake, we multiplied the observed number of infections by the ratio between the total number of B. tropicus collected and those screened for trematode infections.

Finally, to understand how anthropogenic impacts propagate through the studied system we classified the trematode taxa within each community according to their inferred final host(s), namely livestock; birds; multiple final hosts (these are versatile trematode taxa able to infect either mammals, birds or other animals); and unknown host animals (See Table S3 and references herein).

1. Verschuren D, Plisnier P-D, Cocquyt C, Hughes H, Lebrun J, Gelorini V, Rumes B, Mahy G, André L. 2011 Climatic and anthropogenic impacts on African ecosystems. Final report. Research programme science for a sustainable development. Brussels: Belgian Science.

2. De Crop W, Verschuren D. 2021 Mixing regimes in the equatorial crater lakes of western Uganda. Limnologica 90, 125891. (doi:10.1016/j.limno.2021.125891)

3. Tabo Z, Neubauer TA, Tumwebaze I, Stelbrink B, Breuer L, Hammoud C, Albrecht C. 2022 Factors controlling the distribution of intermediate host snails of Schistosoma in crater lakes in Uganda: A machine learning approach. Front. Environ. Sci. 10, 871735. (doi:10.3389/fenvs.2022.871735)

4. De Crop W, Verschuren D, Ryken N, Basooma R, Okuonzia JT, Verdoodt A. 2023 Accelerated soil erosion and sedimentation associated with agricultural activity in crater-lake catchments of western Uganda. Land 12, 976. (doi:10.3390/land12050976)

5.  Renard KG, Foster GR, Weesies GA, McCool DK, Yoder DC. 1997 Predicting soil erosion by water: A guide to conservation planning with the revised universal soil loss equation (RUSLE). U.S. Department of Agriculture, Agricultural Research Service.

6. Ferro V, Porto P. 2000 Sediment delivery distributed (SEDD) Model. J. Hydrol. Eng. 5, 411–422. (doi:10.1061/(ASCE)1084-0699(2000)5:4(411))

7. E.S.A. 2016 Copernicus Sentinel-2A. Data; ESA: Paris, France.

8.  Fick SE, Hijmans RJ. 2017 WorldClim 2: new 1‐km spatial resolution climate surfaces for global land areas. Intl Journal of Climatology 37, 4302–4315. (doi:10.1002/joc.5086)
9. Brown DS. 1994 Freshwater snails of Africa and their medical importance (2nd edn), pp. 218-260; 321-360; 377-419. CRC Press.
10. Tumwebaze I, Clewing C, Dusabe MC, Tumusiime J, Kagoro-Rugunda G, Hammoud C, Albrecht C. 2019 Molecular identification of Bulinus spp. Intermediate host snails of Schistosoma spp. In crater lakes of western Uganda with implications for the transmission of the Schistosoma haematobium group parasites. Parasites & Vectors 12, 565. (doi:10.1186/s13071-019-3811-2
11.  Schols R, Carolus H, Hammoud C, Mulero S, Mudavanhu A, Huyse T. 2019 A rapid diagnostic multiplex PCR approach for xenomonitoring of human and animal schistosomiasis in a ‘One Health’ context. Transactions of The Royal Society of Tropical Medicine and Hygiene 113, 722–729. (doi:10.1093/trstmh/trz067)
12. Hammoud C, Mulero S, Van Bocxlaer B, Boissier J, Verschuren D, Albrecht C, Huyse T. 2022 Simultaneous genotyping of snails and infecting trematode parasites using high‐throughput amplicon sequencing. Molecular Ecology Resources 22, 567–586. (doi:10.1111/1755-0998.13492)

13. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. 2016 DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods 13, 581–583. (doi:10.1038/nmeth.3869)

14. Hsieh TC, Ma KH, Chao A. 2016 iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods Ecol Evol 7, 1451–1456. (doi:10.1111/2041-210X.12613)

15. 42. Kembel SW, Cowan PD, Helmus MR, Cornwell WK, Morlon H, Ackerly DD, Blomberg SP, Webb CO. 2010 Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26, 1463–1464. (doi:10.1093/bioinformatics/btq166)