Aircraft wastewater surveillance correlates with passenger SARS-CoV-2 variants detected during mandatory border testing

Global air travel plays a central role in the rapid international spread of infectious diseases, including SARS-CoV-2. While individual passenger testing has been widely used to monitor viral importation, it is resource-intensive and may miss asymptomatic cases. Aircraft wastewater surveillance has emerged as a non-invasive, scalable alternative for monitoring pathogen circulation among travelers. In this study, we evaluated the ability of aircraft wastewater to reflect SARS-CoV-2 variant distributions among passengers under real-world conditions. Wastewater samples and matched clinical data were collected from 86 international flights (42 from the UK and 44 from the USA) arriving in Israel during May 2022, when mandatory SARS-CoV-2 testing of passengers was still in place. Wastewater samples were processed using standardized concentration and RNA extraction protocols, followed by RT-qPCR detection and whole-genome sequencing. Variant frequencies in wastewater were compared to those obtained from passenger clinical samples using bioinformatic pipelines and statistical analyses, including correlation testing. SARS-CoV-2 RNA was detected in over 90% of aircraft wastewater samples, despite low infection prevalence among passengers (1.8%–3.4%). Sequencing revealed that dominant variants in wastewater (primarily Omicron BA.2, followed by BA.5 and BA.4) were consistent with those identified in passenger samples and with globally circulating lineages. Strong correlations were observed between wastewater-derived and clinical variant distributions (r = 0.84 for UK flights and r = 0.86 for USA flights). Viral load was positively associated with genome coverage, although it did not correlate with passenger numbers or infection rates. Aircraft wastewater surveillance reliably captured SARS-CoV-2 variant profiles among international travelers, even at low infection prevalence. While limitations such as variability in passenger behavior and incomplete genome coverage exist, this approach provides a sensitive and complementary tool to traditional surveillance methods. Integration of aircraft wastewater monitoring with existing genomic surveillance systems could enhance early detection of emerging variants and strengthen global public health preparedness.

Global air travel enables the rapid transport of billions of people across international borders each year, fundamentally shaping the dynamics of infectious disease spread. The aviation network has played a key role in the transboundary transmission of respiratory pathogens, including influenza viruses (1), SARS-CoV-1 (2), and more recently, SARS-CoV-2. In response to the COVID-19 pandemic, public health authorities worldwide implemented a suite of travel-related control measures, ranging from pre-departure testing and vaccination requirements to post-arrival screening and quarantine protocols. However, these interventions were not always sufficient to prevent the movement of asymptomatic or undetected cases, underscoring the need for complementary surveillance strategies to monitor viral dissemination through international travel (3, 4).

Further to the increased utilization of wastewater-based epidemiology approaches for infectious disease surveillance, aircraft wastewater has recently emerged as a promising source for pathogen surveillance at international borders. In 2019, sequencing of aircraft wastewater was first used to detect both enteric and respiratory viruses, revealing viral sig-natures associated with specific geographic regions (5). Although SARS-CoV-2 is primarily transmitted via the respiratory route, fecal shedding has been reported in approximately 30%–60% of infected individuals (6, 7), enabling its indirect detection in wastewater, as a complementary method to individual-level testing. Subsequent investigations have shown that SARS-CoV-2 RNA can be detected in aircraft wastewater even in the absence of confirmed clinical cases among passengers and that emerging variants can be identified at an early stage (8–10). High detection rates of viral RNA in aircraft wastewater, often coupled with successful sequencing of circulating variants, have been consistently reported (11–15). More recently, the approach has been expanded to monitor multiple viral pathogens such as norovirus, rhinovirus, adenovirus, and influenza A, demonstrating its broader applicability for global surveillance (16).

Aircraft wastewater testing represents a scalable and minimally invasive approach for large-scale bio-surveillance. However, its implementation is subject to several constraints, including limited access to lavatory waste systems, potential impairment by disinfectants or cleaning agents, and variability in passenger behavior. Notably, only an estimated 13% of passengers on short-haul flights (<6 h) and approximately 36% on long-haul flights utilize onboard lavatories for defecation, thereby influencing the likelihood of viral shedding into the wastewater system and impacting the sensitivity of pathogen detection (17).

The infrastructure required for effective coverage of global long-haul flights has been modeled, with estimates indicating that thousands of weekly samples would be needed to monitor even 10% of air travelers, depending on factors such as the pathogen’s shedding, its nucleic acid integrity, or flight route and characteristics (18).

Recent studies demonstrate that aircraft wastewater surveillance is a sensitive, non-invasive tool for detecting SARS-CoV-2 infections and variants among international travelers. Studies in France, the UK, and the USA confirmed high positivity rates and successful identification of Omicron sub-lineages via sequencing. Collectively, these findings support aircraft wastewater testing as a scalable approach for global bio-surveillance without disrupting routine airport operations.

In this study, we evaluated the potential of aircraft wastewater surveillance as a complementary tool for monitoring SARS-CoV-2 infections and tracking variant importation among international travelers arriving in Israel. Throughout the SARS-CoV-2 pandemic, Israel ministry of health has been conducting genomic sequencing of clinical and environ-mental samples via the national SARS-CoV-2 sequencing consortium, a collaboration of multiple laboratories across the country. A key part of the surveillance was mandatory testing upon arrival for individuals returning from abroad, which enabled the early detection of imported variants and efforts to curb their spread. As this policy was set to end in late May 2022, alternative methods such as aircraft wastewater surveillance for monitoring incoming variants were investigated. Leveraging a unique opportunity event for direct comparison between wastewater samples and clinical testing of passengers from the same flights, we aimed to assess the accuracy and applicability of this approach for variant tracking from both near and long-haul destinations during a critical transition period in Israel's border testing policy. Our findings may contribute to the growing body of evidence supporting the role of aircraft wastewater as part of global public health surveillance frameworks.

Clinical passenger sample collection and SARS-CoV-2 variant analysis

Until the end of May 2022, all individuals returning to Israel from international travel were required to undergo mandatory SARS-CoV-2 testing upon arrival at Ben Gurion Airport via commercially available real-time PCR tests (most commonly used Seegene Allplex) conducted by the Test&Go service provided by Femi, a contractor selected by the Israel Ministry of Health (iMoH) to implement the airport COVID-19 testing program. All positive samples were sequenced by Israel's national SARS-CoV-2 sequencing consortium laboratories, including the iMoH Central Virology Laboratory (CVL), Sourasky (Ichilov) Medical Center, and Shamir Medical Center (see Sections 3.5 and 3.6 for details on sequencing and downstream analyses), to identify circulating and imported variants and support epidemiologic tracking. This study analyzed 42 direct flights from a nearby origin, the United Kingdom (UK; ∼5 h), and 44 direct flights from a distant origin, the United States of America (USA; ∼13–17 h), all arriving in Tel Aviv during May 2022. Passenger samples from these flights were identified via the Israeli Population and Immigration Authority. After anonymization prior to receipt by the testing laboratories, the samples were tested for SARS-CoV-2, and positive samples were sequenced to determine the SARS-CoV-2 variant. The numbers of total passengers and SARS-CoV-2 infected passengers for each flight are provided in Supplementary Table 1. The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of the Sheba Medical Center institutional review board (7045-20-SMC). Patient consent was waived as the study utilized residual clinical samples and the analysis used anonymous samples.

Wastewater collection and processing for national SARS-CoV-2 surveillance

As part of the national SARS-CoV-2 wastewater surveillance program, 150 sampling sites (mostly in manholes) across Israel, representing approximately 76% of the national population, were monitored twice weekly during April to June 2022 (Supplementary Table 2). Collected samples were transported under refrigerated conditions (4 °C) to four designated laboratories: the Central Virology Laboratory (CVL), Miloda Laboratory, Ariel Kushmaro's Laboratory (Ben-Gurion University), and Jacob Moran-Gilad's Laboratory (Ben-Gurion University) for standardized processing and analysis.

For viral concentration, 20 mL of each sample was filtered through a 0.45 µm, 47 mm diameter electronegative MCE membrane (Merck Millipore Ltd). The membrane was then immediately transferred into 2 mL of lysis buffer (19).

Nucleic acids were extracted from the filters using the NucleoSpin RNA extraction kit (Macherey-Nagel, Germany) in the Miloda, Kushmaro, and Moran-Gilad laboratories, following the manufacturer's instructions, with an elution volume of 35 µL (20). At the CVL, total nucleic acid extraction was performed using the NucliSENS easyMAG system (bioMérieux, France), with an elution volume of 55 µL.

Aircraft wastewater sample processing

Aircraft wastewater was collected from individual international flights using designated vacuum trucks. During the discharge of the truck's contents into the municipal sewage system, a grab sample was collected to ensure flight-specific attribution. Samples were immediately refrigerated (4 °C) and transported to the same designated laboratories (CVL, Miloda, Ariel Kushmaro's Laboratory, and Jacob Moran-Gilad's Laboratory) for standardized SARS-CoV-2 detection and quantification.

Each sample was homogenized, and 45 mL of raw wastewater was processed by centrifugation at 4,696 g for 5 min at 4° C. From the resulting supernatant, 5 mL was transferred into a 15 mL conical tube for RNA extraction. RNA was extracted in duplicate using the Zymo Environ Water RNA Kit (Zymo Research, CA, USA), following the manufacturer's wastewater-specific protocol. The final elution volume for each extraction was 40 μL.

RT-qPCR analysis for SARS-CoV-2 RNA detection

SARS-CoV-2 RNA in wastewater samples from the national surveillance and from aircrafts was detected and quantified using reverse transcription quantitative polymerase chain reaction (RT-qPCR) assays targeting the E and/or N2 genes, as previously described (21, 22). CVL and Miloda laboratories employed the Reliance One-Step RT-qPCR Master Mix (Bio-Rad, Hercules, CA, USA) in a 20 µL final reaction volume, with 8 µL of RNA template per reaction. Ariel Kushmaro's and Jacob Moran-Gilad's laboratories used the One Step PrimeScript III RT-qPCR mix (RR600, Takara, Japan), with 5 µL of RNA template in a 20 µL reaction volume, following protocols optimized for SARS CoV-2 detection in wastewater.

Thermal cycling was performed on different platforms: Bio-Rad CFX96 (Bio-Rad; CVL), QuantStudio™ 5 (Thermo Fisher; Miloda), Applied Biosystems Thermocycler (Thermo Fisher; Ariel Kushmaro), and Mic qPCR system (Bio Molecular Systems; Moran-Gilad). Kushmaro and Moran-Gilad labs used the following cycling conditions: reverse transcription at 52 °C for 7 min, initial denaturation at 95 °C for 10 s, followed by 45 cycles of 95 °C for 5 s and 60 °C for 30s (23). CVL and Miloda followed E gene-specific cycling: 45 °C for 10 min 10 s, followed by 45 cycles of 95 °C for 2 min 20 s, 95 °C for 4 s, and 60 °C for 22 s, with fluorescence acquisition at the end of each cycle (24). All laboratories used MS2 bacteriophage (10⁴–10⁵ copies) as an internal extraction control, spiked into the lysis buffer. Quantification was performed using standard curves generated from serial dilutions of known RNA standards. Cq values from duplicate reactions were averaged and converted to viral concentrations, based on the total extraction volume and sample input. Cq values >40 were considered negative for SARS-CoV-2 RNA. Assay sensitivity and reproducibility were confirmed by repeated testing of RNA standards (25).

Quantification of SARS-CoV-2 RNA in both wastewater and aircraft samples was performed in duplicate. Viral concentrations were performed using standard calibration curves generated from the molecular assay, enabling estimation of viral copies per RT-qPCR reaction. The calculated values were then adjusted by accounting for the nucleic acid elution volume relative to the original sample volume. Finally, the results were normalized to express viral concentrations per liter of wastewater (copies per liter). For national wastewater surveillance samples, viral load calculations were based on all samples collected from the same site within a given epidemiological week (26) (Supplementary Table 2). For sequencing, a single positive replicate per site and sampling week was selected for downstream analysis.

Library preparation and sequencing

Extracted samples underwent library preparation using COVID-seq library preparation kit, as per manufacturer's instructions (Illumina). Library validation and mean fragment size was determined by TapeStation 4200 via DNA HS D1000 kit (Agilent). Libraries were pooled, denatured and diluted to 10pM and sequenced on NovaSeq 6000 platform using an SP flow cell with 2X75 base-pair paired-end reads (Illumina).

Bioinformatic and statistical analyses

Fastq files from both clinical and wastewater samples underwent quality control using FastQC and MultiQC (27) and low-quality sequences were filtered using trimmomatic (28). Sequences were mapped to the SARS-CoV-2 reference genome (NC_045512.2) using the Burrows–Wheeler aligner (BWA) MEM (29). Resulting BAM files were sorted, indexed and underwent quality control using SAMtools suite (30).

Further processing was conducted differently for clinical and wastewater samples, as follows. For clinical samples, consensus fasta sequences were generated using iVar (31), with positions having a depth <5 designated as “N.” SARS-CoV-2 variants were assigned using Nextclade (32). For wastewater sample sequences, which represent a composite signal from multiple individuals, variant frequencies were estimated using a custom python-based algorithm, as previously described (19). Briefly, for each sample, the algorithm processes the BAM file and estimates variant frequencies based on sequencing depth at variant-defining genomic positions. At each such position, the proportion of reads carrying the defining mutation is calculated relative to the total depth. These proportions are then averaged across all defining positions for a given variant to obtain its frequency within the sample. Variant frequencies at the group level were subsequently calculated by averaging the per-sample frequencies across all samples within the group (Supplementary Methods). Supplementary Table 3 lists the defining mutations for each sub-variant circulating at time of the study that were used as inputs for the algorithm.

The globally predominant variants during the study period (April-June 2022) were obtained from CoV-Spectrum (33), which compiles data based on SARS-CoV-2 sequences deposited in public databases.

Additional analyses were conducted in R (version 4.3.2), including Pearson correlation and t.test for statistical significance (base stats package) and data visualization (ggplot2 package).

Sequence data of the aircraft wastewater and passenger clinical samples were deposited to GISAID (accessions: EPI_ISL_20381695–20381780, EPI_ISL_20381808–20382052).

SARS-CoV-2 variant circulation and importation in Israel, May 2022

During the study period of May 2022, the global SARS-CoV-2 landscape was dominated by the Omicron lineage, with BA.2 as the most prevalent variant, followed by BA.5 and BA.4. This pattern was similarly observed in Israel, as indicated by both clinical testing data and nationwide wastewater surveillance (Figure 1, Supplementary Figure 1).

SARS-CoV-2 detection in aircraft passengers and wastewater

To assess whether aircraft wastewater reflects passenger-derived variants, we analyzed wastewater samples from 42 flights arriving from the UK and 44 flights from the USA during May 2022. SARS-CoV-2 RNA was detected in the majority of flights, with 38/42 (90.5%) UK flights and 42/44 (95.5%) USA flights testing positive.

The overall percentage of SARS-CoV-2-infected individuals among the total number of passengers on the aircrafts was low, with 1.8 ± 2.4% infected from an average of 198 ± 35 passengers in the UK flights and 3.4 ± 2.6% infected from an average of 266 ± 44 passengers in the USA flights (Figure 2A and Supplementary Table 1). The average SARS-CoV-2 viral load in aircraft wastewater samples were higher on flights from the USA (7.4 ± 28 million copies per liter) compared to flights from the UK (2.5 ± 6 million copies per liter), although this difference was not statistically significant (Figure 2B and Supplementary Table 1). The viral load was not correlated with either the number of passengers (Figure 2C) or the percentage of infected individuals on corresponding flights (Figure 2D). There was a moderate positive correlation between higher SARS-CoV-2 genomic coverages achieved in samples with higher viral loads (Figure 2E).

Characterization of SARS-CoV-2 variants in aircraft wastewater and passenger samples

We next evaluated the SARS-CoV-2 variants detected in aircraft wastewater and corresponding passenger clinical samples. Whole viral genome sequences were obtained from concentrated wastewater samples and clinical specimens. Due to typically low viral loads in aircraft wastewater, full genome coverage was limited, with UK and USA flight samples achieving mean coverages of 57 ± 33% and 63 ± 36%, respectively, despite sequencing depths exceeding 1,000× (Supplementary Table 1). Despite the overall relatively low coverages, most genomic positions defining the omicron variants (Supplementary Figure 1) were successfully sequenced, enabling variant identification even in samples with lower viral loads/genomic coverages.

Consistent with the global circulation of SARS-CoV-2 variants and circulating variants in the UK and USA during the study period (Figure 1, Supplementary Figure 1), BA.2 was the most frequently detected variant in aircraft wastewater, appearing in 75 ± 37% of samples from UK flights and 83 ± 33% from USA flights, followed by BA.5 (UK: 6 ± 20%, USA: 4 ± 17%) and BA.4 (UK: 5 ± 19%, USA: 0.3 ± 1%) (Figure 3 and Supplementary Table 4). No additional/novel variants were detected.

Comparison between SARS-CoV-2 variants in aircraft wastewater with flight passenger's variants

We next compared variants detected in passenger clinical samples with those identified in the corresponding aircraft wastewater from each flight. Variants identified from positive passenger samples were compared to wastewater-derived variants for the same flights. The overall variant frequencies were similar between wastewater and passenger samples on both UK and USA flights (Figure 3A). Variant frequencies shown represent mean per-sample frequencies within each group, calculated based on sequencing depth at variant-defining positions, and not the proportion of samples in which a variant was detected. Furthermore, a strong positive correlation was observed between variants detected in wastewater and those in passenger samples (r = 0.84, p.value = 4.7 × 10⁻ 30 for UK flights; r = 0.86, p.value = 2.2 × 10⁻⁴⁴ for USA flights; Figure 3B). Notably, the variant distribution identified in aircraft wastewater and in the passenger samples were similar to those observed in all incoming flights from the UK and USA during May 2022. They were also similar to the variants circulating in Israel at the time as reflected by national wastewater surveillance (Figure 3A).

Concurrent with the end of mandatory SARS-CoV-2 testing for returning travelers in Israel at the end of May 2022, this study aimed to evaluate whether aircraft wastewater surveillance could serve as a reliable tool for detecting SARS-CoV-2 variant importation without the need for individual passenger testing. Israel's robust national genomic clinical and environmental surveillance system, provided a unique opportunity to assess this approach within the broader context of variant monitoring.

This study demonstrated that SARS-CoV-2 variant distributions identified in aircraft wastewater closely mirrored those detected in clinical samples from passengers on the same flights. Strong positive correlations were observed for both near (UK) and distant (USA) flights, demonstrating the potential of aircraft wastewater surveillance as a proxy for monitoring variants among international travelers. Although aggregate-level concordance was high, correlations at the level of individual flights were more variable, likely reflecting differences in passenger shedding, lavatory usage behavior, technical variables influencing wastewater disposal across different aviation platforms, and the relatively small number of infected individuals contributing to wastewater in most of flights.

In the present study, a high proportion of aircraft wastewater samples were found to be SARS-CoV-2 positive, with detection rates of 90.5% among flights arriving from the UK and 95.5% among flights from the USA. These findings further support previous studies demonstrating that aircraft wastewater can serve as a highly sensitive matrix for detecting viral introductions (34). When considered alongside the broader literature, our findings reinforce that aircraft and airport wastewater provide a strong but inherently partial signal of pathogen circulation. High detection rates have been consistently reported across settings, including 64.9% positivity in repatriation flights with high predictive value for subsequent clinical cases (9) and up to 93% positivity in aircraft wastewater at major international airports (13). At the same time, modeling studies suggest that aircraft wastewater may capture only 8%–14% of infected travelers, particularly on short-haul flights, due to behavioral and biological constraints (17). These observations support the interpretation of wastewater surveillance as a complementary rather than stand-alone surveillance approach (9, 12, 13, 15). A trend toward higher viral loads was observed in wastewater from distant (USA) compared to near (UK) flights, although this difference was not statistically significant. As expected, viral load was positively correlated with sequencing coverage across both flight groups. In contrast to the high detection rates in aircraft wastewater, the overall prevalence of SARS-CoV-2 infection among passengers was relatively low, averaging 1.8% on flights from the UK and 3.4% on flights from the USA. Reported passenger positivity rates are also low, ranging between 20% and 38% (12). The prevalence in our dataset was the lowest reported to date, likely because passengers were tested only upon entry to Israel and not at later time points. Put together, these data demonstrate that aircraft wastewater surveillance is capable of detecting SARS-CoV-2 and its variants with high sensitivity, even when infection prevalence among passengers is low, supporting its value as a scalable and non-invasive approach for international bio-surveillance.

The integration of molecular detection with next-generation sequencing further enhances the value of this approach by enabling variant identification and early warning capabilities. Previous studies have demonstrated that wastewater-based genomic surveillance can detect emerging variants weeks before clinical identification, track global transmission patterns, and monitor multiple pathogens simultaneously (35, 36). For example, aircraft wastewater surveillance has identified numerous SARS-CoV-2 variants and sub-lineages, as well as additional respiratory and enteric viruses, highlighting its potential as a multi-pathogen surveillance platform (37). Case studies have also shown its ability to detect variant importation despite pre-departure testing and vaccination requirements, underscoring limitations of individual-based border control measures (10, 14).

Notably, SARS-CoV-2 RNA was occasionally detected in wastewater even when no positive passenger samples were identified, possibly due to residual RNA from prior flights, low-level contamination or, as previously reported - infections in passengers not captured by pre-boarding respiratory testing but still shedding virus through feces (8, 9). The presence of viral RNA in aircraft wastewater was not associated with total number of passengers on the flight, highlighting the effect of additional factors such as lavatory use. As noted in a previous study, passenger behavior during lavatory use may pose a limitation in interpreting aircraft wastewater data (17). Accordingly, no correlation was observed between wastewater viral load and flight occupancy (37). Minimizing viral cross-contamination in aircraft wastewater would require dedicated cleaning of onboard holding tanks between flights. However, this would increase turnaround times, maintenance demands, and operational costs, posing challenges for tightly scheduled aircraft and potentially impacting ticket prices, thereby limiting large-scale implementation of such surveillance.

Importantly, the variants detected in aircraft wastewater samples aligned with those circulating in Israel at the time, as indicated by national clinical and wastewater genomic surveillance. They were also representative of variant trends observed in all incoming flights from the UK and USA during the study period, based on sequencing of clinical samples. While this dataset represents only a subset of total air traffic at the time, the findings suggest that aircraft wastewater reliably captured both global and local variant patterns and may even precede the detection of new variants in municipal wastewater (36).

This study offers several contributions that build upon and extend previous work in the field. In particular, the direct, flight-specific comparison of variant profiles between aircraft wastewater and matched passenger clinical samples provided a uniquely controlled assessment of concordance between these two surveillance approaches. While previous studies have successfully demonstrated the feasibility of detecting SARS-CoV-2 RNA or broad lineage patterns in aircraft wastewater, our analysis was conducted under real-world operational conditions during a period of mandatory border testing, allowing for a more detailed evaluation of variant-level agreement. Furthermore, we show that even in the absence of newly introduced variants, aircraft wastewater surveillance can reliably capture circulating lineages with high sensitivity when benchmarked against wastewater and clinical national surveillance genomic data. The inclusion of both short- and long-haul flights also allowed us to explore operational differences, though no substantial variation in correlation was observed between the two.

Limitations of this study, aside from technical challenges associated with aircraft wastewater sampling, include the restricted number of destinations analyzed, as we focused on a subset of short- and long-haul flights without layovers, which may limit the generalizability of the findings. In addition, the study was conducted over a relatively short time frame to take advantage of the final period of mandatory clinical monitoring of passengers arriving in Israel. As a result, longer-term trends in viral evolution could not be assessed, and the variants detected were predominantly BA.2, which was the dominant strain during the study period. An additional limitation is the use of short-read sequencing on composite wastewater samples. As wastewater contains a heterogeneous mixture of viral genomes from multiple individuals, reconstructing complete genomes corresponding to distinct variants or individuals is not feasible, limiting the identification of novel variants despite sensitive detection of circulating mutations. Low viral concentrations in wastewater, often diluted by large liquid volumes, may compromise sequencing performance, leading to reduced genome coverage and decreased sensitivity for variant detection. Finally, because wastewater samples represent a composite signal from multiple individuals and may be influenced by uneven distribution of infections across flights or sampling sites, estimated variant frequencies may not fully reflect true population-level prevalence.

Global modeling suggests that coordinated surveillance at ∼20 major airports, coupled with substantial sampling of inbound flights could form an efficient early warning system for fecal shed emerging pathogens (38). Our findings strengthen this concept, further reinforcing the importance of aircraft wastewater surveillance as a robust, non-invasive, and scalable approach. When integrated with existing national and international genomic monitoring systems, this approach can enhance early detection of emerging variants and contribute to broader public health preparedness and response efforts.