Localized wastewater surveillance showed correlation but no early warning during Bengaluru's Omicron wave.

Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice PLOS Glob Public Health . 2026 Apr 10;6(4):e0004684. doi: 10.1371/journal.pgph.0004684 Search in PMC Search in PubMed View in NLM Catalog Add to search Localized wastewater surveillance showed correlation but no early warning during Bengaluru’s Omicron wave Siva Athreya Siva Athreya 1 International Centre for Theoretical Sciences - TIFR, Bengaluru, India Investigation, Methodology, Supervision, Visualization, Writing – review & editing Find articles by Siva Athreya 1 , Farah Ishtiaq Farah Ishtiaq 2 Tata Institute for Genetics and Society, Bengaluru, India Data curation, Methodology, Writing – review & editing Find articles by Farah Ishtiaq 2 , Tarun Khandelwal Tarun Khandelwal 3 ARTPARK, Indian Institute of Science Campus, Bengaluru, India Data curation, Investigation, Methodology, Software, Visualization, Writing – review & editing Find articles by Tarun Khandelwal 3 , Chitra Pattabiraman Chitra Pattabiraman 3 ARTPARK, Indian Institute of Science Campus, Bengaluru, India Methodology, Resources, Supervision, Visualization, Writing – original draft, Writing – review & editing Find articles by Chitra Pattabiraman 3 , Lakshminarayana Rao Lakshminarayana Rao 4 Centre for Sustainable Technologies, Indian Institute of Science, Bengaluru, India Methodology, Resources, Writing – review & editing Find articles by Lakshminarayana Rao 4 , Rajesh Sundaresan Rajesh Sundaresan 5 Electrical Communication Engineering, Indian Institute of Science, Bengaluru, India Conceptualization, Investigation, Methodology, Project administration, Supervision, Writing – review & editing Find articles by Rajesh Sundaresan 5, * , Reshma Mohan Thattaramppilly Reshma Mohan Thattaramppilly 4 Centre for Sustainable Technologies, Indian Institute of Science, Bengaluru, India Methodology, Writing – review & editing Find articles by Reshma Mohan Thattaramppilly 4 Editor: Mohan Amarasiri 6 Author information Article notes Copyright and License information 1 International Centre for Theoretical Sciences - TIFR, Bengaluru, India 2 Tata Institute for Genetics and Society, Bengaluru, India 3 ARTPARK, Indian Institute of Science Campus, Bengaluru, India 4 Centre for Sustainable Technologies, Indian Institute of Science, Bengaluru, India 5 Electrical Communication Engineering, Indian Institute of Science, Bengaluru, India 6 Tohoku University Graduate School of Engineering School of Engineering: Tohoku Daigaku Daigakuin Kogaku Kenkyuka Kogakubu, JAPAN ✉ * E-mail: [email protected] The authors have declared that no competing interests exist. Roles Siva Athreya : Investigation, Methodology, Supervision, Visualization, Writing – review & editing Farah Ishtiaq : Data curation, Methodology, Writing – review & editing Tarun Khandelwal : Data curation, Investigation, Methodology, Software, Visualization, Writing – review & editing Chitra Pattabiraman : Methodology, Resources, Supervision, Visualization, Writing – original draft, Writing – review & editing Lakshminarayana Rao : Methodology, Resources, Writing – review & editing Rajesh Sundaresan : Conceptualization, Investigation, Methodology, Project administration, Supervision, Writing – review & editing Reshma Mohan Thattaramppilly : Methodology, Writing – review & editing Mohan Amarasiri : Editor Received 2024 Sep 2; Accepted 2025 Dec 29; Collection date 2026. © 2026 Athreya et al This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice PMCID: PMC13068238  PMID: 41961879 Abstract Wastewater surveillance is an effective tool for monitoring the spread of infectious diseases such as COVID-19. In August 2021, a citywide surveillance effort was initiated in Bengaluru to analyze viral loads from 28 sewage treatment plants (STPs). The study found a strong correlation between aggregated viral loads and citywide COVID-19 case counts. However, the lack of localized clinical data limited the ability to assess infection trends at the STP level. In this follow-up study, we incorporate granular clinical data from 198 administrative units in the city. We find similar trends between viral loads at individual STPs and the cases in their catchment areas. A typical confidence interval for the Pearson correlation between clinical cases and wastewater viral loads at an STP is approximately 0.56 – 1.00 , based on the median bounds across the STPs; at the city-level it is 0.67 – 1.00 . However, our analysis shows no reliable indication of a lead or early warning—the viral loads and reported cases rise simultaneously. It is important to note that our study is limited to the first Omicron wave of the pandemic. To quantify lead time, we used correlation and change-point analysis. These results underscore the potential of localized wastewater surveillance for real-time monitoring but highlight its limitation in early outbreak detection. Author summary We aimed to assess the effectiveness of wastewater surveillance in detecting COVID-19 trends at localized scales within Bengaluru during the first Omicron wave. To enable this localized study, we undertook two key tasks: first, we mapped the catchment areas supplying wastewater to the sewage treatment plants (STPs), and second, we incorporated clinical data from 198 administrative units, for the time period when SARS-CoV-2 viral loads were monitored. This mapping of catchment areas and integration of granular clinical data was not done in previous studies. We find that viral loads closely mirrored infection trends at a localized level; however, they did not provide an advance indication of cases, as they rose concurrently with reported infections. Introduction Environmental surveillance of wastewater has proven to be highly effective in detecting diseases in populations, with previous applications including tracking poliovirus and emerging applications such as detecting antimicrobial resistance [ 1 ]. Wastewater surveillance entails continuous, systematic sampling, and computing viral concentrations. Interpretation of these data has helped public health officials understand the prevalence of the disease and design appropriate responses [ 2 ]. During the COVID-19 pandemic, wastewater surveillance has gained significant attention worldwide, particularly due to its potential to serve as an early warning system [ 1 ]. Multiple countries have set up wastewater surveillance systems for COVID-19. Data from countries such as the Netherlands, Spain, Australia, and the USA suggest that SARS-CoV-2 RNA can be detected in sewage in advance of clinical cases [ 3 – 10 ]. While initial studies and reports were limited to developed countries, there has been a growing body of work from urban centers in developing countries supporting the use of routine wastewater surveillance to monitor the COVID-19 pandemic including the Indian cities of Mumbai, Pune, Hyderabad, Ahmedabad, and Bengaluru [ 10 – 14 ]. There has also been work on mapping wastewater catchment areas and setting up SARS-CoV-2 surveillance systems in Bangladesh, Nepal, and Malawi [ 15 , 16 ]. In low-income settings, a large fraction of the sewage is untreated, and the catchment areas for these informal sewage networks are more challenging to map [ 15 ]. Mapping catchment areas is crucial to localized wastewater surveillance, as it enables the systematic identification of specific geographic regions contributing pathogens to the wastewater system. Further, in such low-income settings, it is also useful to identify bellwether sewershed sites for permanent monitoring and to scale-up to subsites during outbreak periods to identify hotspots [ 13 ]. Multiple studies show that wastewater data can be a leading indicator for daily new cases [ 8 , 14 , 17 – 24 ]. These studies have used methods such as correlation, cross-correlation, generalized additive models, and time series analysis [ 8 , 19 – 25 ]. Most of these studies have found correlation between viral loads in wastewater and disease indicators such as daily cases or hospitalizations [ 3 , 8 , 11 , 17 , 19 – 26 ]. Correlation is sensitive to viral shedding characteristics, sampling design, the sewerage network, the size and population of the catchment area. The correlation improves with an increase in sampling frequency [ 26 , 27 ], due to reduction in estimation errors. The lead times, i.e., the number of days by which the wastewater viral load trajectories lead the clinical cases trajectories, reported in different studies vary substantially, with some studies reporting short lead times of two days and others as long as 1–2 weeks [ 8 , 17 , 19 – 25 , 28 ]. The lead times are also influenced by testing policy, access to testing, and delays in reporting cases apart from shedding dynamics of the circulating viral variant. The SARS-CoV-2 virus is shed in respiratory secretions and feces [ 29 – 33 ]. The viral load is the number of RNA copies per unit ml of wastewater. While shedding characteristics may vary across viral variants, a meta-analysis of the Omicron variant found that viral loads peaked 4–6 days after symptom onset and declined by day 10 [ 29 , 32 ]. This is consistent with a higher correlation for SARS-CoV-2 RNA levels in sewage and new cases rather than active cases or cumulative cases [ 26 ]. Infected individuals with no or mild clinical symptoms who are not tested also contribute to the viral load. Our main objective in this study is to quantify the correlation between observed wastewater viral loads at STPs and reported cases in Bengaluru, a city in India with a population of 14 million. We chose the period around the first Omicron wave, i.e., from 15 November 2021 to 31 January 2022. During this period the viral load data in wastewater samples from 26 sewage treatment plants (STPs) were available from [ 12 ] and due to comprehensive testing policy, daily cases data were available at the BBMP (now Greater Bengaluru Authority) website . We first map the catchment areas from Bengaluru for each of these STPs. We then perform correlation analysis and implement a refined change-point detection analysis to evaluate if viral load leads cases in the city during the Omicron wave. Methodology Bruhat Bengaluru Mahanagara Palike (BBMP) is Bengaluru’s administrative body that manages its various civic amenities and services. Bengaluru comprises 198 BBMP wards , which are its local administrative units. The average ward population is approximately 60,000 individuals [ 34 ]. The city produces approximately 1,440 million liters per day (MLD) of wastewater, of which about 85% is treated in 26 STPs. The STPs vary in capacities (measured in MLD) and cover over 183 out of 198 wards ( Fig 1 ). They are designed to treat and process wastewater generated from residential, commercial, and industrial sources. They are spread across the city and are operated by the Bengaluru Water Supply and Sewerage Board (BWSSB). Fig 1. The map of Bengaluru, STP locations, and ward boundaries. Open in a new tab Wards with the same texture send their wastewater to the same set of STPs. The samples were collected from the STPs shown. Wards that are not covered by testing are shown in white. Base layer: Bengaluru BBMP wards [ 35 , 36 ], Bengaluru City STPs [ 37 , 38 ]. License: Open Data Commons Open Database License (ODbL) v1.0. As stated earlier, our main objective is to understand the correlation between observed wastewater viral loads at STPs and reported cases. We chose the period around the Omicron wave for the study because data for both wastewater viral loads and daily cases were available due to diligent testing. The wastewater surveillance in Bengaluru started in August, 2021. Samples were collected from STPs using the ‘grab’ method and tested for the presence of SARS-CoV-2. The data comprises viral load values in copies/ml from 26 STPs. For the period post January 4, 2022 data is also available as part of the supplementary data in [ 12 ]. Sampling was conducted every 10 days before January, 2022 and weekly thereafter. Henceforth, we refer this as ‘weekly’ for simplicity. S3 Table shows the number of samples for each STP during the study period. BBMP collects and records the case data. The ward-level daily case numbers were reported at the BBMP website . Each case is attributed to the patient’s ward of residence. The case data on the BBMP website was originally in PDF format. Upon request, the data was shared with us in machine-readable format. City-wide cases are calculated as the sum of cases reported across all wards. Mapping STP catchments Out of 198 wards, 42 wards were mapped manually using the procedure described below. Mapping for the rest of the wards was given by BWSSB. Manually mapping wards to the respective STPs relied on utilizing a Geographic Information System (GIS) tool known as QGIS (version 3.32.2). The required input files included the geographical and administrative information of the wards in the Bengaluru Urban District, the location and capacity of each STP, and the sewer drainage network map. The information about the 198 wards in Bengaluru was compiled into a geopackage file (.gpkg), which was then imported as a layer into the QGIS project. The locations of the 26 STPs in Bengaluru were sourced from the Open City Urban Data Portal in KML file format and integrated into the QGIS project ( Fig 1 ). The accuracy of the STP locations was cross-checked using a geo-referenced raster file obtained from BWSSB. The sewer system in the Bengaluru Urban District has been designed in alignment with the natural terrain, utilizing the existing ridges and valleys. This sewerage network encompasses approximately 229 km 2 and comprises five primary drainage regions which are commonly referred to as Vrishbhavathi, Arakavathi, Hebbal, Koramangala & Challaghatta, and Suvarnamukh ( Fig 2 ). The sewer line maps in KML format with pipe diameters greater than 300 mm and less than 300 mm were obtained from the Open City Urban Data Portal. This data was also cross-checked with sewerage line maps (georeferenced raster files) obtained from the BWSSB website. A sewer line between two locations makes no distinction between the source and destination. We infer the source and destination from the knowledge of the terrain. The flow direction of sewage from each ward was determined based on the elevation of the valleys. Similarly, an assessment of drainage line density in each ward was conducted, facilitating the mapping of each ward to the corresponding STP. For instance, we infer that Sanjay Nagar ward sends its wastewater to Hebbal STP by inspecting the sewer lines (see S1 Fig . for the map). S2 Fig . shows the sewer line map of the entire city. To the best of our knowledge, this is the first focused effort to map each ward to its STPs in the Bengaluru Urban District. The result of this mapping is given in Section Wards in the catchment of an STP . Fig 2. The Valleys of Bengaluru. Open in a new tab The shapefile for the valleys of Bengaluru was created in QGIS (QGIS.org 3.32.2 ‘Lima’ 2023) using a georeferenced raster file from literature [ 39 ]. Base layer: Bengaluru BBMP wards [ 35 , 36 ]. License: Open Data Commons Open Database License (ODbL) v1.0. Pre-processing of viral loads Wastewater samples were collected at the inlets of the STPs to measure the viral load at each site. We interpolate the sparse data (based on actual dates of collection) to obtain daily viral loads. We use quadratic interpolation to estimate missing values in our data. We implement this using the interpolate() function from the Pandas Python module. If the interpolation outputs any negative values for viral loads, we project it to the nearest value on the set of valid viral loads, that is, we set them to zero. For an STP S , we denote its viral load (possibly interpolated) on day i as v S ( i ). To compute citywide viral load, we combined the viral loads using the following formula: v ( i ) = ∑ S ∈ 𝒮 ( v S ( i ) · M L D S ) ∑ S ∈ 𝒮 M L D S , (1) where 𝒮 is the set of STPs and MLD S is the capacity of STP S ∈ 𝒮 . In other words, the viral load is calculated as the weighted average of the viral loads from all the STPs, with weights corresponding to their capacities. In the above, we use STP capacity rather than daily or weekly flow-rate averages because (1) it represents the hydraulic and organic load the STP is engineered to treat while maintaining the effluent discharge standards, and (2) all STPs are operating continuously at full capacity and the city generates more sewage than the STP capacity. We also combine viral loads from multiple STPs if they have a common catchment area, i.e., if the STPs receive wastewater from the same set of wards. This happens at two locations. Rajacanal (80 MLD) and Horamavu (20 MLD) are two STPs located close to each other that have a common catchment area. Hence, we treat the pair as a single STP and replace their daily viral loads with their weighted average. Similarly, Nagasandra (40 MLD) is a combination of two STPs: Nagasandra Old (20 MLD) and Nagasandra New (20 MLD). See S1 Text for an illustrative example. Pre-processing of case counts The catchment area of an STP may not align perfectly with ward boundaries. To relate viral loads to case counts, we would ideally require cases to be reported at the catchment level. However, since cases are reported at the ward level, we estimate the number of cases within each STP’s catchment area. We describe how we estimate below. For a ward that sends wastewater to multiple STPs, we divide the cases from that ward equally (a simplifying assumption) among the STPs. For an STP S , the case count on a day i is defined as: c ( i ) = ∑ j = 1 K S c ~ j ( i ) k j , (2) where K S is the number of wards in the catchment area of STP S , k j denotes the number of STPs into which the j -th ward drains, and c ~ j ( i ) denotes the number of cases on day i in the j -th ward. We calculate the case counts from 15 November 2021 to 31 January 2022. Computing correlations Previous studies have reported linear and non-linear associations between viral loads in wastewater and COVID-19 case counts in catchment areas. For instance, linear, log-linear and non-linear correlations have been reported in [ 27 , 40 ] and [ 41 ], respectively. In our case, to determine the relationship between viral load and COVID-19 cases, we first plotted the data from all STPs on a scatter plot. The data revealed a linear trend, consistent across STPs. S3 Fig . shows an example using four STPs. Viral load is expected to be proportional to cases because, in a constant population, an increase in infected individuals correlates with more viral copies being shed into the wastewater. If the number of cases doubles, the total viral load in wastewater should also double, assuming similar viral shedding rates [ 29 , 31 ]. Pearson correlation is then used to quantify the strength of the linear relationship, providing a robust measure of how closely variations in viral load align with changes in case numbers. Let 𝐱 = ( x ( 1 ) , … , x ( n ) ) and 𝐲 = ( y ( 1 ) , … , y ( n ) ) be vectors of length n . The Pearson correlation between x and y , denoted as ρ x y , is given by the formula: ρ x y = ∑ i = 1 n ( x ( i ) − x ¯ ) ( y ( i ) − y ¯ ) ∑ i = 1 n ( x ( i ) − x ¯ ) 2 ∑ i = 1 n ( y ( i ) − y ¯ ) 2 (3) where x ¯ and y ¯ are the mean values of the vectors x and y , respectively. A Pearson correlation value of 1 indicates a perfect positive correlation, -1 indicates a perfect negative correlation, and 0 indicates no correlation. We now see how to compute the correlation between cases and viral loads. We begin by defining the following terms for a given Sewage Treatment Plant (STP), S : Time period T : Let T denote the dates from 15 November 2021 to 31 January 2022. There are n = 77 days in this period. Shift parameter η : The parameter η is an integer between -14 and 14 that indicates a shift in the time period T . T η refers to the 77-day period starting from η days before 15 November 2021. For example, T 3 denotes the dates from 12 November 2021 to 28 January 2022. Similarly, T − 2 denotes the dates from 17 November 2021 to 2 February 2022. Case Vector ( 𝐜 = ( c ( 1 ) , … , c ( n ) ) ) : The vector representing the daily COVID-19 cases reported within the catchment area of S during the period T . The number of cases on day i is calculated according to Eq. 2 . Interpolated Daily Viral Load Vector ( 𝐯 ~ η ) : The vector obtained by interpolating viral load measurements taken weekly from S using quadratic interpolation. The viral loads are for the days in the period T η . True Viral Load Vector ( 𝐯 η ) : The hypothetical vector of the true daily viral load values we would have obtained if measurements had been made daily during T η . This is the vector that 𝐯 ~ η approximates. When calculating the correlation between cases and viral loads, we keep the case vector fixed, i.e., the case vector is always for the period T . The viral load vector is shifted relative to the case vector for η ∈ { − 14 , … , 14 } . We are interested in the maximum correlation over all shifts. The maximum correlation ρ 𝐜 𝐯 ~ and the corresponding shift η ⋆ are defined as: ρ 𝐜 𝐯 ~ : = max η ∈ { − 14 , … , 14 } ρ 𝐜 𝐯 ~ η and η ⋆ : = argmax η ∈ { − 14 , … , 14 } ρ 𝐜 𝐯 ~ η . (4) Table 2 (columns 2 and 3) shows the values of the above quantities for each STP. A positive value of η ⋆ indicates that daily viral loads lead the cases. Table 2. Correlation with interpolated viral loads and the range of true correlation value at each STP and citywide. STP Name ρ 𝐜 𝐯 ~ η ⋆ 95% CI for ρ c v Agaram (35 MLD) 0.91 3 0.54-1.00 Bellandur (90 MLD) 0.80 4 0.34-1.00 Chikkabanavara (5 MLD) 0.92 1 0.56-1.00 Chikkabegur (5 MLD) 0.96 7 0.66-1.00 Cubbon Park (4 MLD) 0.98 5 0.70-1.00 Doddabele (60 MLD) 0.89 3 0.50-1.00 Halasuru (2 MLD) 0.87 3 0.47-1.00 Hebbal (100 MLD) 0.82 5 0.38-1.00 Hulimavu (10 MLD) 0.92 3 0.55-1.00 Jakkur (15 MLD) 0.79 7 0.33-1.00 K & C Valley (218 MLD) 0.97 2 0.67-1.00 K & C Valley (60 MLD) 0.94 -1 0.61-1.00 K R Puram Old (20 MLD) 0.97 6 0.69-1.00 Kadabeesanahalli (50 MLD) 0.94 3 0.59-1.00 Kadugodi (6 MLD) 0.93 6 0.58-1.00 Kempambudhi (1 MLD) 0.88 1 0.47-1.00 Kengeri (60 MLD) 0.99 12 0.75-0.99 Lalbagh (1.5 MLD) 0.79 0 0.32-1.00 Mailasandra (75 MLD) 0.93 3 0.58-1.00 Mallathahalli (5 MLD) 0.94 11 0.59-1.00 Nagasandra (40 MLD) 0.97 1 0.67-1.00 Rajacanal (80 MLD) and Horamavu (20 MLD) 0.82 5 0.37-1.00 Sarakki (5 MLD) 0.84 0 0.40-1.00 V Valley (330 MLD) 0.95 2 0.63-1.00 Yelahanka (10 MLD) 0.48 2 0.00-0.88 Yellamallappachetty (15 MLD) 0.88 -4 0.48-1.00 Bengaluru 0.97 3 0.67-1.00 Open in a new tab a ρ c v (resp. ρ 𝐜 𝐯 ~ ) is the correlation value between cases and viral load (resp. interpolated viral load). Validating the sample size and interpolation method The wastewater samples were collected weekly during the study period. Consequently our correlation analysis is on interpolated viral data based on 8–9 samples collected per STP. Thus, there is a need to validate the interpolation method. Ideally, one would estimate the information loss by comparing the interpolated values with the daily viral loads (ground truth). Since daily viral loads are not available for STPs in Bengaluru, we use two indirect ways to justify the interpolation method. There were six STPs in California where daily viral load measurements were made during the Omicron period ([ 42 , 43 ]). As mentioned earlier, the BBMP also reported daily cases ward wise during the Omicron period. We validate the interpolation method using both the datasets described above. We explain in detail the procedure for the STPs in California and the same will apply for the daily cases data from BBMP. For each of the STPs, we apply the following procedure: Step 1: Ground truth We select the viral load data from 11 weeks during the Omicron wave. We denote this sequence by ν . Step 2: Weekly samples From ν , we retain only those viral loads where the collection day is a Monday. The rest of the values are marked as ‘missing’. We denote this sequence by ν w . Step 3: Interpolation We apply quadratic interpolation to ν w to fill up the ‘missing’ values. Thus, we get daily interpolated values for 77 days (11 weeks). We refer to this sequence as ν ~ . Step 4: Correlation We compare the results of the interpolation method with the ground truth. In particular, we find the correlation coefficient between ν and ν ~ . The results of this exercise are shown in S2 Table . The third column shows the correlation between viral load and the interpolated values. The predominantly high values of correlation indicate the effectiveness of interpolation. We also calculate the uncertainty in the viral load estimates. The difference between the estimated viral load and the actual viral load during the baseline period will be much smaller than during the wave. Hence, instead of computing the absolute errors, we use error factors. The error factor is the ratio of the estimated (interpolated) value and the true viral load. During the baseline period, viral load values that are close to zero can lead to high error factors, which skew the distribution. Hence, we first remove the outliers before computing standard error. S4 Fig . shows the weekly mean errors. The y-axis shows the error factor. Each point represents the mean error factor for a given week, and the vertical whiskers show the standard deviation, indicating the variability of the error factor across all points within that week. We also used the above procedure outlined in Steps 1–4 (above) to accurately measure information loss due to interpolation on the BBMP cases data. For each STP in Bengaluru, we consider the case vector c as the ground truth. We retain the cases only on those days when wastewater was sampled in the corresponding STP. We denote this sequence of 8–10 data points as c w . We apply quadratic interpolation to c w to get 𝐜 ~ and then find the correlation coefficient between c and 𝐜 ~ . The results of this exercise are shown in S3 Table . Even when we pick only 8–10 data points, we see a high correlation (>90%) between the interpolated values and the daily case counts. S5 Fig . shows the error bars for case estimates. The weekly error bars in S4 Fig . and S5 Fig . are not directly used in the calculation. However, based on the error bars shown in the figures, we assume that the interpolation method introduces about 25% standard deviation error in each viral load value. This assumption is used in Step 2 below. We next quantify the uncertainties introduced by the interpolation process while calculating the correlations between cases and viral loads. We find the confidence interval of true correlation ρ c v for each STP in three steps: Step 1: Calculate ρ 𝐜 𝐯 ~ . We calculate the correlation between the case vector c and the interpolated viral load 𝐯 ~ . This value is shown in the second column of Table 2 . Step 2: Estimate ρ 𝐯 𝐯 ~ . We see that interpolation perturbs the values of the original vector by about 25% ( S4 Fig . and S5 Fig .). Hence, the question of estimating ρ 𝐯 𝐯 ~ can be re-formulated as estimating the correlation between vectors v and 𝐯 + 0.25 𝐯 ⊙ 𝐳 where z is a vector of independent and identically distributed standard normal values and ⊙ is component-wise multiplication. We find, via simulation, that the correlation is concentrated around 0.895 with a standard deviation of 0.029 ( S6 Fig .). Hence, we use [ 0.895 − 2 × 0.029 , 0.895 + 2 × 0.029 ] = [ 0.837 , 0.953 ] as the 95% confidence interval for ρ 𝐯 𝐯 ~ . Note that this interval is same for all STPs. Our estimate of ρ 𝐯 𝐯 ~ is reasonable given the correlation values for viral loads found in California STPs ( S2 Table ). Step 3: Estimate ρ c v . We bound the value of the true correlation ρ c v in terms of the quantities ρ 𝐯 𝐯 ~ and ρ 𝐜 𝐯 ~ . From the previous step, we have ρ 𝐯 𝐯 ~ ∈ [ 0.837 , 0.953 ] with 95% confidence. To bound the value of true correlation, we use the following inequality that relates pairwise correlations between three vectors proved in [ 44 ]: ρ 𝐯 𝐯 ~ ρ 𝐜 𝐯 ~ − 1 − ρ 𝐯 𝐯 ~ 2 1 − ρ 𝐜 𝐯 ~ 2 ≤ ρ c v ≤ ρ 𝐯 𝐯 ~ ρ 𝐜 𝐯 ~ + 1 − ρ 𝐯 𝐯 ~ 2 1 − ρ 𝐜 𝐯 ~ 2 The above inequality provides a range of values for ρ c v for fixed values of ρ 𝐯 𝐯 ~ and ρ 𝐜 𝐯 ~ . By using the 95% confidence interval for ρ 𝐯 𝐯 ~ , i.e., [0.837,0.953], we get a similar confidence interval for ρ c v which is shown in the last column of Table 2 . Change-point detection using Cumulative SUM (CUSUM) Does a rise in viral load at an STP predict an impending increase in cases? We used a change-point detection approach to answer this question. A change-point detector monitors sequential data for changes in the statistical properties of the data. A good detector quickly detects a change while reducing the chance of a false alarm. Our interest is in both kinds of sequential data: viral loads at an STP and confirmed cases in its catchment area. We describe the general change-point detection problem. Suppose the data points y 1 , … , y n are obtained sequentially. Suppose we are also given a pre-change (reference) distribution f 0 and a post-change distribution f 1 . We wish to determine τ , the change-point location, where y s at time s is distributed (notation ‘ ~ ’) according to f 0 or f 1 as follows: y s ~ f 0 if 0 ≤ s ≤ τ y s ~ f 1 if τ < s ≤ n . The distributions f 0 and f 1 will reflect the statistical properties of the viral loads (or cases) during the period before the wave and at the start of the wave, respectively. At the start of the wave, we do expect a trended change, but to fix ideas, assume a constant post-change distribution. There are two standard performance metrics to evaluate based on the two types of errors that can occur. Firstly, the detector can raise a false alarm when there is no change in the sequential data. This is quantified by average run length (ARL), the expected time until a change is detected when the given sequence comes only from f 0 . Secondly, there could be a delay in detecting a change. This is quantified by worst-case average detection delay (ADD) as defined by Lorden [ 45 ]. We use the cumulative sum algorithm (CUSUM) introduced by Page [ 46 ]. This is a (generalized) likelihood-ratio test that assumes that distributions f 0 and f 1 are known, but the change-point is unknown. Given the sequential data y 1 , … , y n , CUSUM computes the cumulative log-likelihood ratio considering an unknown change-point location τ . The test statistic Z n is defined as: Z n : = max 1 ≤ τ ≤ n ∑ s = τ n log f 1 ( y s ) f 0 ( y s ) for n ≥ 0 The test statistic is easy to compute via a recursive formula: Z n = ( Z n − 1 + log f 1 ( y n ) f 0 ( y n ) ) + , Z 0 = 0 The CUSUM procedure raises an alarm the first time Z n exceeds a pre-defined threshold h . The ARL of CUSUM depends on the threshold h (ARL ∝ exp ( h ) ), and once the ARL is calibrated, the CUSUM algorithm minimizes the ADD in the asymptotic sense [ 47 ]. For this reason, CUSUM is an asymptotically optimal algorithm for hypothesis testing when the distributions f 0 and f 1 are known. The CUSUM algorithm cannot be directly applied to the viral load or case data. The reason is the dependence on time for both cases and viral loads. During post-change, the cases and viral loads increase with time. We choose a different test statistic to remove this dependence on time and embed the current setting within the framework of CUSUM. We set y i = v ( i + 1 ) − v ( i ) , for each i . Note that y i denotes the rate of change of the viral load on the i th day. Although y i ’s are ‘detrended’ they are not independent. In particular, for each i , y i +1 and y i are correlated since both depend on v ( i ). To ensure independence, we use y k for all odd k as they are mutually independent. The pre-change distribution f 0 and post-change distribution f 1 of the y k s are assumed to be normally distributed with mean μ 0 and μ 1 and standard deviation σ 0 and σ 1 , respectively. We estimate the parameters of the pre-distribution f 0 from the values (cases or viral loads) between 15 Nov 2021 and 24 Dec 2021. We estimate the parameters of the post-distribution f 1 from the values between 25 Dec 2021 and 14 Jan 2022 ( S1 Table ). This period corresponds to the rise of the Omicron wave. The table shows that both the viral and case distributions have means close to zero during the period pre-change and switch to positive values in the post-change phase. Assuming the parameter values in S1 Table , we run the CUSUM algorithm to detect change points in both viral load and case data over the same period. We are interested in knowing if the CUSUM detects the change in viral loads earlier than it does with the cases. We say the viral load leads the cases if this happens. To quantify the lead time we first define a few terms. The CUSUM algorithm raises an alarm when it detects a change point, either in cases or viral loads and terminates after that. Alarm dates. Let Cases-CP (cases change point) denote the date the CUSUM algorithm raises an alarm based on case data. Let VCPD be the alarm date based on daily viral load. The lead time Lead-VCPD is the temporal difference between the alarm dates of the two sequences: Lead-VCPD = Cases-CP - VCPD A positive lead time (Lead-VCPD >0) indicates that the alarm date VCPD precedes the alarm date Cases-CP. For example, if VCPD = 2021-12-24 and Cases-CP = 2021-12-30, then: Lead-VCPD = 2021-12-30 - 2021-12-24 = 6 days, i.e., the daily viral load data provides a 6-day early warning signal. We also define an alternate notion of lead time. The earliest sample collection day after VCPD is called VCPC (change-point based on collection date). Consider an example given in Fig 3 . Suppose the samples were collected on 2nd, 9th and 16th day of the month. Hence, viral loads are available only for three days. Suppose the CUSUM algorithm detects a change-point on the 12th day of the month. In reality, no viral load is measured on this day. In practice, the actual rise would not be detected until the next collection day, i.e., the 16th of the month. We define the lead time based on VCPC as: Fig 3. Illustration of the difference between VCPD and VCPC. Open in a new tab The change-point based on interpolated daily viral load is detected on day 12. Hence, VCPD is the 12th of the month. The earliest sample collection day after detection is day 16. Hence, VCPC is 16th of the month. Lead-VCPC = Cases-CP - VCPC The viral load is said to provide an early warning if Lead-VCPC >0. Results Wards in the catchment of an STP We identified the wards within the catchment area of each STP, following the methodology given in Section Mapping STP catchments ). The details are in Table 1 and visualized in Fig 1 . The mapping allows us to attribute the cases in the catchment area to the corresponding STP. Also, this mapping may be of independent interest for the localized surveillance of other pathogens, such as Influenza and Mpox. Table 1. STPs and the list of wards in their catchment. STP-Ward mapping Agaram (35 MLD) 114-Agaram, 125-Marenahalli, 152-Suddagunte Palya, 168-Pattabhiram Nagar, 169-Byrasandra, 170-Jayanagar East, 171-Gurappanapalya, 172-Madivala, 174-HSR Layout, 175-Bommanahalli, 179-Shakambari Nagar, 187-Puttenahalli, 188-Bilekhalli, 189-Hongasandra, 192-Begur, 193-Arakere, 196-Anjanapura Bellandur (90 MLD) 85-Dodda Nekkundi, 86-Marathahalli, 149-Varthuru, 150-Bellanduru, 174-HSR Layout Chikkabanavara (5 MLD) 11-Kuvempu Nagar Chikkabegur (5 MLD) 190-Mangammanapalya, 191-Singasandra Cubbon Park (4 MLD) 93-Vasanth Nagar, 110-Sampangiram Nagar Doddabele (60 MLD) 71-Hegganahalli, 72-Herohalli, 129-Jnana Bharathi ward, 130-Ullalu, 154-Basavanagudi, 159-Kengeri, 160-Rajarajeshwari Nagar, 161-Hosakerehalli, 162-Girinagar, 164-Vidyapeeta ward, 165-Ganesh Mandir ward, 166-Karisandra, 167-Yediyur, 182-Padmanabha Nagar, 183-Chikkalsandra Halasuru (2 MLD) 59-Maruthi Seva Nagar, 79-Sarvagna Nagar Hebbal (100 MLD) 16-Jalahalli, 17-J P Park, 18-Radhakrishna Temple Ward, 19-Sanjaya Nagar, 20-Ganga Nagar, 21-Hebbala, 22-Vishwanath Nagenahalli, 31-Kushal Nagar, 33-Manorayanapalya, 34-Gangenahalli, 35-Aramane Nagara, 36-Mattikere, 37-Yeshwanthpura, 42-Lakshmi Devi Nagar, 45-Malleswaram, 46-Jayachamarajendra Nagar, 48-Muneshwara Nagar Henuur (1 MLD) 23-Nagavara Hulimavu (10 MLD) 193-Arakere, 194-Gottigere Jakkur (15 MLD) 1-Kempegowda Ward, 2-Chowdeswari Ward, 4-Yelahanka Satellite Town, 5-Jakkuru K & C Valley (150 MLD) 109-Chickpete, 110-Sampangiram Nagar, 115-Vannarpet, 116-Nilasandra, 117-Shanthi Nagar, 118-Sudham Nagara, 120-Cottonpete, 139-K R Market, 143-Vishveshwara Puram, 144-Siddapura, 145-Hombegowda Nagara, 146-Lakkasandra, 147-Adugodi, 148-Ejipura, 151-Koramangala K & C Valley (218 MLD) 58-New Tippasandara, 59-Maruthi Seva Nagar, 60-Sagayarapuram, 61-S K Garden, 62-Ramaswamy Palya, 63-Jayamahal, 78-Pulikeshinagar, 79-Sarvagna Nagar, 80-Hoysala Nagar, 87-HAL Airport, 88-Jeevanbhima Nagar, 89-Jogupalya, 90-Halsoor, 91-Bharathi Nagar, 92-Shivaji Nagar, 93-Vasanth Nagar, 112-Domlur, 113-Konena Agrahara, 150-Bellanduru K & C Valley (30 MLD) 148-Ejipura K & C Valley (60 MLD) 170-Jayanagar East, 171-Gurappanapalya, 172-Madivala, 173-Jakkasandra, 174-HSR Layout, 175-Bommanahalli, 176-BTM Layout, 177-J P Nagar, 178-Sarakki, 187-Puttenahalli, 190-Mangammanapalya, 195-Konankunte K R Puram New (20 MLD) 25-Horamavu, 26-Ramamurthy Nagar, 51-Vijnanapura K R Puram Old (20 MLD) 52-K R Puram, 53-Basavanapura Kadabeesanahalli (50 MLD) 50-Benniganahalli, 56-A Narayanapura, 57-C V Raman Nagar, 81-Vijnana Nagar, 82-Garudachar Playa, 84-Hagadur, 86-Marathahalli, 149-Varthuru Kadugodi (6 MLD) 82-Garudachar Playa, 83-Kadugodi, 84-Hagadur Kempambudhi (1 MLD) 142-Sunkenahalli, 155-Hanumanth Nagar Kengeri (60 MLD) 154-Basavanagudi, 159-Kengeri, 160-Rajarajeshwari Nagar, 161-Hosakerehalli, 162-Girinagar, 164-Vidyapeeta ward, 165-Ganesh Mandir ward, 166-Karisandra, 167-Yediyur, 182-Padmanabha Nagar, 183-Chikkalsandra Lalbagh (1.5 MLD) 143-Vishveshwara Puram, 144-Siddapura, 153-Jayanagar, 167-Yediyur Madivala (4 MLD) 176-BTM Layout, 177-J P Nagar Mailasandra (75 MLD) 160-Rajarajeshwari Nagar, 180-Banashankari Temple ward, 181-Kumaraswamy Layout, 184-Uttarahalli, 185-Yelchenahalli, 186-Jaraganahalli, 197-Vasanthpura, 198-Hemmigepura Mallathahalli (5 MLD) 72-Herohalli Nagasandra (40 MLD) 12-Shettihalli, 13-Mallasandra, 14-Bagalakunte, 15-T Dasarahalli, 38-HMT Ward, 39-Chokkasandra, 41-Peenya Industrial Area, 70-Rajagopal Nagar Rajacanal (80 MLD) and Horamavu (20 MLD) 6-Thanisandra, 7-Byatarayanapura, 8-Kodigehalli, 9-Vidyaranyapura, 10-Dodda Bommasandra, 24-HBR Layout, 27-Banasavadi, 28-Kammanahalli, 29-Kacharkanahalli, 30-Kadugondanahalli, 31-Kushal Nagar, 32-Kaval Bairasandra, 47-Devara Jeevanahalli, 48-Muneshwara Nagar, 78-Pulikeshinagar Sarakki (5 MLD) 147-Adugodi, 148-Ejipura, 151-Koramangala, 173-Jakkasandra, 178-Sarakki, 187-Puttenahalli, 195-Konankunte V Valley (330 MLD) 35-Aramane Nagara, 43-Nandini Layout, 44-Marappana Palya, 45-Malleswaram, 64-Rajamahal Guttahalli, 65-Kadu Malleshwar Ward, 66-Subramanya Nagar, 67-Nagapura, 68-Mahalakshimpuram, 69-Laggere, 73-Kottegepalya, 74-Shakthi Ganapathi Nagar, 75-Shankar Matt, 76-Gayithri Nagar, 77-Dattatreya Temple, 92-Shivaji Nagar, 94-Gandhinagar, 95-Subhash Nagar, 96-Okalipuram, 97-Dayananda Nagar, 98-Prakash Nagar, 99-Rajaji Nagar, 100-Basaveshwara Nagar, 101-Kamakshipalya, 102-Vrisabhavathi Nagar, 103-Kaveripura, 104-Govindaraja Nagar, 105-Agrahara Dasarahalli, 106-Dr. Raj Kumar Ward, 107-Shivanagara, 108-Sriramamandir, 109-Chickpete, 120-Cottonpete, 121-Binnipete, 122-Kempapura Agrahara, 123-Vijayanagar, 124-Hosahalli, 125-Marenahalli, 126-Maruthi Mandir ward, 127-Mudalapalya, 128-Nagarabhavi, 131-Nayandahalli, 132-Attiguppe, 133-Hampi Nagar, 134-Bapuji Nagar, 135-Padarayanapura, 136-Jagajivanaramnagar, 137-Rayapuram, 138-Chalavadipalya, 140-Chamrajapet, 141-Azad Nagar, 154-Basavanagudi, 155-Hanumanth Nagar, 156-Srinagar, 157-Gali Anjenaya Temple ward, 158-Deepanjali Nagar, 160-Rajarajeshwari Nagar, 163-Katriguppe Yelahanka (10 MLD) 3-Atturu Yellamallappachetty (15 MLD) 53-Basavanapura, 54-Hudi, 55-Devasandra Open in a new tab Correlation analysis The results of the correlation analysis are summarized in Table 2 . For each STP, we have the maximum correlation value of cases and daily viral loads and the corresponding shift η ⋆ . The median shift is three days. At first glance, it suggests that viral load leads the cases by three days. The observed three-day lead in the interpolated daily viral load data is an artifact of the interpolation process, rather than a true reflection of the relationship between viral loads and case numbers. Since viral load measurements are taken weekly, interpolating this data to a daily frequency introduces a lead that may not accurately represent the actual timing between viral loads and cases. Therefore, this lead should not be considered significant. A lead of seven or more days, however, would be more meaningful, as it aligns with the weekly sampling schedule and could provide useful information for early warning. Change-point analysis Correlation being a retrospective analysis, we adapted a classical change-point analysis on the viral load and cases time series to see if there was an early warning. Table 3 shows the alarm dates for cases and viral loads across all STPs. Leads are measured based on VCPD and VCPC dates (with threshold h = 20). We consider the lead time based on collection dates more accurate, so the entries are sorted according to VCPC lead times. As an analogous indication of statistical power, we vary the threshold h from 10 to 20 and tabulate the mean and standard deviations for lead times ( Table 4 ). Table 3. Comparison of change-points from cases and viral loads. STP name Cases CP VCPD VCPC Lead-VCPD Lead-VCPC Jakkur (15 MLD) 2021-12-30 2021-12-24 2021-12-24 6 6 Hebbal (100 MLD) 2021-12-28 2021-12-24 2021-12-24 4 4 Rajacanal (80 MLD) & Horamavu (20 MLD) 2021-12-26 2021-12-24 2021-12-24 2 2 Yelahanka (10 MLD) 2021-12-26 2021-12-24 2021-12-24 2 2 Halasuru (2 MLD) 2021-12-30 2021-12-24 2021-12-29 6 1 Agaram (35 MLD) 2021-12-29 2021-12-27 2021-12-29 2 0 K R Puram Old (20 MLD) 2022-01-05 2021-12-30 2022-01-05 6 0 Yellamallappachetty (15 MLD) 2021-12-28 2021-12-28 2021-12-28 0 0 Cubbon Park (4 MLD) 2021-12-28 2021-12-28 2021-12-29 0 -1 Lalbagh (1.5 MLD) 2021-12-28 2021-12-24 2021-12-29 4 -1 Sarakki (5 MLD) 2022-01-03 2021-12-24 2022-01-04 10 -1 Chikkabanavara (5 MLD) 2021-12-24 2021-12-26 2021-12-27 -2 -3 Chikkabegur (5 MLD) 2022-01-03 2021-12-24 2022-01-06 10 -3 Mallathahalli (5 MLD) 2022-01-03 2021-12-24 2022-01-06 10 -3 Nagasandra (40 MLD) 2022-01-01 2021-12-26 2022-01-04 6 -3 Kadabeesanahalli (50 MLD) 2021-12-30 2021-12-28 2022-01-03 2 -4 Doddabele (60 MLD) 2022-01-01 2022-01-05 2022-01-06 -4 -5 Kengeri (60 MLD) 2022-01-01 2021-12-24 2022-01-06 8 -5 K & C Valley (218 MLD) 2021-12-28 2021-12-28 2022-01-03 0 -6 Kadugodi (6 MLD) 2021-12-30 2022-01-05 2022-01-05 -6 -6 Bellandur (90 MLD) 2021-12-28 2021-12-24 2022-01-04 4 -7 Mailasandra (75 MLD) 2021-12-28 2022-01-05 2022-01-06 -8 -9 V Valley (330 MLD) 2021-12-28 2021-12-24 2022-01-06 4 -9 Kempambudhi (1 MLD) 2021-12-28 2022-01-07 2022-01-07 -10 -10 Hulimavu (10 MLD) 2021-12-28 2022-01-05 2022-01-11 -8 -14 K & C Valley (60 MLD) 2021-12-24 2022-01-13 2022-01-17 -20 -24 Bengaluru 2021-12-29 2021-12-29 NA 0 NA Open in a new tab a Description of the columns. Cases CP: Change-point based on cases, VCPD: Change-point based on interpolated daily viral load, VCPC: Change-point based on sample collection date. Lead is measured in days. b The alarm from cases at K & C Valley (60 MLD) is a false alarm caused by a blip in the reported cases. We notice that cases remain flat after the alarm by manual inspection. c The alarms from cases for Lalbagh and Yelahanka are also false alarms. The two STPs have a negative post-change rate (see footnote of S1 Table ). d CUSUM is run with threshold h = 20. Table 4. Mean and standard deviation of lead times (in days) for VCPD and VCPC at various STPs. The distribution is over threshold h ranging from 10 to 20. STP Name Lead-VCPD Lead-VCPC Yelahanka (10 MLD) 10.55 (10.72) 7.36 (10.01) Sarakki (5 MLD) 10.36 (2.80) 4.36 (6.50) Halasuru (2 MLD) 6.73 (2.30) 1.55 (3.73) Hebbal (100 MLD) 3.27 (1.54) 1.09 (2.87) Mallathahalli (5 MLD) 6.91 (11.55) -0.45 (13.32) Jakkur (15 MLD) -1.45 (7.09) -2.73 (7.88) Kadugodi (6 MLD) -4.36 (5.77) -4.36 (5.77) Agaram (35 MLD) -1.09 (4.62) -4.55 (6.04) Yellamallappachetty (15 MLD) -3.09 (3.75) -5.36 (4.94) Rajacanal (80 MLD) and Horamavu (20 MLD) -5.27 (5.34) -6.00 (6.21) Nagasandra (40 MLD) 0.91 (6.29) -6.82 (4.91) Kadabeesanahalli (50 MLD) -3.09 (6.84) -7.40 (3.44) Bellandur (90 MLD) 2.00 (6.55) -7.55 (9.18) Cubbon Park (4 MLD) -1.64 (2.38) -8.09 (4.03) K R Puram Old (20 MLD) -4.55 (11.63) -8.60 (10.36) Doddabele (60 MLD) -7.64 (4.25) -8.64 (4.25) K & C Valley (218 MLD) -3.82 (3.46) -9.82 (3.46) Chikkabanavara (5 MLD) -8.55 (6.10) -10.64 (6.97) Lalbagh (1.5 MLD) -5.09 (5.42) -10.82 (5.94) Chikkabegur (5 MLD) 1.45 (8.09) -11.55 (8.09) Kempambudhi (1 MLD) -11.64 (5.71) -12.09 (5.33) Hulimavu (10 MLD) -10.55 (5.12) -15.18 (7.00) Mailasandra (75 MLD) -12.36 (5.17) -15.18 (6.35) Kengeri (60 MLD) -5.82 (7.31) -17.00 (7.07) V Valley (330 MLD) -4.55 (6.33) -17.55 (6.33) K & C Valley (60 MLD) -21.56 (2.95) -24.29 (0.70) Open in a new tab Discussion We use the well-known standard correlation and change-point analysis at the STP level (as opposed to citywide aggregate) due to the mapping of wards to the respective STPs using a Geographic Information System (GIS) tool kit, daily cases data from BBMP, and detailed drainage maps from BWSSB ( Fig. 1 ). From the data we estimated the catchment area associated to each STP, providing a basis for correlating the ward cases with the respective STP viral load. Apart from 3-4 STPs, during the wave, there is high (>0.80) correlation between the observed viral load and the cases observed in the respective catchment area ( Table 2 ). The correlation in Yelahanka is low (0.48) due to a low viral load value between the initial rising phase and the peak; the correlation in Lalbagh and Bellandur is moderate ( 0.79 − 0.80 ) again due to low viral loads but post peak; we do not know the reasons for these low values. The correlation in Jakkur being moderate (0.79) can be attributed to missing data values post the wave’s peak. In our 77 day window, we had 8–9 data points per STP. We then used interpolation to derive daily samples from weekly wastewater data to determine the correlation with the daily cases. We notice a high correlation between the interpolated and actual values of viral load during during the Omicron wave (when there is a well defined peak along with a clear rise and fall in viral loads). However we observe a dip in correlation between actual and interpolated values during baseline period with stable but low viral loads. This is also observed on viral load data from STPs in California, where over the three month period from Nov 2022 to Dec 2023 the method yielded correlations of around 20–30% but with much higher correlations during the Omicron wave ( S2 Table ). We have also addressed the impact of having 8–9 data points and have quantified the interpolation error and uncertainty in correlation obtained in the Section Validating the sample size and interpolation method. The interpolation is needed as the data is available once in seven days and we use a quadratic-interpolation to follow the rising or falling trends closely without introducing a lead bias. To detect a change-point in distribution, we have employed the CUSUM method ( Fig 4 ). CUSUM is asymptotically optimal as long as the data is independent. We have used alternate slopes as our data sequence instead of consecutive values to ensure independence, and this modeling technique provides an advantage over other approaches, for, e.g., [ 12 ]. The change-points for the two sequences differ by at most one day. However working with differences introduces a limitation that there are fewer samples. False alarm rate and power are the parameters of interest for a fixed duration hypothesis testing problem. We have here a sequential hypothesis testing problem. False alarm rate's analogue is the average run length to false alarm (ARL). The analogue of power in the sequential setting is average detection delay (ADD). For a given ARL, the sequential CUSUM approach provides the advantage of accumulating evidence over time to have lower ADD (compared to other approaches that do not accumulate evidence). Just as the power for various false alarm rates, obtained by varying the thresholds, gives the receiver operating characteristics, Table 4 provides the relative ADD for various ARLs obtained by varying the threshold. Fig 4. The output of CUSUM on (composite of) Rajacanal and Horamavu STPs. Open in a new tab Viral loads and cases increased over time during the post-change period. The blue dotted line shows the change-point for cases and the green dotted line shows the change-point for viral loads (VCPC). CUSUM raised an alarm for viral loads two days before it raised an alarm for cases. No robust lead times were observed from viral load data over cases both at the STP level and citywide ( Table 3 and Fig 5 ). This could be due to many reasons. The surveillance systems were already on alert for Omicron and there was good detection of cases due to the availability of RT-PCR testing. The sampling of wastewater was done once a week [ 12 ]. This infrequent sampling of wastewater from the STPs will miss viral load changes in small time windows and it is possible that early warnings are missed. Lead is also influenced by the preparedness of the public health response, by new variants, and changes in viral shedding [ 26 , 27 ]. Lamba et al. [ 12 ] conducted a citywide correlation analysis for a different period from January to June 2022 and found a lead of 7 days. However, this may be a due to limited testing among other factors, see [ 12 ]. Fig 5. A panel showing the rise of viral loads and cases citywide (Bengaluru) and at 26 STPs. Open in a new tab The alarm dates for each STP are indicated with dotted lines (from Table 3 ). The blue and the green dotted lines indicate Cases-CP and Lead-VCPC, respectively. The x -axis is shared across the plots. During the study period, there were no mobility restrictions and other non-pharmaceutical interventions (such as wearing masks and quarantine measures) were recommended but not mandatory. We further assumed that there is a negligible floating population and sufficient homogenization. Under these assumptions an increase in viral load in an STP will correspond to an increase in cases in the corresponding catchment area. Our results are based only on this premise. A unique feature of our work is the mapping of catchment area of each STP in Bengaluru. This helped correlate daily cases at the ward level with viral loads at respective STPs. We were not able to gather such granular data for daily cases in catchment areas along with corresponding STP viral loads in other studies. For example, cases at the catchments are not reported for California’s STPs. Thus, we were limited in providing a comparative analysis. Coverage of STPs in Bengaluru is extensive, capturing wastewater from a large proportion of the population. However, some portion of wastewater remains untreated and is discharged into open drains. The study by [ 48 ] examined viral loads in such open drains, providing complementary insights. Large sewage treatment plants (STPs), like K&C Valley, cover a wide area and mix wastewater from many neighborhoods, making it harder to detect localized outbreaks. This reduces the effectiveness of wastewater surveillance for specific areas. To improve this, we suggest using sentinel manholes in key neighborhoods or high-density areas. Sampling from these manholes would allow for better detection of local viral trends. We compute the citywide viral load by weighted average of viral loads of STPs, weighted by capacities of STPs. A more accurate method to compute the weighted viral load would be to use daily sewage flow data instead of capacities. However, reliable flow data from BWSSB is not available during the Omicron period. Our manual mapping of wards to STPs, based on inspection of the sewerage network and expert consultation, does not easily scale to other cities. It may be possible to automatically map regions to STPs using the sewerage network and the city’s topography. Given a detailed network file, such as a GIS file, and elevation data, flow directions can be traced by following pipe connections and slopes, since wastewater flows downhill. This would allow automated assignment of areas, such as wards, to their respective STPs. While practical challenges like missing data, pumping stations, or undocumented connections may affect accuracy, in principle, accurate and updated network and topographic data would enable systematic and reliable mapping. This approach can be explored as future work to improve the precision and scalability of wastewater surveillance efforts. In summary, as the viral data was limited to 8–9 samples per STP, we have used interpolation to get the daily viral load estimates. We have quantified the consequent uncertainty in correlation with daily cases and provided error bounds. Even with these limitations our analysis during the Omicron wave in Bengaluru suggests that an increase in wastewater viral loads correlates with increase in COVID-19 cases. We also observed this correlation at individual STP levels. Consequently, we conclude that in the absence of testing, continuous wastewater surveillance can help in the monitoring of SARS-CoV2 circulation. However, wastewater surveillance did not give a reliable early warning during the Omicron wave. The frequency of sampling may be increased to daily sampling during periods of infection to enable timely interventions. A more detailed mapping of the sewersheds, e.g., via satellite imaging, can enhance the surveillance area. Supporting information S1 Fig. Drainage pipes from wards leading to the Hebbal STP. (PDF) pgph.0004684.s001.pdf (440.2KB, pdf) S2 Fig. BWSSB sewer lines coverage. (PDF) pgph.0004684.s002.pdf (534.6KB, pdf) S3 Fig. A scatter plot of viral loads and cases from four STPs. (PDF) pgph.0004684.s003.pdf (71.5KB, pdf) S4 Fig. Error bars for viral load estimates at STPs in California. (PDF) pgph.0004684.s004.pdf (30.3KB, pdf) S5 Fig. Error bars for cases estimates at STP catchments in Bengaluru. (PDF) pgph.0004684.s005.pdf (31.4KB, pdf) S6 Fig. Distribution of correlation values between a vector and its noisy version. (PDF) pgph.0004684.s006.pdf (37.4KB, pdf) S1 Table. Parameter values of the pre-and-post distributions of alternate slopes. (PDF) pgph.0004684.s007.pdf (78.4KB, pdf) S2 Table. Correlation between daily viral loads and values interpolated from weekly samples in California. (PDF) pgph.0004684.s008.pdf (50.1KB, pdf) S3 Table. Correlation between daily cases and interpolated cases from weekly samples in Bengaluru. (PDF) pgph.0004684.s009.pdf (44.5KB, pdf) S4 Table. List of files related to datasets, code, and results. (PDF) pgph.0004684.s010.pdf (61.2KB, pdf) S1 Text. An example of computing weighted viral load. (PDF) pgph.0004684.s011.pdf (60.9KB, pdf) S2 Text. A summary of our work in Kannada. (PDF) pgph.0004684.s012.pdf (39.8KB, pdf) Acknowledgments We gratefully acknowledge the support received from BWSSB. We thank Dr. Bhaskar Rajakumar for helping us coordinate with BBMP. We gratefully acknowledge the contribution of Jagadish Midthala towards the conceptualization, data curation, investigation, methodology, software, and in the preparation of the manuscript. Data Availability Data is available at http://dataportal.icts.res.in/wastewater/ . Funding Statement The research of SA is partially supported by the Knowledge Exchange Grant from ICTS-TIFR under project no. RTI4001 of the Department of Atomic Energy, Government of India. The research of RS and TK is partially supported by an IOE grant, CISCO-IISc Centre for Networked Intelligence ( https://cni.iisc.ac.in/ ), and the Isaac Centre for Public Health at the Indian Institute of Science. Wastewater surveillance of SARS-CoV-2 was supported by funding from the Rockefeller Foundation (grant 2021 HTH 018) and the Indian Council of Medical Research grant to (FI) the Tata Institute for Genetics and Society and Tata Trusts. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. References 1. WHO. Status of environmental surveillance for SARS-CoV-2 virus. 2020. https://www.who.int/news-room/commentaries/detail/status-of-environmental-surveillance-for-sars-cov-2-virus 2. CDC. National Wastewater Surveillance System. https://www.cdc.gov/nwss/rv/COVID19-nationaltrend.html 2023. 3. Wolfe MK, Topol A, Knudson A, Simpson A, White B, Vugia DJ, et al. High-Frequency, High-Throughput Quantification of SARS-CoV-2 RNA in Wastewater Settled Solids at Eight Publicly Owned Treatment Works in Northern California Shows Strong Association with COVID-19 Incidence. mSystems. 2021;6(5):e0082921. doi: 10.1128/mSystems.00829-21 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Medema G, Heijnen L, Elsinga G, Italiaander R, Brouwer A. Presence of SARS-Coronavirus-2 RNA in Sewage and Correlation with Reported COVID-19 Prevalence in the Early Stage of the Epidemic in The Netherlands. Environ Sci Technol Lett. 2020;7(7):511–6. doi: 10.1021/acs.estlett.0c00357 [ DOI ] [ PubMed ] [ Google Scholar ] 5. Ahmed W, Angel N, Edson J, Bibby K, Bivins A, O’Brien JW, et al. First confirmed detection of SARS-CoV-2 in untreated wastewater in Australia: A proof of concept for the wastewater surveillance of COVID-19 in the community. Sci Total Environ. 2020;728:138764. doi: 10.1016/j.scitotenv.2020.138764 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Kumar M, Jiang G, Kumar Thakur A, Chatterjee S, Bhattacharya T, Mohapatra S, et al. Lead time of early warning by wastewater surveillance for COVID-19: Geographical variations and impacting factors. Chem Eng J. 2022;441:135936. doi: 10.1016/j.cej.2022.135936 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Shah S, Gwee SXW, Ng JQX, Lau N, Koh J, Pang J. Wastewater surveillance to infer COVID-19 transmission: A systematic review. Sci Total Environ. 2022;804:150060. doi: 10.1016/j.scitotenv.2021.150060 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Wu F, Xiao A, Zhang J, Moniz K, Endo N, Armas F, et al. SARS-CoV-2 RNA concentrations in wastewater foreshadow dynamics and clinical presentation of new COVID-19 cases. Sci Total Environ. 2022;805:150121. doi: 10.1016/j.scitotenv.2021.150121 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Wurtzer S, Marechal V, Mouchel J, Maday Y, Teyssou R, Richard E. Evaluation of lockdown effect on SARS-CoV-2 dynamics through viral genome quantification in wastewater, Greater Paris, France, 5 March to 23 April 2020. Eurosurveillance. 2020;25(50). doi: 10.2807/1560-7917.ES.2020.25.50.2000776 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Naughton CC, Roman FA Jr, Alvarado AGF, Tariqi AQ, Deeming MA, Kadonsky KF, et al. Show us the data: global COVID-19 wastewater monitoring efforts, equity, and gaps. FEMS Microbes. 2023;4:xtad003. doi: 10.1093/femsmc/xtad003 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Wadi VS, Daou M, Zayed N, AlJabri M, Alsheraifi HH, Aldhaheri SS, et al. Long-term study on wastewater SARS-CoV-2 surveillance across United Arab Emirates. Sci Total Environ. 2023;887:163785. doi: 10.1016/j.scitotenv.2023.163785 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Lamba S, Ganesan S, Daroch N, Paul K, Joshi SG, Sreenivas D, et al. SARS-CoV-2 infection dynamics and genomic surveillance to detect variants in wastewater - a longitudinal study in Bengaluru, India. Lancet Reg Health Southeast Asia. 2023;11:100151. doi: 10.1016/j.lansea.2023.100151 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Daniel RF, Kannan SK, Daroch N, Ganesan S, Mozaffer F, Srikantaiah V, et al. Identifying bellwether sewershed sites for sustainable disease surveillance in Bengaluru, India: a longitudinal study. Lancet Reg Health Southeast Asia. 2025;39:100619. doi: 10.1016/j.lansea.2025.100619 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Kumar M, Joshi M, Jiang G, Yamada R, Honda R, Srivastava V, et al. Response of wastewater-based epidemiology predictor for the second wave of COVID-19 in Ahmedabad, India: A long-term data Perspective. Environ Pollut. 2023;337:122471. doi: 10.1016/j.envpol.2023.122471 [ DOI ] [ PubMed ] [ Google Scholar ] 15. Rogawski McQuade ET, Blake IM, Brennhofer SA, Islam MO, Sony SSS, Rahman T, et al. Real-time sewage surveillance for SARS-CoV-2 in Dhaka, Bangladesh versus clinical COVID-19 surveillance: a longitudinal environmental surveillance study (December, 2019-December, 2021). Lancet Microbe. 2023;4(6):e442–51. doi: 10.1016/S2666-5247(23)00010-1 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Ngwira LG, Sharma B, Shrestha KB, Dahal S, Tuladhar R, Manthalu G, et al. Cost of wastewater-based environmental surveillance for SARS-CoV-2: Evidence from pilot sites in Blantyre, Malawi and Kathmandu, Nepal. PLOS Glob Public Health. 2022;2(12):e0001377. doi: 10.1371/journal.pgph.0001377 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Wong YHM, Lim JT, Griffiths J, Lee B, Maliki D, Thompson J, et al. Positive association of SARS-CoV-2 RNA concentrations in wastewater and reported COVID-19 cases in Singapore - A study across three populations. Sci Total Environ. 2023;902:166446. doi: 10.1016/j.scitotenv.2023.166446 [ DOI ] [ PubMed ] [ Google Scholar ] 18. Swift CL, Isanovic M, Correa Velez KE, Norman RS. SARS-CoV-2 concentration in wastewater consistently predicts trends in COVID-19 case counts by at least two days across multiple WWTP scales. Environ Adv. 2023;11:100347. doi: 10.1016/j.envadv.2023.100347 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Olesen SW, Imakaev M, Duvallet C. Making waves: Defining the lead time of wastewater-based epidemiology for COVID-19. Water Res. 2021;202:117433. doi: 10.1016/j.watres.2021.117433 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. D’Aoust PM, Graber TE, Mercier E, Montpetit D, Alexandrov I, Neault N, et al. Catching a resurgence: Increase in SARS-CoV-2 viral RNA identified in wastewater 48h before COVID-19 clinical tests and 96h before hospitalizations. Science of The Total Environment. 2021;770:145319. doi: 10.1016/j.scitotenv.2021.145319 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Kumar M, Joshi M, Patel AK, Joshi CG. Unravelling the early warning capability of wastewater surveillance for COVID-19: A temporal study on SARS-CoV-2 RNA detection and need for the escalation. Environ Res. 2021;196:110946. doi: 10.1016/j.envres.2021.110946 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Larsen DA, Collins MB, Du Q, Hill D, Insaf TZ, Kilaru P. Coupling freedom from disease principles and early warning from wastewater surveillance to improve health security. medRxiv. 2021. doi: 10.1101/2021.06.11.21258797 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Nemudryi A, Nemudraia A, Wiegand T, Surya K, Buyukyoruk M, Cicha C, et al. Temporal Detection and Phylogenetic Assessment of SARS-CoV-2 in Municipal Wastewater. Cell Rep Med. 2020;1(6):100098. doi: 10.1016/j.xcrm.2020.100098 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Peccia J, Zulli A, Brackney DE, Grubaugh ND, Kaplan EH, Casanovas-Massana A, et al. Measurement of SARS-CoV-2 RNA in wastewater tracks community infection dynamics. Nat Biotechnol. 2020;38(10):1164–7. doi: 10.1038/s41587-020-0684-z [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Feng S, Roguet A, McClary-Gutierrez JS, Newton RJ, Kloczko N, Meiman JG, et al. Evaluation of Sampling, Analysis, and Normalization Methods for SARS-CoV-2 Concentrations in Wastewater to Assess COVID-19 Burdens in Wisconsin Communities. ACS EST Water. 2021;1(8):1955–65. doi: 10.1021/acsestwater.1c00160 [ DOI ] [ Google Scholar ] 26. Li X, Zhang S, Sherchan S, Orive G, Lertxundi U, Haramoto E, et al. Correlation between SARS-CoV-2 RNA concentration in wastewater and COVID-19 cases in community: A systematic review and meta-analysis. J Hazard Mater. 2023;441:129848. doi: 10.1016/j.jhazmat.2022.129848 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Rusiñol M, Zammit I, Itarte M, Forés E, Martínez-Puchol S, Girones R, et al. Monitoring waves of the COVID-19 pandemic: Inferences from WWTPs of different sizes. Sci Total Environ. 2021;787:147463. doi: 10.1016/j.scitotenv.2021.147463 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Dejus B, Cacivkins P, Gudra D, Dejus S, Ustinova M, Roga A, et al. Wastewater-based prediction of COVID-19 cases using a random forest algorithm with strain prevalence data: A case study of five municipalities in Latvia. Sci Total Environ. 2023;891:164519. doi: 10.1016/j.scitotenv.2023.164519 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Wu Y, Guo Z, Yuan J, Cao G, Wang Y, Gao P, et al. Duration of viable virus shedding and polymerase chain reaction positivity of the SARS-CoV-2 Omicron variant in the upper respiratory tract: a systematic review and meta-analysis. Int J Infect Dis. 2023;129:228–35. doi: 10.1016/j.ijid.2023.02.011 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Cheung KS, Hung IFN, Chan PPY, Lung KC, Tso E, Liu R, et al. Gastrointestinal Manifestations of SARS-CoV-2 Infection and Virus Load in Fecal Samples From a Hong Kong Cohort: Systematic Review and Meta-analysis. Gastroenterology. 2020;159(1):81–95. doi: 10.1053/j.gastro.2020.03.065 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Parasa S, Desai M, Thoguluva Chandrasekar V, Patel HK, Kennedy KF, Roesch T, et al. Prevalence of Gastrointestinal Symptoms and Fecal Viral Shedding in Patients With Coronavirus Disease 2019. JAMA Netw Open. 2020;3(6):e2011335. doi: 10.1001/jamanetworkopen.2020.11335 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Jones DL, Baluja MQ, Graham DW, Corbishley A, McDonald JE, Malham SK, et al. Shedding of SARS-CoV-2 in feces and urine and its potential role in person-to-person transmission and the environment-based spread of COVID-19. Sci Total Environ. 2020;749:141364. doi: 10.1016/j.scitotenv.2020.141364 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Miura F, Kitajima M, Omori R. Duration of SARS-CoV-2 viral shedding in faeces as a parameter for wastewater-based epidemiology: Re-analysis of patient data using a shedding dynamics model. Sci Total Environ. 2021;769:144549. doi: 10.1016/j.scitotenv.2020.144549 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Directorate of Economics and Statistics. Projected Population of Karnataka 2012-2021. 2020. https://ahara.karnataka.gov.in/foodanalytics/Home/Projected , Accessed 2026 April 03. 35. OpenCity. BBMP Ward Information. https://data.opencity.in/dataset/bbmp-ward-information 2024. 2024 November 14. 36. OpenCity. BBMP Ward Map KML file. https://data.opencity.in/dataset/87b978d1-352e-4b90-aa2c-9991e55d3425/resource/a0329df6-2924-43f4-8fe4-7a6ffcc1d53d/download/806d6b9c-e8d9-4eb0-a3a3-b2ba68ec3cda.kml 2024. 2026 April 03. 37. OpenCity. BWSSB Sewage Treatment Plants (STP) Map; 2024. https://data.opencity.in/dataset/bwssb-sewage-treatment-plant-locations/resource/028ca4bd-7bb7-46bf-8f88-6757b80d9b24 Accessed: 2026 April 03. 38. OpenCity. BWSSB Sewage Treatment Plants (STP) Map - KML File. https://data.opencity.in/dataset/de97d604-7791-43d1-8572-e460cd520ba0/resource/028ca4bd-7bb7-46bf-8f88-6757b80d9b24/download/7aa3272e-287d-4955-ab19-7f0819116509.kml 2024. 2026 April 03. 39. Ramachandra T, Vinay S, Mahapatra D, Varghese S, Aithal B. Water Situation in Bengaluru. Environmental Information System. 2016. [ Google Scholar ] 40. Fitzgerald SF, Rossi G, Low AS, McAteer SP, O’Keefe B, Findlay D, et al. Site Specific Relationships between COVID-19 Cases and SARS-CoV-2 Viral Load in Wastewater Treatment Plant Influent. Environ Sci Technol. 2021;55(22):15276–86. doi: 10.1021/acs.est.1c05029 [ DOI ] [ PubMed ] [ Google Scholar ] 41. Rabe A, Ravuri S, Burnor E, Steele JA, Kantor RS, Choi S, et al. Correlation between wastewater and COVID-19 case incidence rates in major California sewersheds across three variant periods. J Water Health. 2023;21(9):1303–17. doi: 10.2166/wh.2023.173 [ DOI ] [ PubMed ] [ Google Scholar ] 42. Boehm AB, Wolfe MK, Wigginton KR, Bidwell A, White BJ, Hughes B, et al. Human viral nucleic acids concentrations in wastewater solids from Central and Coastal California USA. Sci Data. 2023;10(1):396. doi: 10.1038/s41597-023-02297-7 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Boehm AB, Wolfe MK, Bidwell AL, Zulli A, Chan-Herur V, White BJ, et al. Human pathogen nucleic acids in wastewater solids from 191 wastewater treatment plants in the United States. Sci Data. 2024;11(1):1141. doi: 10.1038/s41597-024-03969-8 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Vos P. Pearson’s correlation between three variables; using students’ basic knowledge of geometry for an exercise in mathematical statistics. International Journal of Mathematical Education in Science and Technology. 2009;40(4):533–41. doi: 10.1080/00207390802419578 [ DOI ] [ Google Scholar ] 45. Lorden G. Procedures for Reacting to a Change in Distribution. Ann Math Statist. 1971;42(6):1897–908. doi: 10.1214/aoms/1177693055 [ DOI ] [ Google Scholar ] 46. Page ES. Continuous Inspection Schemes. Biometrika. 1954;41(1/2):100. doi: 10.2307/2333009 [ DOI ] [ Google Scholar ] 47. Moustakides GV. Optimal Stopping Times for Detecting Changes in Distributions. Ann Statist. 1986;14(4). doi: 10.1214/aos/1176350164 [ DOI ] [ Google Scholar ] 48. Basu P, Choudhury S, Shridhar V, Huilgol P, Roychoudhury S, Nandi I, et al. Surveillance of SARS-CoV-2 RNA in open-water sewage canals contaminated with untreated wastewater in resource-constrained regions. Access Microbiol. 2022;4(1):000318. doi: 10.1099/acmi.0.000318 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] PLOS Glob Public Health. doi: 10.1371/journal.pgph.0004684.r001 Decision Letter 0 Mohan Amarasiri Mohan Amarasiri Academic Editor Find articles by Mohan Amarasiri Author information Copyright and License information Roles Mohan Amarasiri : Academic Editor © 2026 Mohan AmarasiriMohan AmarasiriMohan AmarasiriMohan Amarasiri This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice 10 Oct 2024 PGPH-D-24-02064 Localized wastewater surveillance showed correlation but no early warning during Bengaluru's Omicron wave PLOS Global Public Health Dear Dr. Midthala, Thank you for submitting your manuscript to PLOS Global Public Health. After careful consideration, we feel that it has merit but does not fully meet PLOS Global Public Health’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. The reviewers have raised several important questions related to the limited sample number and the generalization of the results. Please address the points raised by the reviewers carefully and resubmit the modified manuscript. Please submit your revised manuscript by Nov 24 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at [email protected] . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pgph/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. We look forward to receiving your revised manuscript. Kind regards, Mohan Amarasiri Academic Editor PLOS Global Public Health Journal Requirements: 1. We ask that a manuscript source file is provided at Revision. Please upload your manuscript file as a .doc, .docx, .rtf or .tex. 2. Figure 1, 2, 3 and 4: please (a) provide a direct link to the base layer of the map (i.e., the country or region border shape) and ensure this is also included in the figure legend; and (b) provide a link to the terms of use / license information for the base layer image or shapefile. We cannot publish proprietary or copyrighted maps (e.g. Google Maps, Mapquest) and the terms of use for your map base layer must be compatible with our CC-BY 4.0 license. Note: if you created the map in a software program like R or ArcGIS, please locate and indicate the source of the basemap shapefile onto which data has been plotted. If your map was obtained from a copyrighted source please amend the figure so that the base map used is from an openly available source. Alternatively, please provide explicit written permission from the copyright holder granting you the right to publish the material under our CC-BY 4.0 license. Please note that the following CC BY licenses are compatible with PLOS license: CC BY 4.0, CC BY 2.0 and CC BY 3.0, meanwhile such licenses as CC BY-ND 3.0 and others are not compatible due to additional restrictions. If you are unsure whether you can use a map or not, please do reach out and we will be able to help you. The following websites are good examples of where you can source open access or public domain maps: * U.S. Geological Survey (USGS) - All maps are in the public domain. ( http://www.usgs.gov ) * PlaniGlobe - All maps are published under a Creative Commons license so please cite “PlaniGlobe, http://www.planiglobe.com , CC BY 2.0” in the image credit after the caption. ( http://www.planiglobe.com/?lang=enl ) * Natural Earth - All maps are public domain. ( http://www.naturalearthdata.com/about/terms-of-use/ ) Additional Editor Comments (if provided): [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Does this manuscript meet PLOS Global Public Health’s publication criteria ? Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe methodologically and ethically rigorous research with conclusions that are appropriately drawn based on the data presented.? Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe methodologically and ethically rigorous research with conclusions that are appropriately drawn based on the data presented.-->?> Reviewer #1: Partly Reviewer #2: Partly ********** 2. Has the statistical analysis been performed appropriately and rigorously?-->?> Reviewer #1: No Reviewer #2: Yes ********** 3. Have the authors made all data underlying the findings in their manuscript fully available (please refer to the Data Availability Statement at the start of the manuscript PDF file)??> The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--> Reviewer #1: Yes Reviewer #2: Yes ********** 4. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ********** Reviewer #1: This manuscript presents a study analyzing correlations between SARS-CoV-2 viral loads in wastewater and COVID-19 case numbers in Bengaluru, India, during the Omicron wave. The authors examine data at both citywide and localized levels across 26 sewage treatment plants (STPs) from December 15, 2021, to January 15, 2022. The study addresses an important topic and offers some interesting insights, but there are several limitations that need to be addressed to strengthen the manuscript. One major concern is the limited number of viral load data points available. With weekly sampling over a one-month period, there are only 4-5 data points per STP. This small sample size may impact the reliability of the correlation analyses that are central to the study. The authors should consider providing a power analysis to assess if this sample size is sufficient for drawing robust conclusions. If not, the findings and claims may need to be qualified accordingly. The authors should discuss the implications of having few data points, including the potential impact on the uncertainty of correlation estimates, sensitivity to outliers, and statistical power. Limitations related to sample size should be addressed in the discussion. Another issue is the use of quadratic interpolation to estimate daily viral loads from the weekly samples. The reliability of this interpolation approach should be further validated, and uncertainties related to the interpolated values could be quantified. The authors may consider a sensitivity analysis using the raw weekly data to evaluate the impact of interpolation on the results. Given the limited data, some conclusions drawn from the correlation analyses may need to be tempered. The authors should consider focusing conclusions on this specific case study and avoid broad generalizations. Claims could be qualified by emphasizing the uncertainties arising from data limitations. The one-month study period may also constrain the generalizability of the findings. The authors could contextualize how focusing on this specific window within the Omicron wave may impact results compared to examining viral load and case correlations over longer timeframes. If feasible, expanding the analysis to cover multiple COVID waves could strengthen the study. Additionally, the manuscript could benefit from a discussion of potential confounding factors that may influence viral load and case correlations, such as changes in control measures and human mobility patterns over time. Comparing findings to studies of SARS-CoV-2 wastewater surveillance from other regions could provide useful context, even if a full comparative analysis is beyond the current scope. In summary, this study contributes to the growing body of research on wastewater surveillance for COVID-19. However, the conclusions that can be drawn are limited by the small number of data points. Revisions addressing and discussing the implications of these limitations could strengthen the robustness of the findings. I recommend that the authors consider the suggestions above to improve the manuscript. With appropriate revisions, this study could provide a valuable insight into the relationship between wastewater viral loads and COVID-19 cases during the Omicron wave in Bengaluru, while acknowledging the constraints posed by the available data. Reviewer #2: The authors wrote about the localized wastewater surveillance and their application in early warning for SARS-CoV-2. The manuscript needs to have more discussion but overall, the model is interesting. Abstract Please be aware that SARS-CoV-2 is the name of the virus. Covid 19 is the name of the disease. Check the whole document and fix if there are mistakes. Introduction Line 3: poliovirus, not Poliovirus. Line 4: antimicrobial resistance gene? Bacteria? Line 5: Not all virus is an RNA virus. Line 21: … source tracking. What source tracking? Line 31-32: Lead time of 2 days and as long as 1-2 weeks. Can we use this as a guidance? Check the result and see if the model showed at least 2 days lead with a good correlation with clinical cases. Line 35: Please specify what virus. SARS-CoV-2? Method Line 65-66: Please specify why grab method was chosen, what are their limitation, around what time those samples were collected? How did the author measure the RNA copies of SARS-CoV-2? qPCR? What are the primers and probes, or kit used? What PCR machine was used? What is the limit of detection? What is the positive control used? Line 105-121: Did the author try to normalize the data by using flow and population? Please try and see if it works better with the model. What is quadratic interpolation and how did the author did it? What did author mean by clip any negative result? Cutting the data or use 0? Why don’t use a limit of detection number? Line 117-121: How did the author combine 2 STP and replace their daily viral loads with weighted average?? Please describe 1 calculation as example. For clinical cases. Was it the data based on hospital report? Or based on patient residency? Line 148-159: CUSUM. Please mention the full words before abbreviating it. How to measure the average run length? Expected time until a change was detected? Is it the same as what mentioned at line 157? So what was the h-value that applied here? Results Table 1 is too long. Maybe can be mentioned as supplementary information. Please make a summary of table 1. Like how many wards and their range of MLD. Line 182: Correlation analysis was not mentioned in the methodology section. Why was Pearson analysis considered? Line 184: This should be in method section Table 2 and 3: How the author concluded that there is no lead? there are lead number with good correlation there. Figure 5: How was the epi curve data obtained. It should be mentioned at the method section. How to determine the case shift? Figure 6: This should be discussed in the discussion section. How did the data compared based on windows of size? Line 198-219: Need to be in method section. Why did the author explain analysis method at the result section? Just curious, if the viral loads were measured weekly, is it possible to use the weekly average clinical cases for CUSUM analysis? If yes, why it was not conducted? Is it because the window is too small and it will be hard to get correlation? Based on Figure 9, I can see there was early warning where the peak was detected after the VCPD. Discussion. In my opinion, the discussion part is still needing more works. Line 228-231: Just repetition from method. Line 234-243: Early warning is not well defined. If the peak from wastewater followed by peak from clinical cases after several days, does not it count as early warning? Based on Fig 9 I can see that. What is the threshold for early warning? Figure 8 says 4 days. Please discuss from figure 9, what was the range for VDPC for all the STPs, did it consider as early warning. If not, why? Author only mentioned there is a good correlation between the wave of wastewater and clinical but no further discussion. Why are there good correlation? Is it the model? If yes how about comparing the model used here with another model? Please find more literature to discuss. For example, please compare to ARIMA model. An increase in wastewater viral load does not necessarily an outbreak indicator. ********** what does this mean? ). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files. Do you want your identity to be public for this peer review? If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. For information about this choice, including consent withdrawal, please see our Privacy Policy ..--> Reviewer #1: No Reviewer #2: No ********** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/ . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at [email protected] . Please note that Supporting Information files do not need this step. PLOS Glob Public Health. 2026 Apr 10;6(4):e0004684. doi: 10.1371/journal.pgph.0004684.r002 Author response to Decision Letter 1 Article notes Copyright and License information Collection date 2026. PMC Copyright notice 22 Jan 2025 Attachment Submitted filename: PLOS_Rebuttal.pdf pgph.0004684.s013.pdf (305.1KB, pdf) PLOS Glob Public Health. doi: 10.1371/journal.pgph.0004684.r003 Decision Letter 1 Mohan Amarasiri Mohan Amarasiri Academic Editor Find articles by Mohan Amarasiri Author information Copyright and License information Roles Mohan Amarasiri : Academic Editor © 2026 Mohan AmarasiriMohan AmarasiriMohan AmarasiriMohan Amarasiri This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice 8 Apr 2025 PGPH-D-24-02064R1 Localized wastewater surveillance showed correlation but no early warning during Bengaluru's Omicron wave PLOS Global Public Health Dear Dr. Midthala, Thank you for submitting your manuscript to PLOS Global Public Health. After careful consideration, we feel that it has merit but does not fully meet PLOS Global Public Health’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. One reviewer has highlighted the necessity to improve the methodology section. Please submit your revised manuscript by May 08 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at [email protected] . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pgph/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. We look forward to receiving your revised manuscript. Kind regards, Mohan Amarasiri Academic Editor PLOS Global Public Health Journal Requirements: Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice. Additional Editor Comments (if provided): [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author Reviewer #1: All comments have been addressed Reviewer #2: All comments have been addressed ********** publication criteria ? Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe methodologically and ethically rigorous research with conclusions that are appropriately drawn based on the data presented.? Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe methodologically and ethically rigorous research with conclusions that are appropriately drawn based on the data presented.-->?> Reviewer #1: Yes Reviewer #2: Partly ********** 3. Has the statistical analysis been performed appropriately and rigorously?-->?> Reviewer #1: Yes Reviewer #2: Yes ********** 4. Have the authors made all data underlying the findings in their manuscript fully available (please refer to the Data Availability Statement at the start of the manuscript PDF file)??> The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--> Reviewer #1: Yes Reviewer #2: Yes ********** 5. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ********** Reviewer #1: I recommend this manuscript for publication in its current form. The authors have thoroughly addressed all major methodological concerns through expanded analysis, validation, and appropriate uncertainty quantification. The conclusions are well-supported and properly qualified, making a valuable contribution to understanding localized wastewater surveillance during the Omicron wave. Reviewer #2: Reviewer Comments on the Rebuttal Manuscript The author has submitted a revised version of the manuscript in response to previous feedback. The revised manuscript is better written and demonstrates improved clarity. I have some comments remaining for further revision: Abstract Please include the correlation value(s) referenced. Kindly remove the statement mentioning who conducted the surveillance in 2021, as it is not relevant here. Please clarify what is meant by “correspond” in the context used. Author Summary The first bullet point is not a result of this study and should therefore be removed. Introduction The explanation of viral load currently in line 46 should be moved earlier, to line 41, for better context. Please clearly state the aims of the study in the final paragraph of the introduction. Refer to line 63 of the methodology for guidance. Methodology As noted for the abstract, please remove the sentence on line 69 referencing who conducted the surveillance. Line 69: What was the normalization method used? Please specify. Line 73: The researcher’s name mentioned here should instead be included in the Acknowledgments section. Figures 3 and 4 could be moved to the supplementary material. Figures 1 and 2 are sufficient for illustrating the map. Lines 128–132: Please include only the formula; the example calculation can be moved to the supplementary section. When describing the correlation analysis, please explain the assumption of linearity between viral load and COVID-19 case numbers. Why is a linear relationship appropriate here? Figure 5 should be moved to the supplementary material. Results should not be presented in the Methodology section. The explanation of the interpolation method should be made more concise. The associated calculation results (Figures 6 and 7, Tables 1 and 2) should be moved to the supplementary section, as these are results. The same applies to the CUSUM analysis: Figures 8 and Table 3 should be relocated to the supplementary material. For Table 5, please define what "PCv" refers to. Discussion The interpolation method used for a small dataset is interesting. Please discuss the interpretation of the results presented in Figures 6 and 7. What does the error factor represent? What values are considered good or poor in this context? How small can the dataset be for this interpolation method to still be effective? Please include a discussion of the study’s limitations and suggestions for improvement in future work. Line 393: Is the observation discussed here possibly due to limitations in the sampling strategy or data availability? Would more complete data potentially lead to different outcomes? ********** what does this mean? ). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files. Do you want your identity to be public for this peer review? If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. For information about this choice, including consent withdrawal, please see our Privacy Policy ..--> Reviewer #1: Yes: Charin ModchangCharin ModchangCharin ModchangCharin Modchang Reviewer #2: No ********** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/ . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at [email protected] . Please note that Supporting Information files do not need this step. PLOS Glob Public Health. 2026 Apr 10;6(4):e0004684. doi: 10.1371/journal.pgph.0004684.r004 Author response to Decision Letter 2 Article notes Copyright and License information Collection date 2026. PMC Copyright notice 5 May 2025 Attachment Submitted filename: PLOS_Rebuttal_auresp_2.pdf pgph.0004684.s014.pdf (252.9KB, pdf) PLOS Glob Public Health. doi: 10.1371/journal.pgph.0004684.r005 Decision Letter 2 Mohan Amarasiri Mohan Amarasiri Academic Editor Find articles by Mohan Amarasiri Author information Copyright and License information Roles Mohan Amarasiri : Academic Editor © 2026 Mohan AmarasiriMohan AmarasiriMohan AmarasiriMohan Amarasiri This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice 15 Aug 2025 PGPH-D-24-02064R2 Localized wastewater surveillance showed correlation but no early warning during Bengaluru's Omicron wave PLOS Global Public Health Dear Dr. Midthala, Thank you for submitting your manuscript to PLOS Global Public Health. In accordance with your request, we are returning the manuscript to your account so that you may revise it in light of the disparity in viral load values reported in your submission and in doi: 10.1016/j.lansea.2025.100619. Please submit your revised manuscript by Sep 29 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at [email protected] . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pgph/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter explaining the changes made to the manuscript. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. We look forward to receiving your revised manuscript. Kind regards, Steve Zimmerman, PhD PLOS Staff Editor Journal Requirements: If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. Additional Editor Comments (if provided): [Note: HTML markup is below. Please do not edit.] Reviewers' comments: [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/ . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at [email protected] . Please note that Supporting Information files do not need this step. PLOS Glob Public Health. 2026 Apr 10;6(4):e0004684. doi: 10.1371/journal.pgph.0004684.r006 Author response to Decision Letter 3 Article notes Copyright and License information Collection date 2026. PMC Copyright notice 29 Sep 2025 Attachment Submitted filename: response-to-review.pdf pgph.0004684.s015.pdf (40.9KB, pdf) PLOS Glob Public Health. doi: 10.1371/journal.pgph.0004684.r007 Decision Letter 3 Mohan Amarasiri Mohan Amarasiri Academic Editor Find articles by Mohan Amarasiri Author information Copyright and License information Roles Mohan Amarasiri : Academic Editor © 2026 Mohan AmarasiriMohan AmarasiriMohan AmarasiriMohan Amarasiri This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice 21 Nov 2025 PGPH-D-24-02064R3 Localized wastewater surveillance showed correlation but no early warning during Bengaluru's Omicron wave PLOS Global Public Health Dear Dr. Rajesh, Thank you for submitting your manuscript to PLOS Global Public Health. After careful consideration, we feel that it has merit but does not fully meet PLOS Global Public Health’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Please submit your revised manuscript by Dec 21 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at [email protected] . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pgph/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. We look forward to receiving your revised manuscript. Kind regards, Mohan Amarasiri Academic Editor PLOS Global Public Health Journal Requirements: If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice. Additional Editor Comments (if provided): [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author Reviewer #1: All comments have been addressed Reviewer #2: All comments have been addressed ********** publication criteria ? Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe methodologically and ethically rigorous research with conclusions that are appropriately drawn based on the data presented.? Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe methodologically and ethically rigorous research with conclusions that are appropriately drawn based on the data presented.-->?> Reviewer #1: Yes Reviewer #2: Yes ********** 3. Has the statistical analysis been performed appropriately and rigorously?-->?> Reviewer #1: Yes Reviewer #2: Yes ********** 4. Have the authors made all data underlying the findings in their manuscript fully available (please refer to the Data Availability Statement at the start of the manuscript PDF file)??> The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--> Reviewer #1: Yes Reviewer #2: Yes ********** 5. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ********** Reviewer #1: The authors have fully and transparently resolved the data discrepancy and other issues raised, without affecting the study’s conclusions. I recommend acceptance of the manuscript. Reviewer #2: This manuscript addresses a timely and important topic in covid19 surveillance by examining the relationship between wastewater viral loads and clinical case trends across multiple sites. As wastewater-based epidemiology continues to evolve as a complementary tool for infectious disease monitoring, especially for enteric and respiratory viruses, studies that evaluate its accuracy, temporal dynamics, and analytic approaches are essential. The authors present relevant data with clear public health implications; however, several methodological clarifications, expanded discussion points, and additional statistical context are required to enhance the rigor, interpretability, and impact of the findings. The comments below are intended to strengthen the manuscript and support its suitability for publication in PLOS Pathogens. Abstract Please include the confidence-interval values for city-level clinical data versus aggregated viral load so that the similarity with localized viral load trends can be justified. Introduction Line 4: Please revise this sentence. Antimicrobial resistance (AMR) and antimicrobial resistance genes (ARG) surveillance represents a recent application of wastewater surveillance. It should not be presented as a previous or early use of environmental surveillance. The cited reference is correct; however, the interpretation requires adjustment. Line 29: Please define the criteria used to classify a “strong correlation.” Line 33: Please elaborate on the reasons why increased sampling frequency strengthens correlation analysis. Line 34: Please provide a clear definition of “lead times.” Methods Lines 118–119: Please clarify the Limit of Detection (LoD) and Limit of Quantification (LoQ). If the LoD values are available, please explain why they were not used in place of zero. In addition, please clarify whether applying the LoD threshold would alter the results of the correlation analysis. Lines 122–124: Please explain why the sewage treatment plant (STP) capacity was used rather than daily or weekly flow-rate averages. Figure 4: Please include the units for viral load. In addition, the final portion of the figure appears to show decreasing clinical cases while viral loads increase. This pattern appears inconsistent with other STPs shown in Figure 5. Please discuss any potential explanations. Discussion Lines 343–349: I recommend expanding this section by introducing standard correlation strength categories (for example, low, moderate, and high) and discussing the rationale behind these classifications. Please also describe the statistical methods used, why they are appropriate, and whether they have been commonly applied in similar studies. Highlight which STP demonstrated the highest and lowest correlations and consider providing potential explanations, such as geographic context or differences in data availability. Lines 355–357: Please clarify the reason for the low viral loads observed during this period. Review lines 420–422 and assess whether the reported error bounds are sufficiently narrow to support confidence in the findings. Lines 360–362: Please justify the interpolation method used. A brief explanation regarding why interpolation is appropriate for this dataset would strengthen this section. Lines 363–372: This portion reads more like a methods description rather than a discussion. Please revise to address the advantages of the modelling approach, cite comparable studies that have applied similar models, and discuss alternative models that could have been used, including relative strengths and limitations. ********** what does this mean? ). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files. Do you want your identity to be public for this peer review? If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. If you choose “no”, your identity will remain anonymous but your review may still be made public. For information about this choice, including consent withdrawal, please see our Privacy Policy ..--> Reviewer #1: No Reviewer #2: No ********** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] PLOS Glob Public Health. 2026 Apr 10;6(4):e0004684. doi: 10.1371/journal.pgph.0004684.r008 Author response to Decision Letter 4 Article notes Copyright and License information Collection date 2026. PMC Copyright notice 21 Dec 2025 Attachment Submitted filename: PLOS_Wastewater_Manuscript_Response.pdf pgph.0004684.s016.pdf (64.4KB, pdf) PLOS Glob Public Health. doi: 10.1371/journal.pgph.0004684.r009 Decision Letter 4 Mohan Amarasiri Mohan Amarasiri Academic Editor Find articles by Mohan Amarasiri Author information Copyright and License information Roles Mohan Amarasiri : Academic Editor © 2026 Mohan AmarasiriMohan AmarasiriMohan AmarasiriMohan Amarasiri This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice 29 Dec 2025 Localized wastewater surveillance showed correlation but no early warning during Bengaluru's Omicron wave PGPH-D-24-02064R4 Dear Prof Rajesh, We are pleased to inform you that your manuscript 'Localized wastewater surveillance showed correlation but no early warning during Bengaluru's Omicron wave' has been provisionally accepted for publication in PLOS Global Public Health. Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests. Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated. IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they'll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact [email protected] . Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Global Public Health. Best regards, Mohan Amarasiri Academic Editor PLOS Global Public Health *********************************************************** Reviewer Comments (if any, and for reference): Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials S1 Fig. Drainage pipes from wards leading to the Hebbal STP. (PDF) pgph.0004684.s001.pdf (440.2KB, pdf) S2 Fig. BWSSB sewer lines coverage. (PDF) pgph.0004684.s002.pdf (534.6KB, pdf) S3 Fig. A scatter plot of viral loads and cases from four STPs. (PDF) pgph.0004684.s003.pdf (71.5KB, pdf) S4 Fig. Error bars for viral load estimates at STPs in California. (PDF) pgph.0004684.s004.pdf (30.3KB, pdf) S5 Fig. Error bars for cases estimates at STP catchments in Bengaluru. (PDF) pgph.0004684.s005.pdf (31.4KB, pdf) S6 Fig. Distribution of correlation values between a vector and its noisy version. (PDF) pgph.0004684.s006.pdf (37.4KB, pdf) S1 Table. Parameter values of the pre-and-post distributions of alternate slopes. (PDF) pgph.0004684.s007.pdf (78.4KB, pdf) S2 Table. Correlation between daily viral loads and values interpolated from weekly samples in California. (PDF) pgph.0004684.s008.pdf (50.1KB, pdf) S3 Table. Correlation between daily cases and interpolated cases from weekly samples in Bengaluru. (PDF) pgph.0004684.s009.pdf (44.5KB, pdf) S4 Table. List of files related to datasets, code, and results. (PDF) pgph.0004684.s010.pdf (61.2KB, pdf) S1 Text. An example of computing weighted viral load. (PDF) pgph.0004684.s011.pdf (60.9KB, pdf) S2 Text. A summary of our work in Kannada. (PDF) pgph.0004684.s012.pdf (39.8KB, pdf) Attachment Submitted filename: PLOS_Rebuttal.pdf pgph.0004684.s013.pdf (305.1KB, pdf) Attachment Submitted filename: PLOS_Rebuttal_auresp_2.pdf pgph.0004684.s014.pdf (252.9KB, pdf) Attachment Submitted filename: response-to-review.pdf pgph.0004684.s015.pdf (40.9KB, pdf) Attachment Submitted filename: PLOS_Wastewater_Manuscript_Response.pdf pgph.0004684.s016.pdf (64.4KB, pdf) Data Availability Statement Data is available at http://dataportal.icts.res.in/wastewater/ . Articles from PLOS Global Public Health are provided here courtesy of PLOS ACTIONS View on publisher site PDF (1.9 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• On dedicated institutions and funding to support disaster risk reduction: Reflections on local instrumentation in Chile.#659900
• Self management strategies for hypertensive patients on treatment in selected clinics at Ga-Dikgale, Capricorn District, Limpopo, South Africa.#659901
• From integration to use: a qualitative study of barriers and facilitators to postpartum family planning use in northeast Ethiopia—a multistakeholder perspective.#659902
• Universal Health Coverage in Pakistan: Challenges, Reforms, and the Road to 2030.#659903
• Association of social restrictions with fear of cancer recurrence in hematopoietic stem cell transplantation patients: an exploratory analysis of the potential chain mediation role of illness perception and cognitive emotion regulation.#659904