Skip to main content

Characteristics of the urban sewer system and rat presence in Seattle

Cite this dataset

Guo, Xiaocong et al. (2022). Characteristics of the urban sewer system and rat presence in Seattle [Dataset]. Dryad.


Rats are abundant and ubiquitous in urban environments. There has been increasing attention to the need for evidence-based, integrated rat management and surveillance approaches because rats can compromise public health and impose economic costs. Yet there are few studies that characterize rat distributions in sewers and there are no studies that incorporate the complexity of sewer networks that encompass multiple sewer lines, all comprised of their own unique characteristics. To address this knowledge gap, this study identifies sewer characteristics that are associated with rat presence in the city of Seattle’s urban sewer system. We obtained sewer baiting data from 1752 geotagged manholes to monitor rat presence and constructed generalized additive models to account for spatial autocorrelation. Sewer rats were unevenly distributed across sampled manholes with clusters of higher rat presence at upper elevations, within sanitary pipes, narrower pipes, pipes at a shallower depth, and older pipes. These findings are important because identifying features of urban sewers that promote rat presence may allow municipalities to target areas for rat control activities and sewer maintenance. These findings suggest the need to evaluate additional characteristics of the surface environment and identify the factors driving rat movement within sewers, across the surface, and between the surface and the sewers. 


Data was collected in the port city of Seattle, Washington USA (47.6°N, 122.3°W) between February 2016 and September 2019 as a part of Seattle’s ongoing rat sewer baiting program. In the baiting program, manholes across the city were geotagged in map grids, where all grids in one zone were baited before moving to the next zone. The method for monitoring a manhole includes an initial assessment with four non-toxic Talon Weather BlocTM bait blocks. Blocks were suspended from the manhole so that they hung just above the sewer surface. Bait consumption was measured 10 days after the initial visit to monitor rat presence. Rats were considered present in manholes if some bait was consumed and/or signs of rodents (e.g., rodent gnaw marks, rat droppings) were observed.

Data regarding the consumption of non-toxic bait (rats were considered either present or absent in each manhole) were joined with three publicly accessible municipal datasets. These datasets included 10 manhole characteristics (point features), 21 sewer line characteristics (line features), and 2 surface characteristics. Two weather characteristics, temperature (average monthly temperature) and precipitation (cumulative monthly precipitation), were obtained from the Western Regional Climate Center and Seattle Weather Blog.

Usage notes

We used RSudio (version 1.3) for statistical analyses and ArcGIS Pro (version 2.6.0) for spatial analyses.


Canadian Agricultural Partnership