The deployment of surface-based sensors/samplers is a common practice for emission and air quality monitoring purposes and the proper section of sites for the measurement equipment is critical to an accurate characterization of the emissions. Typically the approach used to make a sampler siting decisions utilizes a large number of dispersion model computations where the initial conditions are varied based on a climatological record for the area. The results of the dispersion simulations are then used to map the areas where emissions are most likely to be transported. While this approach works well for many applications, it becomes computational costly to use this approach in situations where higher fidelity solutions involving more complex dispersion models are required. One such situation is the positioning of chemical sampling instruments in urban environments where the availability of suitable sites and the non-intuitive dispersion patterns associated with the wind flow around the buildings and through the urban canyons make site selection difficult. This presentation illustrates a more computational efficient methodology for identifying the optimal locations for air quality monitoring equipment deployed in this complex challenging environment. The method involves using a series of coupled technologies to map the probability of detection (POD) for a given emission detection threshold. The approach involves the following elements: 1. A high-resolution (40 km horizontal spatial resolution) gridded climatological reanalysis; 2. a multi-dimensional pattern extraction and classification technique know as Self Organizing Maps (SOM) that is used to characterize the variety of weather patterns relevant for atmospheric transport and dispersion and their frequencies of occurrence; 3. the construction of building-aware wind flow fields for the urban environment for the SOM weather patterns; 4. interior dispersion modeling that utilizes the wind-loading pressure from the building aware wind model to identify likely material exfiltration paths; 5. and simulations from a Lagrangian particle dispersion model to map the exterior dispersion patterns. The exterior dispersion patterns and associated frequency of occurrence are then combined to estimate the map of the POD for a given detection threshold. The method is flexible and can be tuned to allow the detailed characterization of POD for a given sampler detection threshold and sampling period (e.g. sampling duration, season, time of day). An example of this methodology is illustrated for a single facility in an urban location surrounded by numerous multi-story buildings.