The site selection process for a geomagnetic observatory placement is essentially a rigorous effort to eliminate or minimize all possible sources contaminating observatory measurements now and for the foreseeable future. The performance of a geomagnetic observatory critically depends on the signal-to-noise ratio of the recorded magnetic field components. All EM noise sources that distort this ratio must be identified and considered in site selection. Site selection for geomagnetic observatories traditionally involves evaluating multiple environmental, geological, and infrastructural factors, but without classification, these criteria are often assessed ad hoc, leading to inefficiencies and suboptimal choices. Classification addresses these by enabling systematic, data-driven prioritization of criteria, reducing subjectivity, and facilitating scalable GIS-based multi-criteria evaluations across large regions.

In this study, we first carry out a detailed review and classification of EM noise sources that affect geomagnetic measurements. These sources are grouped into three main categories: (a) artificial EM noise, (b) natural EM noise related to geological, crustal magnetic anomalies and atmospheric processes, and (c) natural hazards that threaten long-term observatory operation. Artificial EM noise includes power transmission and distribution networks (three phase high-voltage lines and single phase medium-voltage feeders), electrified railways and DC metro systems, buried metallic pipelines with cathodic protection, radio and radar communication systems, and electric security fences. For each of these sources, published case studies and theoretical considerations are used to describe the frequency content, dominant field components, amplitude decay with distance, and recommended minimum distances (e.g., tens of meters to several kilometers) required to avoid strong contamination of geomagnetic records.

Natural EM noise and geophysical conditions are also analyzed as essential criteria. Topographic variations and associated charge accumulation on surface irregularities, spatial changes in crustal magnetic susceptibility inferred from airborne magnetic data, and large-scale atmospheric and ionospheric phenomena can all deform the EM field and reduce the interpretability of observatory data. Moreover, because one of the main tasks of a geomagnetic observatory is to provide long time series for studies of core dynamics, the site must remain stable over time and be protected from destructive geological processes. Therefore, active faults, seismic hazard zones, mass-movement–prone slopes, and other geohazards in Tehran Province are mapped and incorporated as exclusion or low-suitability zones. In the next step, these conceptual criteria are translated into spatial information layers for Tehran Province. Ten main layers are generated for artificial EM noise sources, including the locations of high-voltage transmission lines, oil and gas pipelines, railways and DC metro lines, cities and villages, active mines, industrial zones, power plants, and transformer substations. These data are compiled from national geological and mining databases and from open, collaborative geographic databases and are validated and edited using satellite imagery and GIS tools. Additional layers representing topographic slope derived from SRTM-based digital elevation models, surface water bodies, crustal magnetic anomalies, lithological units, active faults, and seismic hazard zones are also prepared.