Leak detection event aggregation and ranking systems and methods

A method comprising:

• collecting a set of training data for training a risk model configured to rank natural gas leaks by hazard level, the training data including a hazard level value and values of a set of hazard-predictive features for each of a plurality of natural gas leaks, the values of the set of hazard-predictive features being determined according to data collected by a vehicle-borne gas concentration measurement device performing a sequence of geospatially-referenced mobile gas concentration measurements along one or more survey paths, the collected data including gas concentration and location;
• and
• configuring a set of parameters of the risk model according to the training data to generate predicted hazard level values for the plurality of natural gas leaks that match hazard level values in the training data.

A non-transitory computer-readable medium encoding instructions which, when executed by a computer system comprising a hardware processor and an associated memory, cause the computer system to:

• collect a set of training data for training a risk model configured to rank natural gas leaks by hazard level, the training data including a hazard level value and values of a set of hazard-predictive features for each of a plurality of natural gas leaks, the values of the set of hazard-predictive features being determined according to data collected by a vehicle-borne gas concentration measurement device performing a sequence of geospatially-referenced mobile gas concentration measurements along one or more survey paths, the collected data including gas concentration and location;
• and
• configure a set of parameters of the risk model according to the training data to generate predicted hazard level values for the plurality of natural gas leaks that match hazard level values in the training data.