Inverse thermal history modeling is an effective tool to explore plausible time-temperature (t-T) histories that can be used to describe the geologic history of a sample. Although in some inverse modeling exercises the input thermochronology data is consistent with a single set of t-T histories with similar heating and cooling trends, more commonly inverse models identify a range of paths with different and distinct heating and cooling histories but similarly good fits to the data. Each set of these “path families” typically requires a different geologic interpretation to explain the observed heating and cooling trend, so it is important to identify and consider all possible path families consistent with the regional geology that fit the modelled dataset before selecting a preferred geologic interpretation. Although the inverse model results are always consistent with measured data, a model’s ability to detect all possible path families is partly controlled by the model design – for example the choice of initial conditions, monotonicity settings, and forced time-temperature windows. In this exercise using the thermal history modeling program HeFTy, we illustrate the effects of model design on the inverse model results of a set of multi-chronometer datasets from the southern Patagonian Andes. We use model design to maximize the number of path families identified through inverse modeling. Once individual paths are classified according to path families, we use independently constrained regional geology to discriminate among the diverse plausible set of path families and evaluate different available geologic scenarios. Our exercise illustrates that models restricting exploration of all path families may not identify the true cooling history of the sample. Initially, it may appear challenging to interpret inverse model results that include multiple path families, but we argue that iterating between independent geologic data and modeling provides an effective tool to test the geologic plausibility of alternative heating and cooling histories. Although this exercise is executed using HeFTy, maximizing the identification of all possible path families should be an important component of model design in inverse modeling exercises using all inverse modeling programs.

Numerical thermal history modelling has become a core approach used for interpretation of low-temperature thermochronometry data. Modelling programs can find rock time-temperature (t-T) paths that fit the input data while incorporating independent geologic information about a sample’s history and leveraging the factors that impact the kinetics of each thermochronometric system (e.g., grain size, radiation damage, and composition). HeFTy (Ketcham, 2005) and QTQt (Gallagher, 2012) are two of the commonly used tools for both forward and inverse t-T modeling. The modelling process involves making key decisions about (i) data input, (ii) initial set-up of model space and parameters, (iii) kinetic model(s) (i.e. annealing, diffusion, radiation damage), and (iv) additional t-T constraints. In addition, users need to have an understanding of the statistical methods underlying the modelling approach to be able to interpret the model outputs and the relationship between the observed and predicted data. However, these modeling tools currently lack clear and accessible entry-points for all users—experienced and new thermochronologists alike—and thus for many geoscientists, there is a substantial barrier to the modeling, interpretation, and publication of thermochronologic datasets. Here we present a suite of simple forward and inverse models that we recommend everyone perform before embarking on t-T modeling in HeFTy and/or QTQt for the first time. At the core of the exercises are the six different t-T paths used by Wolf et al. (1998) to illustrate the partial-retention behavior of the apatite He system; however, this approach can easily be applied to other systems as well. This exercise not only illustrates the fundamental behavior of thermochronologic systems but also guides users through the main functionality of the modelling programs. Despite the apparent simplicity of this exercise, users will experience most of the challenges and opportunities common to thermal history modeling, including: how to enter data; error handling; how to use geologic constraints in t-T space; the non-unique nature of cooling ages; the power of grain size and eU variability; the limitations on a model’s ability to resolve the ‘right’ rock thermal history; and how to evaluate the sensitivity of model results to all these factors. These exercises were introduced in the Thermo2020/1 Sunday workshops for both QTQt and HeFTy and are more fully fleshed out in two publications currently in preparation.