Geothermal heat flow (GHF) is a critical parameter for understanding the thermal structure and dynamics of the lithosphere, offering key insights into geophysical processes and geothermal energy potential. This study investigates the spatial variability of GHF in Germany by applying a Bayesian Markov Chain Monte Carlo method to estimate key thermal parameters, including crustal and mantle thermal conductivities, crustal heat production, and mantle heat flow. The analysis integrates data on surface heat flow, surface temperatures, and the lithosphere-asthenosphere boundary (LAB) depth.   To address the limitations posed by the sparse and uneven distribution of direct borehole measurements, comprising only 595 GHF records, we incorporated a wide range of geophysical and geological datasets, such as gravity, magnetics, seismic velocity, topography, and proximity to faults and volcanic regions. These datasets were analyzed using a Quantile Regression Forest (QRF) approach, which enabled robust GHF estimations while accounting for uncertainties and providing reliable prediction intervals. This methodology significantly improves upon traditional Curie depth-based methods, providing a more accurate and comprehensive GHF model for Germany.   The probabilistic multi-geoobservable modelling presented here enhances our understanding of the geothermal regime in Germany, contributing to a more precise assessment of its geothermal resources and the thermal state of its lithosphere.

Surface heat flow provides essential information on the thermal state and thickness of the lithosphere. Southern Africa is a mosaic of the best-preserved and exposed crustal blocks, assembled in the early late Archean and then modified by a series of major tectono-thermal events, both of Precambrian and Phanerozoic. Understanding the thermal and compositional structure of the southern African lithosphere provides crucial information for the actual causes, processes of lithospheric stability, and modification. Temperature plays a major role in the distribution of the long-wavelength crustal magnetic anomalies. Curie depth, interpreted as the depth to $580^{\circ}$C, provides a valuable constraint on the thermal structure of the lithosphere. Due to the sparse distribution of surface heat flow data, we examine the degree to which the thermal structure of the crust can be constrained from the Curie depth in southern Africa. The Curie depth is estimated from magnetic anomaly data using spectral methods in combination wavelet analysis; a Bayesian approach is applied to address the uncertainty. Subsequently, the obtained Curie depth is used to estimate the surface heat flow, and the outcome is compared to available heat flow measurements. Unlike other cratonic regions, the shallowest Curie depth and low effective elastic thickness values observed over the Kaapvaal Craton suggest thermal reworking of the cratonic lithosphere in this region.