| Abstract |
The former gas exploration well ?Horstberg-Z1? was the first in the N-German basin to be used for testing single-well, waterbased, proppant-free techniques of petrothermal heat extraction from deep-seated, tight sedimentary layers (Jung et al. 2005). Besides other options tested (?huff-puff?, large-scale circulation), maybe the most innovative, interesting, and promising one (Orzol et al. 2005) was that of a so-called ?frac circulation?. It relied on first creating a large-area fracture, primarily by means of massive water injection into a 4 km deep lowly-permeable sandstone layer. This ?waterfrac? was supposed, and confirmed (Wessling et al. 2009, Tischner et al. 2010) to propagate into adjacent tight claystone-sandstone layers, more or less vertically, and become arrested, upwards, within a significantly more permeable sandstone layer. A special wellbore completion provides for ?two wells within one hole?: the inner tubing is screened within the waterfrac initiation layer (lowly-permeable sandstone), whereas the wellbore annulus is screened within the waterfrac-arresting layer (more permeable sandstone). Producing / injecting the hot / cold fluid simultaneously through well annulus / inner tubing requires reliable packer technology and sufficient thermal insulation between upwards / downwards circulating fluids. The memorable waterfrac + tracer test conducted, almost a decade ago, at Horstberg is still inspiring some new answers to ?old? questions like: why single-well? why fracturing? why tracers? does this only work at Horstberg, or can it work almost anywhere else in the N-European sedimentary basin? Heat and tracer transport within waterfrac and matrix turn out to fit into a surprisingly simple description, as the plain arithmetic sum of certain ?petrothermal?- and ?aquifer?-type contributions, whose relative weighting can vary from site to site, depending upon stratigraphy and wellbore geometry. At Horstberg, within the particular formations tested (?Solling?, ?Detfurth?, ?Volpriehausen?, comprising mainly claystone and sandstone layers), thermal lifetime results to be petrothermally dominated, while tracer residence times appear to be ?aquifer?- dominated. Despite this incongruence, thermal lifetime can reliably be predicted from tracer test results. What cannot be determined from ?waterfrac flow-path tracing?, is precisely the waterfrac aperture; aperture uncertainty, however, does not impede upon thermal lifetime predictability. Results of a semianalytical approach are confirmed by numerical simulations using a FE model that includes more details of hydrogeological heterogeneity for the Horstberg site. |