Urgent environmental challenges have created a need to more fully leverage the subsurface for its water, energy, and carbon-sequestering resources. Doing so safely and economically will, however, require robust and adaptable tools for engineering subsurface fluid flow and chemical reactions. Microbially-enabled technologies have shown great promise to this end, but despite years of research, understanding of microbial transport and activity in natural rock systems remains incomplete. This dissertation aims to advance fundamental understanding of these processes, with a particular focus on developing microbially-induced carbonate precipitation (MICP) for controlling subsurface fluid flow. To achieve the aforementioned goal of extracting mechanistic information from empirical studies while also capturing the relevant complexities of the natural subsurface, several creative experimental approaches are utilized. In Chapter 2, the effect of pore structure on the efficacy of microbial mineralization reactions is assessed, with an eye towards understanding how precipitate mobilization might influence permeability reduction. Modifying the pore structure of natural limestone rock cores allows the influence of elevated flow velocity through preferential flow paths to be isolated, and a mechanism of shear-induced precipitate removal from pore walls is proposed to explain stark differences from previously modeled results. In Chapter 3, the additional variable of rock mineralogy is incorporated, and its effects are decoupled from those of porosity and pore structure using flowthrough columns packed with homogenized sandstone grains. Sandstone clay content is found to work in tandem with pore size to mediate microbial attachment to solid surfaces, with significant downstream effects on the quantity and distribution of precipitates. These effects are attributed in part to the accumulation of precipitates in suspension, a mechanism currently excluded from most MICP models. In Chapter 4, a custom microfluidics platform incorporating natural rock chips is used to investigate biofilm formation in artificial fractures, with the goal of characterizing the influence of biofilm development on microbial mineralization processes. The rock properties that promote robust biofilm development are found to differ from those found to promote planktonic cell attachment in Chapter 3. Furthermore, in MICP-capable biofilms, biofilm development is found to significantly influence the quantity, timing, and morphology of carbonate precipitates. Finally, in Chapter 5, a newly optimized DNA extraction method is applied to quantitatively characterizing the phenomenon of biomass encapsulation by carbonate precipitates for the first time. A nonlinear relationship between calcium concentration, precipitation rate, and microbial growth is identified, and a modeling exercise is carried out in order to probe the mechanistic basis for this nonlinearity. Taken together, these findings have important implications for how microbes, biofilms, and microbially-mediated precipitation processes are understood and modeled in complex porous media. The potential applications of this fundamental knowledge are numerous, spanning groundwater treatment, metal resource recovery, hydrogen production, sustainable concrete alternatives, and carbon sequestration, to name a few examples.