Mitigating climate change will require unprecedented deployment of a wide variety of technologies, many of which are in the early stages of development. Assessing the long-term environmental and natural resource risks, benefits of many of these technologies is challenging due to their changing capabilities, costs and resource requirements. This dissertation consists of three articles that explore the nature of technological change in the assessment of rapidly expanding greenhouse gas (GHG) mitigation technologies using hybrid life cycle assessment (LCA).

Thin-film photovoltaics (PV), including cadmium telluride (CdTe) and copper indium gallium selenide (CIGS), are rapidly-growing renewable electricity generation technologies. Thin films have improved significantly recently, and similar improvements are expected in the future, warranting re-evaluation of the environmental implications of PV to update and inform policy decisions. The life cycle impacts of thin films can be reduced by several likely cost-reducing technological changes: (1) module efficiency increases (2) module dematerialization (3) changes in upstream energy and materials production and (4) end-of-life recycling of balance of system (BOS). Results show comparable environmental and resource impacts for CdTe and CIGS. Compared to the US electricity mix in 2010, both perform at least 90% better in 7 of 12, and at least 50% better in 3 of 12 impact categories, with comparable land use, and increased metal depletion unless BOS recycling is ensured. Technological changes, particularly efficiency increases, contribute to 35-80% reductions in all impacts by 2030.

Efficient lighting based on light emitting diodes (LEDs) is another technology that is increasingly being deployed to save energy and thus mitigate GHG emissions. Efficient LED light sources currently have lower environmental and natural resource impacts than traditional light sources, and are expected to see dramatic improvements in luminous efficacy and possibly their life cycle environmental impacts in the future. Aggressive deployment of these increasingly efficient LED lamps and luminaires along with decarbonization of electricity generation can reduce the life cycle GHG emissions from the global provision of lighting by more than a factor of seven while at the same time stabilizing metal depletion and allowing global demand for lighting services to grow by 2.5 - 2.9 times by 2050.

LCA results show that technological changes can influence the assessment of GHG mitigation technologies, but existing models are not adequate for predicting the rates of technological changes and how those changes affect the life cycle environmental impacts and resource requirements of technologies. Technology learning curves use empirical observations to show how the costs of technologies decline with increased cumulative production, but do not distinguish the direct and supply chain by which cost reductions are achieved. The final chapter of this dissertation proposes a mathematical framework to separate the effects of learning on the intermediate inputs to a technology from the effects of learning on value added, and incorporates those technological learning effects throughout the supply chain of a technology based on the computational structure of LCA. An example for CdTe PV shows that technology learning throughout the supply chain can contribute to observed learning rates for GHG mitigation technologies.