There is an urgent societal need for the development of high-accuracy models that quantitatively assess the sensitivity of landscapes to climate-change over time-scales of decades to millennia and at high-resolutions. Progress is being hampered, in part, by the difficulty of integrating the necessarily diverse range of disciplines whose processes are dynamically coupled as environments respond to climate change. Such models are ideally calibrated with data on landscape change under past climate change and from well-characterized catchment observatories. Current impact models deal typically with event-specific characterizations and/or a relatively limited number of processes often within generic environments. This session calls for contributions toward the development of more integrated models, including the demonstration of coupling, internal dynamics of processes and calibration on paleo-records of landscape change. It encourages site-specific applications and evidence of performance success.