The land surface plays a critical role in the global cycles of energy, water, carbon and other elements essential for life. The models used for both weather forecasting and climate prediction include land-surface schemes. Land-surface models (LSMs) have evolved to incorporate many individual processes, and most LSMs now incorporate demographic elements derived from the parallel development of dynamic global vegetation models (DGVMs). However, the diversity and complexity of global ecosystems means that current LSMs require many hundreds of parameters whose values are mostly poorly constrained and are difficult to calibrate by any practical method. This problem has slowed the advancement of LSMs such that they are neither improving in their fidelity to observations, nor converging in their future predictions. New approaches to reducing the complexity, including the adoption of eco-evolutionary optimality theories as a basis for process understanding, are being explored but there is still a long way to go to develop more robust, parameter sparse LSMs. In this session, we invite contributions that address the problems of parameterisation, explore new approaches to develop simpler model frameworks, provide insights into how the ever increasingly wealth of data can be used for testing individual processes, or showcase applications of new theoretical approaches in an LSM context.