The commentary of Bronstert (2004) cannot be labeled with NASA's LUCC scientific questions. The paper itself provides the research questions in the hydrological modeling domain with an emphasis on rainfall-runoff modeling. Several key factors that require special attention, as pointed out by the author, include (1) drivers, (2) appropriate spatiotemporal scales of observation¹, (3) incorporation of feedback, if necessary, (4) modeling aims (normal or extreme events), and (5) scenario-driving modeling.

The commentary provides a nice summary into six points. If interested, they can be found in the last page of the commentary.

Footnote
¹Aylward (2005) points out a related issue for appropriate spatiotemporal scales: most "existing studies have been undertaken at small scales (less than 10 km²) in headwater basins and over relatively short durations, making accurate extrapolation and 'upscaling' difficult.".

References
Aylward, B., 2005, Towards watershed science that matters, Hydrologic Processes, 19, 2643-2647
Bronstert, A, 2004, Rainfall-runoff modelling for assessing impacts of climate and land-use change, Hydrological Processes, 18, 567-570 [doi]