Plant-soil-water models are applied in two case studies. Attention is given to the desired level of accuracy in (agro-)hydrological. research when applied in problem-oriented studies. In the case studies it is shown that when decision criteria are only roughly known and when only little information is available about the other aspects relevant to the problem, simple plant-soil-water models should be used.A simple water balance and crop growth model was applied to simulate production of maize in southern Mozambique. For different sowing strategies, varying from a scattered one to sowing once a year, the available maize for consumption for an average family farming unit wasdetermined. Different model parameters were varied to study their impact upon sowing strategies.For maximizing yearly consumption the preferred strategy almost fully depended on losses by pests and diseases and post-harvest losses. However, regarding the decision criterion of minimizing the periods with food shortage the preferred sowing strategy greatly depended on water-availability and potential production levels.Both field and model studies were carried out to study the water balance of bog relicts in the Netherlands. In the field experiments special attention was given to the hydrophysical properties of the upper peat layers and to the evapotranspiration ofSphagnum papillosumand Molinia caerulea .A simple model was applied to simulate the water level fluctuations in a bog relict under different water management options. Both external measures (e.g. hydrological bufferzone's) and internal measures (water conservation) were analysed. Which option has to be preferred depends on the hydrological conditions in the bog relict and on the ecological constraints for bog regeneration, which are not fully known.For an understanding of the hydrology of bogs and bog relicts detailed knowledge of the plant-soil-water relations is required. However, given the high spatial diversity in these areas, simple plant-soil-water models should be used.