During the second half of this century offshore sand extraction has become commonplace. There has been an important increase of sand demand to use on construction and coastal protection as well as for industrial purposes. On the other hand, the development in the dredging field has helped to the expansion of the process. Since nearshore dredging is an alteration in a very dynamic environment with important consequences, it is sought to find the minimal depth where a trench can be dredged without affecting the coastal stability and neither considers an excessive restriction. The trivial solution would be to extract in deep bottoms, but the price of these operations (they become more expensive with increasing distance offshore) and also sediment availability makes it unrealistic. Is in this context of uses and resources where this work is done. First, trying to recognise which are the potential effects of nearshore dredging on coastal stability and next, looking if there is data enough to prove them. Then, after identifying the main effects, a methodology to prevent these impacts is proposed, i.e. some engineering rules to minimise the impacts and select the zones to be dredged. Theoretically nearshore dredging and the resulting trench can affect the hydrodynamics, the sediment transport and control the coastal behaviour and, as a consequence, the coastal stability, through different mechanisms: the interaction with longshore sediment transport, the beach drawdown, the interception of the onshore sediment transport in the inner shelf, the modification of wave characteristics, the modification of wave's field of velocity and the trench propagation to the coast. There are different depth criterion to prevent each of these impacts but at the same time, the ones associated to some of them are exceeded by others. Thus, the way to select the depth criterion to prevent all of them must be based on the most restrictive one. On the other hand, it has been observed that other impacts just would verify in particular kind of beaches or coasts (e.g. the interception of onshore transport) or within long time scales (e.g. trench propagation) and although potentially they would be able to affect the coastal stability, they will mainly verify in very specific coastal stretches and, in consequence, they are not very common. Finally, and after analysing several reported coastal responses, three effects have been identified as the most likely to happen and therefore, some rules are necessary to be sure they will be avoided: beach drawdown, trench propagation and wave's modification. To avoid the called beach drawdown, trenches should be done seawards of the depth of closure d,. Therefore it is necessary to find this threshold depth of significant vertical changes, but since it has been demonstrated that it is time and space dependant it is not a trivial task. However, a first assessment of the minimum depth can be done using Hallermeier's equation fed by extremal wave conditions, selected to be representative of a return period according to the life period of the trench, e.g. Tretum>25 years. It has been tried to represent trench propagation with a process-based model, but the results have been just qualitative and not quantitative. However, it can exist so to keep sediment transport gradients small and avoid it, trenches should have little depth or be in depths where the transport rates are small. A general criterion could be to dredge seawards of the depth that represents the beginning of significant sediment transport, which at the same time would avoid the interception of the onshore transport in the inner shelf. The third effect is the modification of wave characteristics. It induces longshore sediment transport gradients that suppose a change in shoreline development as it has been observed in nature. This effect will be significant in long coasts such that their behaviour is conditioned by the longshore sediment transport, but there it will be less important if trenches are shallow and not very wide (less than 400m, Viguier et al., 1984).