Throughout the world, several millions of small ponds exist for water supply, irrigation, flood control or to control water quality downstream. The reduced flow velocity in these ponds causes sedimentation of tranported particles. For most ponds this is a negative impact as their retention capacity decreases due to sedimentation processes. Sediment volumes in small ponds can be used to reconstruct sediment yield values and to study the spatial variation in sediment yield over large areas. Especially in developing countries, this technique can be very helpful in establishing large data sets on sediment delivery as there are often no resources for expensive monitoring programmes. However, when such studies are undertaken, one has to take into account the efficiency of the pond in trapping sediments. This trap efficiency is dependent on the characteristics of the inflowing sediment and the retention time of the water in the pond, which in turn are controlled by pond geometry and runoff characteristics. Because trap efficiency is one of the most important properties of a pond or reservoir, it has been studied for quite some time. This article provides an overview of the different methods available to estimate the trap efficiency of reservoirs and ponds. The first set of methods are empirical models that predict trap efficiency, mostly of normally ponded large reservoirs using data on a mid to long-term basis. These models relate trap efficiency to a capacity/watershed ratio, a capacity/annual inflow ratio or a sedimentation index. Today, these models are the most widely used models to predict trap efficiency, even for reservoirs or ponds that have totally different characteristics from the reservoirs used in these models. For small ponds, these models seem to be less appropriate. They also cannot be used for predicting trap efficiency for a single event. To overcome these restrictions, different theoretical models have been developed based on sedimentation principles. These can be very simple, such as the overflow rate method, but also very complex when runoff and sediment are routed through a pond with incremental time-steps. The theoretical-based models are probably more capable of predicting trap efficiency for small ponds with varying geometric characteristics, and some of them also provide data on effluent sediment concentrations and quality. However, when reconstructing sediment yield values using sedimentation rates over a period of a few months to a few year (the mid-term basis), one needs a trap efficiency value for the whole period, not for one single event. At present only limited research has been done on establishing mid-term trap efficiency models based on theoretical principles. This is probably the most important gap in trap efficiency research.