Optimum sparse array design for maximum output signal-to-noise ratio (MaxSNR) and signal-to-interference ratio (MaxSINR) have been shown to yield significant performance improvement compared to random or environmental-independent structured arrays, with the same number of antennas. We examine the MaxSNR problem in presence of local scatterings which may follow specific deterministic or statistical scattering models. It is shown that if the scatterers assume a Gaussian distribution centered around the source angular position, the optimum array configuration for maximizing the SNR is the commonly used uniform linear array (ULA). If the scatterers circulate the source, the optimum design yields sparse array topologies with superior performance over ULAs. Simulation results are presented to show the effectiveness of array configurability in the case of both Gaussian and circularly spread sources.