The aim of this study was an assessment of feasibility of conversion of sewage holding (SH) tanks to rainwater harvesting (RWH) tanks in Poland. Such a conversion may partly solve the problem of water scarcity for irrigation of plants in individual small gardens and reduce tap water consumption. Seven methods of RWH tanks sizing were applied to an example of a small harvesting system of the roof area equal to the garden irrigation area of 100 m² for three different irrigation doses. A new criterion was introduced to optimize the tank capacity. Economic optimization was provided for new RWH tanks and for the tanks adapted from abandoned SH tanks. Results obtained for a system sited in west-central Poland in an average year have shown that design capacity of RWH tanks varied markedly between sizing methods. The conversion of SH tanks to RWH tanks is profitable, especially for irrigation due to scarcity of water in relatively dry west-central regions. Conversion of individual SH tanks in a good technical state to RWH tanks is relatively simple and cheap. The potential increase in storage volume due to the conversion of individual SH tanks to RWH tanks could reach all over Poland 215–350 dam³ per year, and individually can save up to 18–25% of total annual water use.

Scarcity of freshwater is one of the major issues which hinders nourishment in large portion of the countries like Ethio-pia. The communities in the Dawe River watershed are facing acute water shortage where water harvesting is vital means of survival. The purpose of this study was to identify optimal water harvesting areas by considering socioeconomic and biophysical factors. This was performed through the integration of soil and water assessment tool (SWAT) model, remote sensing (RS) and Geographic Information System (GIS) technique based on multi-criteria evaluation (MCE). The parame-ters used for the selection of optimal sites for rainwater harvesting were surface runoff, soil texture, land use land cover, slope gradient and stakeholders’ priority. Rainfall data was acquired from the neighbouring weather stations while infor-mation about the soil was attained from laboratory analysis using pipette method. Runoff depth was estimated using SWAT model. The statistical performance of the model in estimating the runoff was revealed with coefficient of determination (R2) of 0.81 and Nash–Sutcliffe Efficiency (NSE) of 0.76 for monthly calibration and R2 of 0.79 and NSE of 0.72 for monthly validation periods. The result implied that there's adequate runoff water to be conserved. Combination of hydrological model with GIS and RS was found to be a vital tool in estimating rainfall runoff and mapping suitable water harvest home sites.