This paper deals with the combination of web based quality assurance and data fusion processes to support mutual validation of crowdsourced observations and authoritative data. Results stem from the EU FP7 funded Citizen Observatory Web (COBWEB) project. COBWEB targets new methods and tools for collecting Citizen Science and Crowdsourced data for policymaking. Major issues relate to uncertainty in the quality and accuracy of such data. We present a design for a framework and an implementation that incorporates techniques to qualify and add value to crowdsourced data with a view to addressing some of the uncertainty and isolation issues with these sources. We show an example of a Citizen Observatory use case that utilizes data on the distribution of Japanese Knotweed collected by volunteers via a mobile app. Following a service-oriented design, the approach offers generic components which could also be transferred to similar applications for spatial data validation.