11 I. Data integration and comparison in landscape archaeology: towards analysis beyond sites and valleys Peter Attema, University of Groningen Summary I discuss the importance of data integration in landscape archaeology, taking into account issues of scale, feasibility and relevance. Referencing a current initiative – the Rome Hinterland Project (RHP) – undertaken by an international consortium of researchers to combine landscape archaeological data from field surveys in the landscapes around Rome, I highlight the potential of the archaeological record of the Marche for the integration of field data from landscape survey. I discuss the example of, and progress made within, the RHP to illustrate the feasibility of data integration. Introduction Many landscape archaeologists nowadays record, process and analyse large amounts of site and artefact data to discover patterns in settlement and land use dynamics in their fieldwork areas, so as to study the longue durée of landscapes. Systematic recording of surface archaeology in the Mediterranean has a long research history, rooted in the processual paradigm of the 1960s and 1970s. From the beginning, its practitioners have been keen on standardization of field and recording procedures, and on the validation and representativeness of data obtained from the field. There has also been a continual interest in studying the multiple biases that affect the surface archaeological record (van Leusen 2002) and the specific regional traditions of field survey (see for Italy especially Cambi and Terrenato 1994; Terrenato 1996). The attitude of practitioners from the current generation of landscape archaeologists, who criticise the sources of their own data and that of data from other researchers, has led to exploration of the potential for comparison of datasets obtained from systematic surveys and topographical work, and its potential for comparison and integration (Alcock and Cherry 2004; Attema et al. 2010; Witcher 2008). Key to the success of comparison and integration is formal classification according to set criteria, a condition that is rarely met in case studies presented in the literature. If undertaken, however, it allows aggregate and comparative analyses of archaeological sites and finds recorded in the landscape (Attema et al. forthcoming). I use the example of a recent initiative by an international consortium to build an overarching database – the Rome Hinterland Project Database (RHPdb) – which aims at the integration of three major survey projects based in the hinterland of Rome, in which large amounts of field and artefact data have been collected since the 1960s: the Tiber Valley Project (TVP) (Patterson et al. forthcoming), the Suburbium project (Carafa and Capanna 2019) and the Pontine Region Project (PRP) (Attema et al. 2019). Central to the structure of the RHPdb is the classification table for sites according to function and the classification table of ceramics according to type, and their associated dates. The uniform classification of sites and artefacts is a precondition for merging data to perform aggregate and comparative analyses. Judging from the landscape archaeological papers presented at the conference and in this volume on the (surface) archaeology of the landscapes of the Picenum and ager Gallicus, the intensity and quality