“…To set the background, we recall that over the last decades considerable work has been done on extending classical statistical methods to the case of data embedded in functional Hilbert spaces (see, e.g., Ramsay and Silverman, 2005 , Ferraty and Vieu, 2006 , Horváth and Kokoszka, 2012 , Wang et al, 2016 and references therein). Moreover, an important body of literature has focused on developing a consistent theoretical framework for geostatistical modeling and spatial prediction for functional data, developing the infinite-dimensional counterparts of techniques such as spatial variography and Kriging ( Giraldo et al, 2010 , Nerini et al, 2010 , Giraldo et al, 2011 , Ruiz-Medina, 2012 , Menafoglio et al, 2013 , Caballero et al, 2013 , Ignaccolo et al, 2014 , Menafoglio and Petris, 2015 , Menafoglio et al, 2016b , Menafoglio et al, 2016a ). Recent works have specifically targeted the case of constrained functions and other object data, in a stream of literature that we refer to as Object Oriented Spatial Statistics (O2S2) ( Menafoglio and Secchi, 2017 and references therein).…”