Data-driven logistic-based weighting of geochemical and geological evidence layers in mineral prospectivity mapping

Prospectivity analysis of orogenic gold deposits in Saqez-Sardasht Goldfield, Zagros Orogen, Iran

“…Carranza) Manuscript submitted to Ore Geology Reviews 9 spatial evidence using different i and s values were pointed out by Yousefi and Nykänen (2016 Masters, 1993;Zadeh, 1993;Tsoukalas and Uhrig, 1997;Burrough and McDonnell, 1998;Kemp et al, 2001; Nykanen et al, 2008a,b; Almasi et al, 2015a,b; Yousefi and Carranza, 2015a,b,c;Yousefi and Nykänen, 2016).…”

“…Then, we trialed different types of logistic functions, namely large, small, and near, that were proposed by Tsoukalas and Uhrig (1997) and applied by Yousefi and Carranza (2015a,b,c, 2016b) and Yousefi and Nykänen (2016), in order to quantify diverse relationships between spatial evidence values and the deposit-type sought.…”

“…To reduce bias in the assignment of weights to continuous-value spatial evidence, various researchers (e.g., Nykänen et al, 2008a;Yousefi et al, 2012Yousefi et al, , 2013Yousefi et al, , 2014Yousefi and Carranza, 2015a, b, c, 2016a;Yousefi and Nykänen, 2016) have applied logistic functions to assign fuzzy weights to indicator features without using locations of KMOs and without discretization of evidential values into some arbitrary classes based on expert opinion. While this practice overcomes imprecise evaluation of the relative importance of evidential values, as portrayed by simplification and discretization of continuous-value evidential data into some arbitrary classes, it is also subjective because using a single logistic function for weighting spatial evidence values does not consider the fact that diverse deposit-types or mineral systems form by diverse geological processes.…”

“…The weights assigned to classes of discretized evidential values may be based on (a) expert judgment directly, (b) locations of known mineral occurrences (KMOs), (c) a combination of (a) and (b), or (d) subjectively-defined functions, so indirectly-assigned by analyst (e.g., Luo, 1990;Bonham-Carter, 1994;Cheng and Agterberg, 1999;Luo and Dimitrakopoulos, 2003;Porwal et al, 2003a Porwal et al, ,b,c, 2004Porwal et al, , 2006Carranza et al, 2005;Carranza, 2008Carranza, , 2014Porwal and Kreuzer, 2010;Mejía-Herrera et al, 2014;Carranza and Laborte, 2016;McKay and Harris, 2016). All these methods impart bias due to discretization of continuous spatial values, use of subjective expert judgments, and sparse or incomplete data on locations of KMOs in knowledge-and data-driven MPM (Coolbaugh et al, 2007;Lusty et al, 2012;Ford et al, 2016).To reduce bias in the assignment of weights to continuous-value spatial evidence, various researchers (e.g., Nykänen et al, 2008a;Yousefi et al, 2012Yousefi et al, , 2013Yousefi et al, , 2014Yousefi and Carranza, 2015a, b, c, 2016a;Yousefi and Nykänen, 2016) have applied logistic functions to assign fuzzy weights to indicator features without using locations of KMOs and without discretization of evidential values into some arbitrary classes based on expert opinion. While this practice overcomes imprecise evaluation of the relative importance of evidential values, as portrayed by simplification and discretization of continuous-value evidential data into some arbitrary classes, it is also subjective because using a single logistic function for weighting spatial evidence values does not consider the fact that diverse deposit-types or mineral systems form by diverse geological processes.…”

“…Therefore, understanding mineral system processes of the deposit-type sought in the area facilitates eliciting of exploration features to be used for MPM. In this regard, diversity of relationships between various mineral deposits and their corresponding spatial evidence values results in different conceptual models of mineral prospectivity.Thus, translation of the parameters of a conceptual model, which derived from understanding mineral system of the corresponding deposit-type sought, to weighted evidence layers is a critical undertaking There are various type of uncertainties that adversely affect prospectivity analysis of mineral deposits (e.g., McCuaig et al, 2010;Yousefi and Carranza, 2015a), of which the noteworthy are: (a) uncertainty due to missing evidence (Zuo et al, 2015); (b) uncertainty resulting from imprecise weighting to spatial evidence values (e.g., Nykänen et al 2008a;Yousefi and Nykänen, 2016;Yousefi, 2017); (c) uncertainty in selection of subjectively-defined functions to be used for weighting spatial values (Luo, 1990;Luo and Dimitrakopoulos, 2003); (d) uncertainty due to complexity of geological setting (Van Loon, 2002) and anomaly patterns (Cheng, 2007;Yousefi et al, 2013); (e) uncertainty due to dissimilarities of geological settings (McCuaig et al, 2010;Lisitsin et al, 2013); (f) uncertainties due to incomplete knowledge in the understanding of mineral system processes (McCuaig et al, 2010); (f) uncertainty due to complex relationships between indicator features and mineral deposits (e.g., Yousefi et al, 2013), and (g) uncertainty due to inaccurately and imprecisely Prospectivity analysis of orogenic gold deposits in Saqez-Sardasht Goldfield, Zagros Orogen, Iran (by A. Almasi, M. Yousefi, E. J.M. Carranza) Manuscript submitted to Ore Geology Reviews 17 presenting of geological features in exploration datasets (Lisitsin et al, 2013).…”