Rute Coimbra

Regression trees for modeling geochemical data-An application to Late Jurassic carbonates (Ammonitico Rosso)

Research based on ancient carbonate geochemical records is often assisted by multivariate statistical analysis, among others, used for data mining. This contribution reports a complementary approach that can be applied to paleoenvironmental research. The choice to use a machine learning method, here regression trees (RT), relied in the ability to learn complex patterns, integrating multiple types of data with different statistical distributions to obtain a knowledge model of geochemical behavior along a paleo-platform.The Late Jurassic epioceanic deposits under scope are represented by six stratigraphic sections located in SE Spain and on the Majorca Island. The used database comprises a total of 1960 data points corresponding to eight variables (stable C and O isotopes, the elements Ca, Mg, Sr, Fe, Mn and skeletal content). This study uses RT models in which the predictive variables are the geochemical proxies, whilst skeletal content is used as a target variable. The resulting model is data driven, explaining variations in the target variable and providing additional information on the relative importance of each variable to each prediction, as well as its corresponding threshold values.The obtained RT revealed a structured distribution of samples, organized either by stratigraphic section or sets of nearby sections. Averaged estimated skeletal abundance confirmed the initial observations of higher skeletal content for the most distal sections with estimated values from 18% to 27%. In contrast, lower skeletal abundance from 5% to 15% is proposed for the remaining sections. The geochemical variable that best discriminates this major trend is ?18O, at a threshold value of -0.2?, interpreted as evidence for separation of water-mass properties across the studied areas. Other four variables were considered relevant by the obtained decision tree: C isotopes, Ca, Sr and Mn, providing new insights for further differentiation between sets of samples.

Disentangling shallow-water bulk carbonate carbon isotope archives with evidence for multi-stage diagenesis: An in-depth component-specific petrographic and geochemical study from Oman (mid-Cretaceous)

Disentangling shallow‐water bulk carbonate carbon isotope archives into primary and diagenetic components is a notoriously difficult task and even diagenetically screened records often provide chemostratigraphic patterns that significantly differ from global signals. This is mainly caused by the polygenetic nature of shallow‐water carbonate substrates, local carbon cycle processes causing considerable neritic–pelagic isotope gradients and the presence of hiatal surfaces resulting in extremely low carbonate preservation rates. Provided here is an in‐depth petrographic and geochemical evaluation of different carbonate phases of a mid‐Cretaceous (Barremian–Aptian) shallow‐water limestone succession (Jabal Madar section) deposited on the tropical Arabian carbonate platform in Oman. The superposition of stable isotope signatures of identified carbonate phases causes a complex and often noisy bulk carbon isotope pattern. Blocky sparite cements filling intergranular pores and bioclastic voids evidence intermediate to (arguably) deep burial diagenetic conditions during their formation, owing to different timing or differential faulting promoting the circulation of fluids from variable sources. In contrast, sparite cements filling sub‐vertical veins reveal a rock‐buffered diagenetic fluid composition with an intriguing moderate enrichment in 13C, probably due to fractionation during pressure release in the context of the Miocene exhumation of the carbonate platform under study. The presence of abundant, replacive dedolomite in mud‐supported limestone samples forced negative carbon and oxygen isotope changes that are either associated with the thermal breakdown of organic matter in the deep burial realm or the expulsion of buried meteoric water in the intermediate burial realm. Notwithstanding the documented stratigraphically variable and often facies‐related impact of different diagenetic fluids on the bulk‐rock stable isotope signature, the identification of diagenetic end‐members defined δ13C and δ18O threshold values that allowed the most reliable ‘primary’ bulk carbon isotope signatures to be extracted. Most importantly, this approach exemplifies how to place regional shallow‐water stable isotope patterns with evidence for a complex multi‐stage diagenetic history into a supraregional or even global context.