Diagnostic Evaluation Of Waterlogged Archaeological Wood From "Nanhai No.1" Wooden Shipwreck

Comprehensive and accurate evaluation of the preservation state of waterlogged archaeological wood（WAW）is a fundamental study for the research fields of wood science and cultural heritage conservation.This thesis was focused on the WAW samples collected from an ancient Chinese shipwreck Nanhai No.1.The decay class of WAW was graded through wood anatomical,physical,and chemical diagnostic analysis.The mechanism of anatomical and chemical changes in the cell wall was then investigated based on different preservation states of the WAW samples.Furthermore,non-destructive and real-time monitoring of WAW moisture content（MC）was realized through hyperspectral imaging（HSI）method.This study explores the multidisciplinary diagnostic methods for the scientific evaluation of the preservation state of WAW and reveals the effecting mechanism of alteration in the cell wall structure on the properties of WAW.The outputs of this work can provide a theoretical basis and technical support for the conservation of wooden cultural heritage from the marine context.The main conclusions of the study are listed as follows:（1）A total of 7 wood species were identified from 29 WAW samples,including 3 softwood and 4hardwood species,and the wood species of the main component of the shipwreck was softwood Pinus massoniana.（2）The decay class of the samples was graded as moderate and severe decay according to the anatomical features.Based on the physical analysis,the maximum water content（MWC）of the severely decayed samples was higher than 500%and the MWC of the moderate decayed samples was lower than250%.In tandem with hierarchical cluster analysis（HCA）and principal components analysis（PCA）,FTIR measurements classified four clusters of samples,including groups of anomalous samples,moderately decayed softwood,severely decayed softwood,and severely decayed hardwood.The consistence results of decay grading were obtained through wood anatomical,physical and chemical analysis.（3）The erosion decay（EB）was evident by wood anatomical analysis in Nanhai No.1 wood samples The specific feature of EB decay in the softwood transverse section showed the sound cell walls adjacent to degraded ones forming a distinct chequered degradation pattern.With the EB attack,the S_2L layer of the tracheid cell wall remained intact in compression wood.（4）The comparative analysis of the anatomical structure of the sound wood（SW）,moderated decayed samples（K1NH7 and K1NH13）,and severely decayed samples（101NH and 107NH）of Pinus massoniana indicated the main reason for the better-preservation of compression wood was due to the high lignification of S_2Llayer.On the other hand,the secondary cell wall of severely decayed samples showed an apparent“spongy”feature in structure with the only intact middle lamellae remaining.The result of the evaluation of the preservation state was SW（sound wood）>K1NH7（moderate）>K1NH13（moderate）>101NH（severe）=107NH（severe）.（5）The analysis of the scanning electron microscope and energy dispersive spectrometer（SEM-EDS）indicated that the sulfur element was widely distributed in the wood cell wall and compounds of sulfur and iron were deposited in wood tissue and cell with different morphologies.（6）The chemical alteration of the degraded wood cell wall in different preservation states of Pinus massoniana showed the selective depletion of polysaccharides,depolymerized cellulose molecule,damage to crystal structure,and the break ofβ-O-4 chemical bond of lignin with the increase of decay.Combined with the MWC analysis the result of the evaluation the preservation state was SW（sound wood）>K1NH7（moderate）>K1NH13（moderate）>107NH（severe）>101NH（severe）.（7）The study on hygroscopicity properties indicated that the inorganic deposition in the moderate decay sample K1NH7 deeply influenced the hygroscopicity of wood cell walls and much more moisture adsorption sites were exposed in the moderate decay samples K1NH13 due to the degradation of polysaccharide.For severely degraded samples,the moisture adsorption of the monolayer was lower than SW and significant condensation of capillary moisture appeared when the relative humidity was higher than 70%.（8）Hyperspectral imaging（HSI）as a non-destructive method was used to real-time monitor the moisture content change of WAW.A linear algorithm named partial least square regression（PLSR）and a nonlinear machine learning algorithm named random forests regression（RFR）were successfully used to establish models with high prediction capacity of the changes in moisture content.