Modelling dendro-anthracological parameters with dendrochronological reference datasets: Interrogating the applicability of anthraco-typology to assess Aleppo pine (Pinus halepensis Miller) wood management from archaeological charcoal fragments

TL;DR: In this paper, the authors adapted the anthraco-typological methodology to conifer wood anatomy and evaluated its applicability on the archaeological charcoal assemblages of Aleppo pine, through complete dendrological analysis of the referential material.

About: This article is published in Journal of Archaeological Science.The article was published on 2020-12-01 and is currently open access. It has received 4 citations till now. The article focuses on the topics: Anthracology & Aleppo Pine.

Summary

Introduction

• Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation Highlights Dendro-anthracological and anthraco-typological methods are tested for Pinus halepensis.
• In order to distinguish between small fragments deriving from branches and trunks, 25 radial growth (ring width) is also measured.
• This allowed for the 5 application of the anthraco-typological principles to the obtained dendro-anthracological datasets 6 according to the results of the dendrochronological referential.

2.2. Archaeological charcoal assemblages analysed 13

• The anthraco-typological approach was applied to five different assemblages of dispersed charcoal 14 fragments from three sites in Mallorca, dating from the Late Bronze Age to the Late Iron Age/Early 15 Roman period .
• As a consequence, the pine fragments may potentially represent both firewood 21 and/or timber remains and be subsequently overrepresented in the anthracological spectra.
• This is 22 the case of the two assemblages from Building 51 in Ses Païsses – one corresponding to the habitat 23 level and one to the roof collapse – and the assemblage of the Talaiot 3 in Son Fornés, a tower-24 shaped public monument inside the village.
• 25 The other two assemblages originate from archaeological deposits formed after a deliberate 26 abandonment.
• The assemblage from the interior of Building 25 in Ses Païsses corresponds to 27 dispersed charcoal fragments linked to metallurgical activity: firewood residues covering a 28 relatively long time span.

2.3.1. Measurement of the charcoal-pith distance 2

• Different methodologies have been proposed to calculate the distance between the charcoal 3 fragment and the pith of the trunk/branch from where it originated and to infer a minimum 4 diameter for the original wood ( x2) (see a detailed analysis in Dufraisse et al. 2019).
• Angle between the two rays <2º were excluded; 30 and the correction factors applied were: 31 iii).
• No reliable measurements of charcoal-pith distance >12.5 cm were considered; 34 v) A correction factor taking into account wood shrinkage during carbonisation (20%) was 35 applied to the measurement of the charcoal-pith distance.
• An average value for the ring 25 width of all the fragments for every projected diameter class in each assemblage was also 26 calculated.

2.4 Anthraco-typology: theoretical principles and referential model 30

• In order to easily compare the dendro-anthracological data between the archaeological charcoal 31 fragments and the referential model, the principles of anthraco-typology (i.e. the combination of the 32 dendro-anthracological parameters) were applied (Dufraisse et al. 2018a; Dufraisse and Coubray 33 2018).
• Considering the wood anatomy of gymnosperms and the combination of two dendro-34 anthracological parameters (charcoal-pith distance and growth rate), a total of four anthraco-groups 35 were defined based on two different variables of the charcoal fragments: projected diameter classes 1 and average ring width .
• 4 Moreover, the dendrochronological referential of the Aleppo pine in Mallorca allows the 5 interrogation of different growth rates for both organs, trunks and branches, discriminating between 6 fast and slow growth (see section 3.1).
• Accordingly, by combining these two parameters, measured 7 both on the referential and on the archaeological materials, four anthraco-groups were considered as 8 discriminating between small and large diameters (</> 7 cmn) and fast and slow growth (</> 9 threshold established after the dendrochronological referential) .

3.1.1 Growth trends of Aleppo pine populations in Mallorca 18

• The trunk cores of the three stations presented similar growth trends, characterised by fast or very 19 fast growth in juvenile trees that is progressively reduced due to the age effect .
• They presented different ring-width values.
• 7 8 Figure 5: Results of the dendrochronological measurements of trunk cores and branches from 9 referential stations: a) ring width vs. cambial age; b) ring width vs. diameter classes.
• Considering these results, ring width was also plotted on samples above and 19 under 7 cm of diameter for both trunks and branches .
• There is a quantitative difference in the radial growth of trunks and branches for Aleppo 35 pine that clearly defines differentiated growth rates for both organs of the tree: < 1.5 mm for 1 branches and > 1.5 mm for trunks.

3.2. Anthraco-typology: establishment of the anthraco-groups according to the 4

• Minimum diameter estimation results (charcoal-pith distance and ADmodel) 23.
• The 24 percentage of fragments corresponding to each diameter class are expressed both before and after 25 the application of the ADmodel.
• It is significant to remark that the trends shown by both values did 26 not present significant variances .
• The Iron Age assemblages of Building 25 (Ses Païsses) and Talaiot 3 (Son Fornés) also 6 show the exploitation of small woods, although the projected diameters are oriented towards larger 7 values, between 4 and 7 cm.

3.3.2. Radial growth results (ring width and diameter classes) 12

• The results obtained by measuring the average ring width of the fragments of every diameter class 13 were variable for each assemblage .
• In Building 51 (Ses Païsses) both assemblages 14 (occupation layer and faller roof) show remarkably narrow ring widths, mostly under 1 mm or less, 15 in all diameters classes.
• In the Iron Age assemblages from Building 25 (Ses Païsses) and Talaiot 3 16 (Son Fornés), ring width is remarkably higher (more than double) than in Building 51 and clearly 17 above the established 1.5 mm threshold.
• Finally, the ring width measured at Na Galera is variable 18 along the different diameter classes: it is low and predominantly under the 1.5 mm threshold for 19 diameter classes smaller than 7 cm; only a few fragments with diameters above a 7 cm ring width 20 clearly exceed this value.
• The results of the charcoal-pith distance and the ring width measurements were sorted, according to 24 the anthraco-typological principles and the definition of the anthraco-groups, and reviewed against 25 the results obtained from the referential study .

4.1. Dendrochronological reference dataset: modelling of the dendro-anthracological 8

• Two different referential datasets were created from the pine populations in Mallorca – one 10 reflecting radial growth in pine trunks, the other comparing the radial growth of trunk and branches 11 in the same individuals.
• In all cases, the radial growth of branches was significantly lower than that 25 of trunks for the same cambial age, a pattern that is also clear when replacing the cambial age for 26 the diameter classes .
• The applicability of anthraco-typology to Aleppo pine archaeological charcoal fragments 15 Considering the results of the referential dendrochronological model, the authors can see that the 16 applicability of the dendro-anthracological parameters and tools and the anthraco-typology 17 principles offer a relevant opportunity to better interrogate the Aleppo pine archaeological charcoal 18 fragments to identify past uses of this tree.
• Anthraco-groups 3 (large diameter >7 cm and 16 slow growth <1.5 mm, representing the external part of mature trunks) and 4 (large diameter >7 cm 17 and fast growth >1.5 mm, also representing the external part of mature trunks) presented minor 18 values in both assemblages.
• The fact that anthraco-group 4 is 27 only present in the roof layer and that the values of anthraco-group 2 are higher here, suggests that 28 trunks would have been used as construction timber, burnt after the fire event that destroyed the 29 building, rather than as fuel (which, in the case of pine, would focus on branches).

5. Conclusions 17

• This study shows that the anthraco-typology approach created for temperate angiosperms with ring-18 porous wood anatomy (Quercus petraea/robur, Dufraisse et al. 2018a) is applicable to 19 Mediterranean gymnosperms (Pinus halepensis), provided that an appropriate dendrochronological 20 referential is built.
• This study, 3 therefore, confirms that the development of dendrochronological referential datasets for different 4 species and regions is needed to adapt the anthraco-typological principles to every tree species.
• In the case of the dump at Na Galera, anthraco-typology suggests 17 two potentially different origins for the pine charcoal fragment assemblage: firewood debris 18 deposited after the cleaning of the hearth, and charcoal fragments derived from the burnt and 19 fragmented timber from a neighbouring building destroyed by fire.

Figures

Table 4: Results of the dendro-anthracological parameters measured on Aleppo pine archaeological charcoal assemblages (projected diameter classes and average ring width) according to anthraco-typology principles.

Table 3: Results of the charcoal-pith distance measurements expressed by diameter classes and modelled according to the ADmodel.

Table 2: Description of the archaeological charcoal assemblages analysed

Table 1: Description of the forest stations sampled for the referential dendrological study

Frequently asked questions

Q1. What contributions have the authors mentioned in the paper "Modelling dendro-anthracological parameters with dendrochronological reference datasets: interrogating the applicability of anthraco-typology to assess aleppo pine (pinus halepensis miller) wood management from archaeological charcoal fragments" ?

In this study the authors have adapted the anthraco-typological methodology to conifer wood anatomy and evaluated its applicability on the archaeological charcoal assemblages of Aleppo pine, through complete dendrological analysis of the referential material. Analysis of this dataset has proved that anthraco-typological methods are applicable to Mediterranean conifers and to further interrogate pinewood exploitation in prehistoric Mallorca. 

Q2. What future works have the authors mentioned in the paper "Modelling dendro-anthracological parameters with dendrochronological reference datasets: interrogating the applicability of anthraco-typology to assess aleppo pine (pinus halepensis miller) wood management from archaeological charcoal fragments" ?

In this case, the possibility to characterise the presence/absence of firewood and/or timber 11 remains allows us to explain the formation of the archaeological layers and identify different wood 12 uses for the same species. 29 Finally, it is important to note that after dendrochronological reference datasets and the testing of 30 applicability of anthraco-typology, further studies look to improve the interrogation of the 31 anthracological record. The development of dendro-anthracological datasets will allow us to 32 interrogate the impact of wood exploitation on past woodlands and offer new insights to identify 33 more suitable fragments for anthraco-climatic analyses ( or select the most reliable charcoal 34 fragments for isotopic analysis ) contributing to the future development of archaeological wood 35 charcoal analysis. In the case of the dump at Na Galera, anthraco-typology suggests 17 two potentially different origins for the pine charcoal fragment assemblage: firewood debris 18 deposited after the cleaning of the hearth, and charcoal fragments derived from the burnt and 19 fragmented timber from a neighbouring building destroyed by fire. 

Q3. How many dispersed charcoal fragments were found in the interior of the building?

The assemblage from the interior of Building 25 in Ses Païsses corresponds to 27 dispersed charcoal fragments linked to metallurgical activity: firewood residues covering a 28 relatively long time span. 

Q4. How many diameters are shown in the assemblages of building 51?

In Building 51 (Ses Païsses) both assemblages 14 (occupation layer and faller roof) show remarkably narrow ring widths, mostly under 1 mm or less, 15 in all diameters classes. 

Q5. What evidence is there that fuel debris was cleaned on a regular basis?

the fact that big articulated charcoal 5 fragments were detected during the excavation of the pit, together with the clear predominance of 6 branch exploitation attested by anthraco-typology, suggests that fuel debris was cleaned on a 7 frequent basis and that once deposited in the pit charcoal suffered little mechanical disturbance. 

Q6. What was the method used to obtain the distance and angle between two ligneous rays?

The measurements of the distance and the angle between 14 two ligneous rays were obtained using a semi-automatic system based on four landmarks integrated 15 into the Nikon NIS Elements software.