Advances in biocultural geography of olive tree (Olea europaea L.) landscapes by merging biological and historical assays

Sample collection

Eighteen samples of mono-varietal Ogliarola campana EVOO were directly collected in Campania region (Southern Italy), among olive production areas ascribed to four regional PDOs (Fig. 1): Irpinia (IR, 7 samples), Colline Salernitane (CS, 4 samples), Penisola Sorrentina (PS, 5 samples) and Cilento (CI, 2 samples). The unbalanced experimental design was constrained by local and uncontrolled factors, such as the uneven number of farms with presence of the cultivar and the varying willingness of producers to be recruited in the study. The Ogliarola campana cultivar (also referred to as Ogliarola) is also known as Uogliarola and Minucciola in the Sorrento peninsula (Penisola Sorrentina) and as Ogliara in Cilento. The samples of EVOO were individually collected from olive mills during 2013 harvest and each sample was representative of some farms located in the surrounding area. The mono-varietal identity of Ogliarola campana was verified in the orchards where the olives came from, by morphological identification of olive trees, with the support of local agronomists. Three random samples of EVOO were collected from each olive mill and delivered to the analytical laboratories in bottles of dark glass, then stored at room temperature for some weeks, until analysis. The geographic coordinates of each EVOO sample were referred to the olive mill position.

Determination of carbon isotope and elemental compositions

The analysis of stable carbon isotopes was performed with a CF-IRMS (Isoprime GV, Elementar UK Ltd, Cheadle, UK) coupled with an elemental analyser (NA1500; Carlo Erba, Milan, Italy). In details, sub-samples of about 500 μg of oil were quantitatively combusted in the elemental analyser. The resulting CO₂ was admitted through helium continuous flow to the isotope ratio mass spectrometer. The determination of the isotope ratios (R = ¹³C/¹²C) of both samples and internal standards allowed the calculation of the carbon isotope composition (δ¹³C) values of the sample, anchored to the reference scale of IAEA standard VPDB. Both gaseous (RM 8562, RM 8563, RM 8564) and solid IAEA standards (NBS-22 fuel oil and IAEA-CH6 Sucrose) were used for pristine determinations of the internal standards on the VPDB scale. The isotopic determinations were then expressed as δ notation, which was calculated as the relative deviation of the isotope ratio of a sample from that of the international standard, using the expression: δ = (R_s– R_std)/R_std, where R_s is the isotope ratio of the sample and R_std is the isotope ratio of the international standard. The standard deviation of replicate measurements was ± 0.1‰. The obtained δ¹³C values, as usual, were multiplied by 1000 and shown in “per mil” units (‰).

The analysis of elemental composition was performed with a ICP-OES (optical emission spectrometry) iCAP 7200 instrument (Thermo Scientific Inc., Waltham, MA, USA), equipped with an ASX-520 auto sampler (Cetac Technologies Inc., Omaha, NE, USA). The samples were prepared and analysed according to the method of Camin et al.^9,10. Briefly, 15 g of oil were extracted with 10 ml of extracting solution, which was prepared with 6,7% H₂O₂, 1% HNO₃ and 0,2% HCl, in ultrapure water. The elemental analysis was carried out using radial and axial acquisition of the elements. The trace elements evaluated in the analysis were among the most frequently detected in olive oil, according to Beltran et al.¹⁴. The limit of detection (LOD) of each element was calculated as three times the standard deviation of the signal recorded in ten replicates of the blank samples. The LOD ranged from 0.06 µg/kg (Ba) to 45.65 µg/kg (Na). The relative standard deviation (RSD%) of the analytical method was calculated on five replicated extractions and analyses of an oil sample. The RSD% ranged from 2% (Ca) to 36% (Al).

Climatic and geographic data

The database of the regional agricultural service (Regione Campania, Assessorato Agricoltura, Italy) was used to obtain climatic data from meteorological stations representative of the four production areas: Mirabella Eclano (IR), Forio d’Ischia (PS), Battipaglia (CS) and Stella Cilento (CI). The xerothermic index X_i^28,29 of each site was calculated using the formula:

where, T_med is the monthly mean temperature and P is the monthly precipitation in mm.

The local climatic data of the sampling areas were interpolated from Worldclim datasets (www.worldclim.org). Other GIS layers include the administrative boundaries from Regione Campania (https://sit2.regione.campania.it/content/ctr), the digital elevation model and the aerial photographs from Ministero dell’Ambiente (http://www.pcn.minambiente.it). The Qgis GIS software⁵² was used to retrieve latitude, longitude, altitude and distance from the sea of the sampling sites. The geological characterization of the soils was derived from Carta Geologica d’Italia 1:50000, sheets: 433, 466, 467 and 503, issued by ISPRA⁵³. The setting of a final biocultural map, in accordance with the previous evaluation and validation of the variables utilized, was performed in R⁵⁴, DataGraph form Visualtools (https://www.visualdatatools.com/DataGraph/) and finally Qgis⁵².

Cultural and historical analysis

We set up a method to analyse toponymy. Some anthropic and cultural elements related to the olive tree were taken as biocultural indicators, assuming that they are preserved in a semantic frame. This method combines information from medieval sources and from toponymy, in order to characterize the contemporary biocultural landscapes. Starting from the etymology of the word Ogliarola, we carried out a bibliographic search in the agricultural contracts of the Middle Ages within the current production areas of Ogliarola in Campania. In some cases, we also examined the relevant terms occurring in the history of certain areas (e.g., Montis Corvini) that do not correspond to the current localization of olive groves. This historical period is testified by an abundant written documentation about agricultural landscapes. The keywords and word roots utilized to investigate the toponyms of vegetation (phyto-toponymy)²⁵ were related to Ogliarola (Ogli-) and to the olive plant (oliv-, olib-, olev-, etc.) (see the etymology by Rhizopolou⁵⁵). We also investigated the terms related to practices of olive cultivation and oil production. For instance, the toponyms Zapino and Torchiara are referred to the orchard cultivation practices and to the oil extraction method, respectively.

The Codices, i.e. the documental manuscripts collected in a book of agricultural and other kind of contracts, were also examined: Codex Diplomaticus Verginianus (CDV, referring to Irpinia), Codex Diplomaticus Amalfitanus (CDA, referring to Penisola Sorrentina), Codex Diplomaticus Cavensis (CDC, referring to Colline Salernitane and Cilento) and Codice Solothurn (CoS, referring to Colline Salernitane and Cilento). In the case of CDV, the comprehension of historical sources was supported by the librarians of the Biblioteca del Monumento Nazionale di Montevergine (Mercogliano, Italy). Several original texts were found in the volume of Filangieri di Candida⁵⁶. On-line resources provided by ALIM (Archivio Latinità Italiana nel Medioevo) project were utilized for the reading and interpretation of CDC.

The historical search was limited to the sources dating back to the medieval period, from IX to XIV century Current Era (CE). A cultural-historical database was set up for each current production area of Ogliarola and allowed the count of the following elements, also included in the evaluation of the terroir score:

• agricultural contracts,

• historical toponyms,

• historical places with past presence of olive orchards,

• present toponyms,

• medieval cities or stations (stationes),

• medieval structures for care of poor people (hospitia et hospitalia).

We considered the present toponyms when referred to practices of cultivation specifically associated to olive orchards (e.g., Torchiara and Sanza, close to Salerno) but we ignored those related to generic agricultural practices (e.g., Pastena, from the Latin verb pastenare, which means to plough). Furthermore, we took into account all the toponyms tightly linked to the cultivar Ogliarola.

Statistical analysis and Score evaluation

Quantitative analytical variables (elemental composition, climatic data, stable isotope composition) were tested for normality of data distribution by Shapiro-Wilk normality test using R⁵⁴. Data distribution was considered significantly deviating from normality at p-value <0.01. The normally distributed variables were further analysed by ANOVA and Fisher’s Least Significant Difference (LSD) post-hoc test using R. The variables with not normal distribution were analysed by the non-parametric Kruskal-Wallis rank sum test using R. Principal Component Analysis (PCA) was performed on quantitative data and on the historical dataset using R and FactoMineR package⁵⁷. Samples and variables with missing data were excluded a priori from the PCA.

In order to obtain a single index representing both anthropic data and analytical measurements on Ogliarola EVOOs, we propose an original score evaluation method, based on the relative counting for the anthropic observations and on the Shannon’s information for the physical variables. A detailed description of this method, with literature citations, is provided in the Supplementary information 2, which also includes a Python script of the algorithm. Here we present the principle and a simplified description of the main steps. Briefly, we reduced the bulk of data to eight variables only, scaled within the 0…1 range (the greater the better). The values were plotted along the arms of a radar plot, then a synthetic score was evaluated, according to the area spanned by the plot. We applied the score evaluation procedure to the anthropic and physical variables:

Anthropic variables: the six variables (number of observations) are scaled to 0…1, according to the maximum value of the variable in all the PDOs, dividing the number of observations of a given variable by the maximum one. Table 3 contains, for each PDO, the number of observations and the scaled value. For ease of reading of the radar plots, each variable has been assigned an abbreviation label, defined in Table 3.

Physical variables: Each PDO has a distribution of the measurements over a certain number of Ogliarola EVOO samples. The information content of such distribution is evaluated in terms of Shannon’s information or syntropy, a “negative version” of the entropy: a higher typicality of the values is associated to a higher syntropy^58,59. We selected the δ¹³C (treated as a single variable) and the elemental composition (averaged over all the elements) as relevant physical variables. We chose these variables because the δ¹³C is correlated to plant physiology and is an indicator of the climatic, geomorphologic and agricultural conditions. The presence of trace elements is also related to the plant cultivation environment and to the management practices. Given the large number of the elemental measurements, we summed up these variables, averaging the respective syntropies.

The final score, named terroir score, was evaluated according to the area of the radar plot (Fig. 1). Each arm of the plot represents the value of a variable. The plot area is scaled with respect to the maximal regular polygon (an octagon in this case), so that if all the variables relative to a PDO have a value of one, the scaled area measures exactly one. On the other hand, if the radar plot area is shrunk (all or nearly all the variables with zero-value) the area is close to zero. The score is finished by a logarithmic transformation to compensate saturation of the highest values and expressed in percent of the maximum allowed area. Due to the experimental design of our case study, without replicated samples in each PDO area, it was not possible to estimate a confidence interval of the terroir score. In order to assess the score reliability, we conducted a sensitivity analysis with an ensemble of Gaussian perturbations to a set of simulated measurements (Supplementary information 2). The graphical output of the radar plots was done using DataGraph form Visualtools (https://www.visualdatatools.com/DataGraph/). The terroir score and the associated plot characterize the biocultural profile and the typicality of a given production area.

Source: Ecology - nature.com