Forest inventory and palynological analysis are complementary methodologies for vegetation sampling in an archaeological site of Central BrazilAbstract
Archaeological sites are under threat due to the expansion of agriculture and livestock over the Cerrado, currently restricted to narrow strips of vegetation. Palynological research provides a systemic view of plant diversity and contributes to the analysis of anthropic processes. Forest inventories associated with the characterization of pollen spectra in regions with archaeological sites are fundamental for the conservation of natural vegetation and the protection of shelters. This study aimed to conduct palynological identification and a forest inventory in an archaeological site in the Serranópolis region, southwestern Goiás. The objective was to understand the relationship between native vegetation and the current anthropization of the region surrounding the archaeological site. In the forest inventory, 174 trees were measured, distributed across 21 families, totaling 38 species. In the palynological survey, 715 pollen grains were recorded, divided into 14 botanical families. The families Fabaceae and Malvaceae were found both in the pollen rain collection and in the forest inventory; 11 types/botanical families were found exclusively in the pollen rain, while 18 families were identified only through the forest inventory methodology.
Keywords:
Cerrado; Serranópolis; pollen; plant species
Resumo
Os sítios arqueológicos sofrem ameaças, em razão do avanço da agropecuária sobre o Cerrado, atualmente limitado a faixas de vegetação. Pesquisas palinológicas possibilitam uma visão sistêmica da diversidade vegetal e contribuem para a análise de processos antrópicos. Inventários florestais associados à caracterização dos espectros polínicos, nas regiões com ocorrência de sítios arqueológicos, são fundamentais para a conservação vegetal natural e proteção dos abrigos. Este estudo teve como objetivo a identificação palinológica e a realização do inventário florestal na área de um sítio arqueológico na região de Serranópolis, sudoeste de Goiás. Buscou-se compreender a relação entre a vegetação nativa e a atual antropização da região no entorno do sítio arqueológico. No inventário florestal, foram mensuradas 174 árvores, distribuídas em 21 famílias, totalizando 38 espécies. No levantamento palinológico, foram registrados 715 grãos de polén, distribuídos em 14 famílias botânicas. As famílias Fabaceae e Malvaceae foram encontradas tanto na coleta de chuva polínica quanto no inventário florestal, 11 famílias botânicas foram encontradas somente na assembleia resultante da chuva polínica, enquanto no inventário florestal foram identificadas 18 famílias registradas exclusivamente por meio desta metodologia.
Palavras-chave:
Cerrado; Serranópolis; pólen; espécies vegetais
1. Introduction
The Brazilian Cerrado comprises 200 million hectares of tropical savanna and represents a global biodiversity hotspot under intense pressure from agribusiness expansion and rapid native vegetation loss (Strassburg et al., 2017). The Cerrado is a globally important tropical savanna that provides essential ecosystem services, accounting for approximately 43% of Brazil’s surface water resources outside the Amazon Basin and functioning as a critical carbon reservoir. Despite harboring more than 4,800 endemic species, the biome faces one of the highest rates of land conversion worldwide, with deforestation rates that have reached up to 2.5 times those recorded in the Amazon. This scenario has been driven primarily by the rapid expansion of agribusiness and cattle ranching, resulting in the loss of approximately 46% of the original native vegetation (Strassburg et al., 2017).
Accurate characterization of these remnants is essential for biodiversity conservation, ecological restoration, and environmental monitoring. Floristic surveys and forest inventories remain the primary tools for describing vegetation structure and composition, providing quantitative information on species richness, density, basal area, and biomass (Scolforo and Mello, 2006). These surveys constitute the foundation for conservation planning and ecological assessments. However, conventional inventories have important limitations because they generally focus on woody vegetation, frequently underestimating herbaceous species and graminoids, which represent a substantial proportion of Cerrado biodiversity. Furthermore, species identification in the field may be hindered by the presence of sterile individuals, seasonal phenological variation, sampling intensity, and plot distribution, reducing the representativeness of the vegetation and making comparisons among different surveys more difficult, as well as limiting taxonomic resolution in highly diverse communities (Kiataqui et al., 2022).
Given these limitations, modern palynology, based on the analysis of modern pollen rain, has emerged as an important complementary approach for vegetation characterization. Modern pollen rain reflects the reproductive activity of plant species and records taxa that may be absent from or undetected during conventional vegetation surveys, particularly herbaceous, shrubby, and anemophilous species (Faegri et al., 1989; Cassino and Ledru, 2021). In addition, pollen spectra have been widely used to reconstruct vegetation dynamics, assess landscape heterogeneity, understand pollen dispersal patterns, calibrate vegetation–pollen relationships, identify resources used by pollinators, and establish modern analogues for paleoenvironmental reconstructions (Absy et al., 2014; Cassino and Ledru, 2021).
Despite these advantages, the relationship between vegetation composition and pollen assemblages is not straightforward. Pollen production, dispersal, and preservation differ considerably among taxa, resulting in an uneven representation of plant groups in pollen records. Anemophilous taxa, such as Poaceae and Cyperaceae, tend to be overrepresented, whereas many entomophilous tree species produce comparatively smaller amounts of pollen. Consequently, modern pollen spectra should be interpreted in light of their depositional biases and in conjunction with direct vegetation surveys (Cassino and Ledru, 2021). Whereas forest inventories directly record the species present within the sampled area, modern pollen rain integrates signals from both local and regional vegetation, reflecting differences in species abundance, pollination mechanisms, pollen production, dispersal, and preservation (Faegri et al., 1989; Cassino et al., 2015; Cassino and Ledru, 2021). Rather than replacing floristic inventories, pollen analysis complements them by incorporating plant groups that are difficult to detect using conventional sampling methods, while also expanding the spatial representativeness of vegetation.
Floristic similarity between forest inventories and modern pollen rain constitutes an important indicator of the ecological representativeness of modern pollen records. Previous studies have shown that this relationship varies according to vegetation structure, species population density, pollination systems, and landscape configuration (Calcote, 1995). In heterogeneous ecosystems such as the Cerrado, this relationship becomes even more complex because of the coexistence of different phytophysiognomies and the high diversity of life forms (Cardoso et al., 2026). Thus, integrating floristic inventories with modern pollen rain analyses provides a more comprehensive characterization of plant diversity, reduces methodological biases, and contributes to environmental monitoring, biodiversity conservation, and the reconstruction of vegetation dynamics over time (Mendes et al., 2021; Cassino and Ledru, 2021).
The integration of forest inventories and modern pollen analyses therefore provides a more comprehensive assessment of plant diversity than either method alone. Comparative studies have shown that these approaches often identify distinct sets of taxa, demonstrating their complementarity for ecological monitoring, biodiversity conservation, and the interpretation of environmental change (Cassino and Ledru, 2021; Mendes et al., 2021). Such integrated approaches are particularly valuable in landscapes undergoing rapid anthropogenic transformation, where understanding both the current vegetation structure and pollen dispersal patterns contributes to monitoring the ecological consequences of agricultural expansion.
Although modern pollen rain studies integrated with floristic inventories have been conducted in the Brazilian Atlantic Forest (Behling et al., 1997; Behling and Negrelle, 2006), such studies remain scarce in the Cerrado biome. To the best of our knowledge, no published study has integrated a standardized forest inventory with modern pollen rain sampling using pollen traps at the same archaeological site in the Cerrado. This highlights the novelty of the present study by establishing a direct link between current vegetation structure and its palynological representation in a specific archaeological context.
This integrated perspective is particularly relevant in the archaeological landscapes of the Cerrado in central Brazil. Human occupation of this biome began approximately 11,000 years BP with populations associated with the Itaparica Tradition, whose settlements extended across large areas of Central Brazil (Barbosa and Schmitz, 2008). Archaeological evidence indicates that these populations made extensive use of native vegetation for food, fuel, fibers, and construction materials, exploiting species that remain characteristic components of Cerrado vegetation today, including palms (Arecaceae) and species of the genus Hymenaea (Barbosa and Schmitz, 2008). More recently, evidence has suggested that Indigenous populations actively influenced the structure and composition of Cerrado vegetation, in a manner similar to the processes described for the Amazon and Atlantic Forest ecosystems (Stewart et al., 2021).
The Serranópolis Archaeological Complex, in the state of Goiás, represents one of the most important records of early human occupation in central South America. However, the native vegetation surrounding these archaeological sites is currently threatened by agricultural expansion, which has fragmented the seasonal dry forests into isolated remnants largely restricted to rocky slopes and riparian corridors (Rubin et al., 2016; Resende et al., 2019). Although floristic surveys and palynological studies have been independently applied in Cerrado ecosystems, studies integrating these approaches in archaeological landscapes remain scarce, limiting a broader understanding of vegetation structure and its implications for biodiversity conservation and paleoecological interpretation.
herefore, the present study compared vegetation characterization obtained through a conventional forest inventory and modern pollen analysis in a Cerrado remnant associated with the archaeological site GO-Ja-02 (Gruta do Diogo I), located in Serranópolis, Goiás, Brazil. The complementarity of these two sampling methods in documenting plant diversity was evaluated, and their contributions to vegetation conservation, environmental monitoring, and the interpretation of long-term human–environment interactions were discussed.
2. Methodology
2.1. Study area description: archaeological sites of the Serranópolis Region
The region of Serranópolis, in southwestern Goiás, comprises a significant set of archaeological sites studied since the 1970s. Research conducted between 1970 and 1990 identified dozens of rock shelters with lithic, rock art, ceramic, and burial records, some dated to the Pleistocene-Holocene boundary (Rubin et al., 2024). Available data indicate that the first human groups occupied the Cerrado from nuclei associated with the Itaparica Tradition, including Serranópolis, with chronologies around 12,000 to 11,000 years BP. These hunter-gatherer populations exploited plant resources for multiple purposes, such as firewood, fibers, and shelter coverings, and their diet included palm fruits (Arecaceae) and jatobás (Hymenaea spp.) (Barbosa and Schmitz, 2008).
The Itaparica technological tradition, characterized by well-elaborated unifacial scrapers, which occupied the shelters and developed a simple economy in an environment with climatic stability, extends from the beginning of the Holocene at about 11,000 years BP until around 8,500 years BP, when it is replaced by a new cultural horizon called the Serranópolis Tradition, marked by a lithic industry with less elaborate instruments with poorly retouched flakes and the use of bones and shells as raw material. This new phase of occupation lasts until about 6,500 years BP, from which point in the Serranópolis region, an occupational hiatus occurs that extends until about 2,400 years BP, when the record of the first ceramic groups in the Cerrado biome begins, eventually using the shelters in a period when the environment was supposedly similar to the current one (Rubin et al., 2016)
After an occupation hiatus, ceramic evidence reappears around 2,500 years BP, attributed to groups of the Jataí phase, Una Tradition, who may have practiced hunting, fishing, gathering, and incipient agriculture (Rubin et al., 2024). Currently, the archaeological sites of Serranópolis restrict native vegetation to narrow strips near drainage systems and shelters. In this context, the conservation of vegetation remnants is fundamental for protecting shelters. Moreover, integrated studies of forest inventory and palynology become strategic for understanding and conserving the cultural and natural landscape (Rubin et al., 2016; Resende et al., 2019; Santos et al., 2019).
The study area is located in the municipality of Serranópolis, in southwestern Goiás, where archaeological research began in the 1970s (Figure 1). The area defined for this study consists of a predominantly flat landscape, almost entirely anthropized, except for hillside slopes and vegetation along river margins, where forest formations remain preserved. The other areas are occupied by extensive agriculture and livestock. Across these anthropized landscapes, basalt outcrops are found, which characterize the type of soil in these agricultural areas: red latosol in cultivated areas and quartzarenic neosol where pastures extend (Sousa and Lobato, 2004).
Map of the Serranópolis Archaeological Complex showing the 50-m radius around the forest inventory plot and the 50-km radius representing the potential source area of modern pollen rain.
The region, initially flat, transitions into a more rugged relief formed by small hills. This area, also anthropized, is occupied by extensive cattle ranching, predominantly covered with Urochloa spp. The described region contains a drainage network formed by the Inacinho and Bela Vista streams and the Verde River. Near the shelter, the altitude increases, and further ahead sandstone outcrops appear, which constitute the GO-Ja-02 shelter.
At the GO-Ja-02 archaeological site, the Cerrado area, with sandy soils beginning at the front of the shelter, is characterized by evergreen to semi-deciduous dry forest, with tree species reaching approximately 15 to 30 meters in height. This is another forest formation with predominance of a shrub–tree stratum, classified as deciduous seasonal forest, popularly known as “Mata Seca” (Seasonally Dry Forest physiognomy sensu Ribeiro and Walter, 2008)
The superficial soil of the site, along the “Mata Seca” surrounding the excavation, is covered by a thin layer of litter (organic matter) and composed of quartz and iron oxides, with black to brown coloration. This is a sandy soil (Sousa and Lobato, 2004). Iron oxides (mainly hematite and goethite) appear in discrete proportions, often as films or specks on quartz grains. They are the main agents responsible for brown to reddish hues when organic matter decreases (Scheinost and Schwertmann, 1999). The black to dark brown coloration at the surface indicates a superficial A horizon enriched by organic matter from litter, typical of seasonally dry vegetation areas, where nutrient cycling is concentrated in the first decimeter of soil (Embrapa, 2021; Frazão et al., 2008).
The climate of the region, according to Köppen’s classification, is type Aw, tropical with a dry season, characteristic of Cerrado regions, with two well-defined seasons: dry (May to September) and rainy (October to April), with annual precipitation around 1,500 mm (Ribeiro and Walter, 2008).
This study area is located in the Paraná Sedimentary Basin, N/NW segment, composed of sedimentary deposition throughout the Paleozoic and Mesozoic. In this region, outcrops of rocks from the São Bento Group (sediments and basaltic flows related to the Paraná Sedimentary Basin) is formed by a succession of sandstones from the Botucatu Formation, deposited during the Cretaceous, which are interbedded with basaltic lava flows from the Serra Geral Formation (Rubin et al., 2024). The relief of this region results from tectonic effects and climatic events acting on the rocks, with contours defined by erosive processes related to the Mesozoic geotectonic history of South America, uplift during the Tertiary, and regional planation in the Cretaceous (Rubin et al., 2024).
2.2. Forest inventory
A forest census was conducted within a 50 m radius from the center of the cave (Archaeological Site – GO-Ja-02, totaling approximately 7.854 m2 (0,785 ha), sampled (Figure 2). This methodology was chosen as it is the most efficient for representing the vegetational remnants of the region (Pereira et al., 2017). In the area, only vegetation fragments associated with the rocky formation near the site were found. The vegetation is mostly characterized as Deciduous Seasonal Forest (“dry forest”), according to the classification of Ribeiro and Walter (2008).
GO-JA-02 shelter, location of the archaeological excavation, where the forest inventory was conducted and the pollen collector was installed.
The inclusion criterion for individuals classified as trees in the census was a circumference equal to or greater than 15 cm (Venturoli et al., 2015; Martins et al., 2016). Botanical identification was carried out in the field to the closest known species level using botanical identification guides (Silva-Júnior and Pereira, 2009; Kuhlmann, 2018). For species that could not be identified in the field, botanical material was collected and sent to a herbarium for future identification. Species classification followed the most recent APG system (APG, 2016). The inventory results were presented in a table showing species with absolute and relative frequency, by botanical family, and the number of species per family.
Sampling sufficiency was evaluated using the non-parametric species richness estimators Chao1, first-order Jackknife (Jackknife 1), and Bootstrap, calculated with PAST (Paleontological Statistics) software, version 4. These estimators were used to estimate the potential richness of the tree community based on the frequency of rare species recorded in the inventory. In addition, a species accumulation curve was generated to assess sampling representativeness and evaluate the tendency toward stabilization of the observed species richness.
2.3. Palynological analysis
The survey of pollen types in the surroundings of the Serranópolis archaeological complex was divided into three stages: the first consisted of the construction and installation of the pollen collector; the second involved the collection of palynomorphs; and the third comprised the preparation and identification of botanical families.
In the first stage, the construction and installation of the pollen rain collector was carried out. It is important to emphasize that this represents the creation of a new tool, since there are few methodological references in the scientific literature regarding the construction of such equipment. The materials used to build the pollen rain collector were: (i) one discarded iron ring from civil construction; (ii) one poultry feeder in the shape of a funnel; (iii) one 70 mm PVC cap, perforated with a drill, used at the smaller base of the funnel to store the collected pollen; (iv) one 25 mm laboratory filter inserted inside the 70 mm PVC cap to separate the collected pollen from rainwater; (v) one galvanized mesh placed at the upper part of the funnel to filter incoming objects, allowing only particles of pollen size (200 µm) to pass; and finally (vi) one treated eucalyptus post, 4 meters high, used as the base for the collector, fixed into the ground in front of the GO-JA-02 shelter. At the top of the post, the iron ring was attached, supporting the collector with its wider opening facing upward, where palynomorphs are deposited by wind action, and the narrower base with the 70 mm PVC cap facing downward, where pollen accumulates on the laboratory filter. These filters are collected monthly and replaced with new ones (Figure 3).
Pollen rain collector installed in front of the GO-JA-02 shelter, in the archaeological complex located in Serranópolis, Goiás. The device was built using repurposed materials, including an iron ring, plastic funnel, PVC cap, laboratory filter, and galvanized mesh. The equipment was fixed to a four-meter treated eucalyptus post.
After completing the installation stage, the collection of pollen material deposited in the collector was carried out. Collections were conducted at two distinct moments. Initially, a preliminary collection, referred to as a “test collection,” was performed and subsequently discarded, covering the period from August 27, 2023, to September 8, 2023. Following this, two separate collections were conducted, which are the focus of analysis in this study. These collections encompassed one period during the rainy season and another during the dry season. The first collection, corresponding to the rainy season, was carried out between September 9, 2023, and October 8, 2023. The second collection, corresponding to the dry season, was conducted between March 24, 2024, and June 12, 2024.
Once the collections were completed, preparation of the palynomorph assemblage began at the Paleoecology Laboratory of the Center for Biological Studies and Research (CEPB), located at Campus II of PUC Goiás. The methodology used for material preparation followed Barberi (2001), aiming to concentrate the pollen contained in the samples to enable identification and counting of pollen grains.
Preparation of the pollen material began with sieving (filtration) to separate macro-remains from palynomorphs using distilled water (H2O D). Subsequently, the filters, where pollen had been deposited by wind action, were dissolved in a heated solution of 5 mL of 10% KOH with 50 mL of distilled water, following Salgado-Labouriau’s (2007) protocol. After this procedure, the pollen material was sieved again. The next step involved combining the pollen material immersed in distilled water into a single beaker and distributing the contents into 16 test tubes for centrifugation at 2,200 rpm for five minutes. This procedure was repeated eight times for each prepared collection until all pollen material was concentrated at the bottom of a single test tube. Glycerin was then added to the material, which was transferred to a storage tube.
Once preparation was finalized, microscope slides were prepared, and the fifth and final methodological stage, taxonomic identification of palynomorph types, was carried out. Identification was performed through microscopic analysis at 40x magnification, using identification keys (Salgado-Labouriau, 1973) and a recent palynological
atlas (Lorente et al., 2017). To reach the saturation curve, 300 grains per level/collection were analyzed (Salgado-Labouriau, 1973). In total, more than 715 grains were counted and identified, with 385 corresponding to the rainy season collection and 330 to the dry season collection.
3. Results
3.1. Forest inventory
A total of 174 individuals, representing 38 species distributed across 21 botanical families, were identified. Fabaceae was the richest family, comprising 10 species, followed by Sapindaceae with three species. Anacardiaceae, Malvaceae, Moraceae, Myrtaceae, and Erythropalaceae were each represented by two species, whereas the remaining 14 families were represented by a single species. The most abundant species were Astronium urundeuva, Agonandra brasiliensis, Hymenaea courbaril, Astronium fraxinifolium, Sterculia chicha, and Heisteria ovata, which together accounted for 52% of all recorded individuals (Table 1; Figure 4).
Botanical species name, absolute and relative frequency found in the forest formation of the archaeological site Gruta do Diogo 1, Serranópolis, Goiás.
Tree species identified in the forest inventory conducted at the archaeological site GO-Ja-02, Serranópolis, Goiás, Brazil.
Non-parametric species richness estimates indicated an expected richness of 48–52 species using the Chao1 estimator (95% CI: 44–58), 46–50 species using the first-order Jackknife estimator (95% CI: 43–53), and 43–47 species using the bootstrap estimator (95% CI: 41–49). These estimates indicate that the inventory sampled approximately 73–78% of the expected tree species richness in the study area. Furthermore, the species accumulation curve showed a tendency toward stabilization, suggesting that sampling approximately 40–60 additional individuals would likely allow the curve to approach a plateau, thereby increasing the probability of detecting rare species.
3.2. Palynological analysis
The pollen collection showed low richness of botanical families (n: 14) that rely on atmospheric agents to disperse the pollen of their flowers (Table 2; Figure 5). The results obtained demonstrated, during the rainy season, low diversity of botanical families, with a predominance of Fabaceae pollen. In the dry season, there was also low diversity of botanical families, identified from the palynomorphs, but with species different from those recorded in the rainy season, and a high abundance of Poaceae.
Botanical family, absolute and relative frequency found in the pollen rain collection during the rainy season (09/09/2023 to 08/11/2023) and the dry season (23/03/2024 to 12/06/2024) at the archaeological site Gruta do Diogo I, Serranópolis, Goiás.
3.3. Tree x Pollen analysis
Forest and pollen inventory surveys of the same region show that only 4 botanical families are similar in both surveys, with only Fabaceae, Euphorbiaceae, and Malvaceae showing a greater number of occurrences of pollen grains or tree pollen (Table 3; Figure 6).
Comparison of occurrences between botanical families of trees and pollen grains at the archaeological site Gruta do Diogo 1, Serranópolis, Goiás.
Comparison of botanical families recorded in the forest inventory and identified in the modern pollen rain at the archaeological site GO-Ja-02, Serranópolis, Goiás, Brazil.
4. Discussion
In the forest inventory, 174 trees were measured, corresponding to 0.02 tree per m2, distributed across 38 species belonging to 21 botanical families. In the palynological survey, 600 pollen types were identified, divided into 14 botanical families. The methodologies Applied, forest inventory and pollen rain collection, revealed distinct diversity, abundance, and frequency patterns.
From the forest inventory, greater diversity was obtained was obtained for woody species, with predominance of the species Astronium urundeuva, Agonandra brasiliensis, Hymenaea courbaril, Astronium fraxinifolium, Sterculia chicha, and Heisteria ovata. The palynological study, in turn, presented less expressive species diversity; however, since it was not limited to woody species, it showed greater predominance of Fabaceae (tree/shrub) during the rainy season and Poaceae (herbaceous) during the dry season. The strong representation of Poaceae in the modern pollen rain is also consistent with studies conducted in African forest–savannah mosaics. In Ghana, Julier et al. (2018) found that Poaceae pollen provided the clearest palynological distinction along a forest–savannah gradient, despite incomplete correspondence between pollen assemblages and the vegetation recorded within the plots. The high abundance of Poaceae in the present study may therefore reflect not only the local herbaceous component, which was excluded from the tree inventory, but also its high pollen productivity, anemophilous dispersal, and the extensive pasture and open vegetation surrounding the archaeological site.
It is noteworthy that this diversity, derived from non-forest habits, was found only in the palynological study. The surveys carried out jointly in the same area demonstrate two complementary methods, since if only one method had been used, it would not have been possible to encompass the entire floristic richness of the region. For this reason, pollen rain collection is essential, especially to integrate non-forest species into the survey.
Similar patterns have been reported in tropical forest and savanna ecosystems in Bolivia and Ghana. Gosling et al. (2018) compared modern pollen rain obtained from pollen traps with vegetation inventories conducted in 1ha plots encompassing tropical rainforest, seasonally dry semi-deciduous forest, savanna, and forest–savanna transition zones. Their results showed that palynological richness was higher than vegetation richness, whereas pollen diversity was consistently lower than plant diversity. This discrepancy was attributed to differences in pollination strategies, pollen productivity and dispersal, canopy openness, taxonomic resolution, and the larger source area represented by modern pollen rain. These findings are consistent with the low taxonomic overlap observed in the present study and reinforce that modern pollen rain and forest inventories represent distinct, yet complementary, components of plant diversity.
An unprecedented contribution of this study is the integrated application of forest inventory and modern palynological sampling. Although modern pollen rain studies associated with floristic inventories have been conducted in the Brazilian Atlantic Forest (Behling et al., 1997; Behling and Negrelle, 2006), investigations using this approach are still incipient in the Cerrado biome. Furthermore, studies integrating, within the same archaeological site, a standardized forest inventory with modern pollen rain sampling using pollen traps remain scarce. This knowledge gap highlights the innovative nature of the present study, as it establishes a direct correspondence between the composition and structure of the current vegetation and its palynological representation within a specific archaeological context. Existing studies have generally focused either on floristic and forest inventories or on palynological surveys, each of which has individually contributed to understanding local vegetation. In this regard, Freitas et al. (2015) identified a total of 47 palynomorph taxa, including pollen grains and other palynomorphs, from the archaeological site Evaristo I, representing both arboreal-shrub and herbaceous vegetation. Similarly, the present study identified botanical families associated with both woody and herbaceous components of the vegetation, reinforcing the usefulness of modern palynological analyses for characterizing vegetation in archaeological landscapes. Forest is the same vegetation type surrounding the study area investigated here. This occurs because, in addition to the possibility of botanical families incorporating greater diversity of plant growth forms, pollen rain is embedded in the spatial context in which a pollen grain can be dispersed by wind (anemophily) up to a distance of 50 km (Salgado-Labouriau, 1973). Thus, the results obtained in palynology encompass a larger area than that of the forest inventory, which was limited to a 50 m radius around the archaeological site GO-Ja-02. For this reason, as demonstrated by Freitas et al. (2015), palynology broadens the understanding of the landscape, including a greater diversity of Cerrado phytophysiognomies.
On the other hand, regarding the forest inventory carried out, which represented only forest species, 18 families were identified exclusively in the forest inventory, corresponding to 62% of the botanical families recorded. Therefore, the methods, when applied in isolation, are not capable of demonstrating floristic richness in its true diversity. This reinforces that palynological and forest studies encompass distinct portions of vegetation, thereby demonstrating their complementarity.
As reported in the forest inventory results, the families with the highest occurrence were: Anacardiaceae, Opiliaceae, Fabaceae, Anacardiaceae, Malvaceae, and Erythropalaceae. Of these, two were found both in the pollen rain collection and in the forest inventory: Fabaceae and Malvaceae, which correspond to 33.3% of the main species found in the forest inventory and 52.62% of the types/families identified in the pollen rain.
The floristic and palynological survey in the archaeological site provides relevant information regarding local environmental conservation. Although small areas were once considered low priority for biodiversity conservation, under the justification that they lacked the capacity to maintain ecological processes, recent studies demonstrate that vegetation fragments have ecological relevance, as they possess conservation value in terms of diversity and structure of tree communities, which are essential for the conservation of flora of relevant species (Carneiro et al., 2023).
Among the 38 species identified in the forest inventory, six present relevance determined by other factors. The species Astronium urundeuva accounted for 17.82% of those identified in this forest survey. It is a heavily exploited species, as its timber potential led to extensive exploitation for civil and rural construction (Venturoli et al., 2015; Xavier et al., 2017), reinforcing the need for its conservation in remaining Cerrado areas, such as around the GO-JÁ-02 shelter. Regarding the ecological interactions of plant species with fauna and feeding habits, the fruits of the following species present in the forest inventory, Heisteria ovata, Hymenaea courbaril, and Sterculia chicha, are consumed by birds, marsupials, and primates, which in turn act in the dispersal of their seeds (Moura, 2023; Ferreira, 2023). These are therefore key species for conservation, due to their ecological functions.
Considering that the botanical research was conducted within the Serranópolis archaeological complex, specifically in the surroundings of the GO-Ja-02 rock shelter, the botanical data collected allow us to establish connections with the ancestral peoples who once occupied this region. The study by Freitas et al. (2018) on the construction of reference collections of useful/economic plants as a foundation for archaeobotanical research demonstrates how such data can be used to infer practices of indigenous groups regarding plant utilization. Archaeobotanical evidence dated to 4,500 years BP from the Alcobaça Site (Pernambuco) revealed babassu palm fruits, while findings from the Pedra do Cachorro Site in Buíque (Pernambuco) suggest subsistence strategies of hunter-gatherer groups based on the collection of fruit-bearing plants, including babassu palms (Attalea speciosa), as well as the use of jatobá (Hymenaea courbaril) wood and seeds.
Excavations carried out in the 1970s at the Serranópolis complex, as noted in the introduction, uncovered palm fruit pits (Arecaceae) and jatobá seeds (Hymenaea spp.), likely reflecting the dietary habits of these human groups (Barbosa and Schmitz, 2008). In the present forest inventory, 17 specimens of jatobá (Hymenaea courbaril) were identified. This species belongs to the same genus (Hymenaea) as that recovered in the excavations, reinforcing its role as a dietary component of indigenous groups inhabiting the study area, as well as those occupying present-day Piauí, according to archaeobotanical studies in Cerrado archaeological sites (Freitas et al., 2018).
Palynomorphs recovered from sediments of the GO-Ja-02 excavation provide further insights into possible interactions between early human occupations and vegetation, while also contributing to understanding vegetation dynamics in the region. From a plant ecology perspective, future research should include long-term monitoring of vegetation development in situ, alongside assessments of anthropogenic areas and their seasonal changes or impacts (dry and rainy seasons). Such monitoring would enable integration of these ecological observations with the palynological and forest inventory data presented here, thereby strengthening hypotheses on vegetation evolution and human-environment interactions in the Cerrado biome.
5. Conclusion
Through the application of both methodologies, it was possible to achieve broader inclusion of species and botanical families, thereby providing a more comprehensive representation of local plant diversity.
The comparison between the methods revealed low taxonomic overlap, with only three families (Fabaceae, Malvaceae, and Euphorbiaceae) occurring simultaneously in both surveys. The remaining families were exclusive to each methodology, clearly demonstrating the complementary nature of the approaches.
The presence of ecologically relevant species, together with palynological information on regeneration, underscores the importance of this fragment for biodiversity conservation and for safeguarding the integrity of the archaeological site.
Finally, the study fulfills a dual role: it documents the current status of the Cerrado remnant surrounding the GO-JÁ-02 shelter, located in a highly anthropized area, and provides data that support hypotheses regarding past interactions between human occupations and local vegetation.
Acknowledgements
The author, CMSN, thanks CNPq for the continuation of the research productivity grant.
Data Availability Statement
All data underlying the results presented here are contained in the published article.
References
-
ABSY, M.L., CLEEF, A.M., D’APOLITO, C. and SILVA, M.F.F., 2014. Palynological differentiation of savanna types in Carajás, Brazil (southeastern Amazonia). Palynology, vol. 38, no. 1, pp. 78-89. https://doi.org/10.1080/01916122.2013.842189
» https://doi.org/10.1080/01916122.2013.842189 -
ANGIOSPERM PHYLOGENY GROUP – APG, 2016. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG IV. Botanical Journal of the Linnean Society, vol. 181, no. 1, pp. 1-20. https://doi.org/10.1111/boj.12385
» https://doi.org/10.1111/boj.12385 - BARBERI, M., 2001. Mudanças paleoambientais na região dos cerrados durante o Quaternário Tardio: o estudo da Lagoa Bonita, DF Brasília: Universidade de Brasília. Dissertação de Mestrado.
- BARBOSA, A.S. and SCHMITZ, P.I., 2008. Ocupação indígena do Cerrado: esboço de uma história. In: S.M. SANO, S.P. ALMEIDA and J.F. RIBEIRO, eds. Cerrado: ecologia e flora Brasília, DF: Embrapa Informação Tecnológica, vol. 1, pp. 48-68.
-
BEHLING, H. and NEGRELLE, R.R.B., 2006. Vegetation and pollen rain relationship from the tropical Atlantic rain forest in southern Brazil. Brazilian Archives of Biology and Technology, vol. 49, no. 4, pp. 631-642. https://doi.org/10.1590/S1516-89132006000500013
» https://doi.org/10.1590/S1516-89132006000500013 -
BEHLING, H., NEGRELLE, R.R.B. and COLINVAUX, P.A., 1997. Modern pollen rain data from the tropical Atlantic rain forest, Reserva Volta Velha, South Brazil. Review of Palaeobotany and Palynology, vol. 97, no. 3-4, pp. 287-299. https://doi.org/10.1016/S0034-6667(96)00073-5
» https://doi.org/10.1016/S0034-6667(96)00073-5 -
CALCOTE, R.R., 1995. Pollen source area and pollen productivity: evidence from forest hollows. Journal of Ecology, vol. 83, no. 4, pp. 591-602. https://doi.org/10.2307/2261627
» https://doi.org/10.2307/2261627 -
CARDOSO, J.C.F., TREVIZAN, R., MARUYAMA, P.K., CAETANO, A.P.S., GONÇALVES, R.V., ANTONINI, Y. and OLIVEIRA, P.E., 2026. Pollination and plant reproduction in the Cerrado, the world’s most biodiverse savanna. Biological Reviews of the Cambridge Philosophical Society, vol. 101, no. 1, pp. 74-105. https://doi.org/10.1111/brv.70073 PMid:40957635.
» https://doi.org/10.1111/brv.70073 -
CARNEIRO, M.S., RIBEIRO, B.R., CAMPOS, C.C.F., ALMEIDA, H.S., ARAUJO, M.L., SANTOS, F.A.M. and RAMOS, F.N., 2023. Small forest fragments and their importance for conservation of tree communities in Atlantic Semideciduous Forest. Iheringia. Série Botânica, vol. 78, pp. e2023019. https://doi.org/10.21826/2446-82312023v78e2023019
» https://doi.org/10.21826/2446-82312023v78e2023019 -
CASSINO, R.F., MARTINHO, C.T. and DA SILVA CAMINHA, S.A., 2015. Modern pollen spectra of the Cerrado vegetation in two national parks of Central Brazil, and implications for interpreting fossil pollen records. Review of Palaeobotany and Palynology, vol. 223, pp. 71-86. https://doi.org/10.1016/j.revpalbo.2015.09.002
» https://doi.org/10.1016/j.revpalbo.2015.09.002 -
CASSINO, R.F. and LEDRU, M.P., 2021. Quantitative reconstruction of vegetation cover from modern pollen rain in the Cerrado biome of Brazil. Palaeogeography, Palaeoclimatology, Palaeoecology, vol. 574, pp. 1-16. https://doi.org/10.1016/j.palaeo.2021.110462
» https://doi.org/10.1016/j.palaeo.2021.110462 -
EMBRAPA, 2021 [viewed 27 April 2026]. Neossolo Quartzarênico. Agência Embrapa de Informação Tecnológica [online]. Available from: https://www.embrapa.br/agencia-de-informacao-tecnologica/tematicas/bioma-cerrado/solo/tipos-de-solo/aq-/-nq
» https://www.embrapa.br/agencia-de-informacao-tecnologica/tematicas/bioma-cerrado/solo/tipos-de-solo/aq-/-nq - FAEGRI, K., KALAND, P.E. and KRZYWINSKI, K., 1989. Text-book of pollen analysis 4th ed. New York: John Wiley & Sons, 328 p.
- FERREIRA, J.O.V., 2023. Dispersão induzida de sementes por vertebrados frugívoros em floresta em restauração no sudeste brasileiro Campinas: Universidade Estadual de Campinas. Dissertação de Mestrado.
-
FRAZÃO, L.A., PÍCCOLO, M.C., FEIGL, B.J., CERRI, C.C. and CERRI, C.E.P., 2008. Propriedades químicas de um Neossolo Quartzarênico sob diferentes sistemas de manejo no Cerrado. Pesquisa Agropecuária Brasileira, vol. 43, no. 5, pp. 641-648. https://doi.org/10.1590/S0100-204X2008000500012
» https://doi.org/10.1590/S0100-204X2008000500012 - FREITAS, A., NASCIMENTO, A.L., AGE, M., MORAES, L.A.D., FILHO, S.P. and REIS, R.L., 2018. Piauí de plantas e gentes: construção de coleções de referência de plantas úteis/econômicas como base para estudos arqueobotânicos. Revista do Laboratório de Arqueologia e Paleontologia da UEPB, vol. 1, pp. 83-116.
- FREITAS, A.G., BARBERI, M. and RUBIN, J.C.R., 2015. Chuva de pólen moderna em Serranópolis (GO): subsídios para interpretações paleoambientais. Habitus, vol. 13, no. 1, pp. 115-132.
-
GOSLING, W.D., JULIER, A.C.M., ADU-BREDU, S., DJAGBLETEY, G.D., FRASER, W.T., JARDINE, P.E., LOMAX, B.H., MALHI, Y., MANU, E.A., MAYLE, F.E. and MOORE, S., 2018. Pollen-vegetation richness and diversity relationships in the tropics. Vegetation History and Archaeobotany, vol. 27, no. 2, pp. 411-418. https://doi.org/10.1007/s00334-017-0642-y PMid:31983811.
» https://doi.org/10.1007/s00334-017-0642-y -
JULIER, A.C.M., JARDINE, P.E., ADU-BREDU, S., COE, A.L., DUAH-GYAMFI, A., FRASER, W.T., LOMAX, B.H., MALHI, Y., MOORE, S., OWUSU-AFRIYIE, K. and GOSLING, W.D., 2018. The modern pollen–vegetation relationships of a tropical forest–savannah mosaic landscape, Ghana, West Africa. Palynology, vol. 42, no. 3, pp. 324-338. https://doi.org/10.1080/01916122.2017.1356392
» https://doi.org/10.1080/01916122.2017.1356392 -
KIATAQUI, F.K., NORONHA, S.E. and SIMON, M.F., 2022. Contribution of the Brazilian National Forest Inventory to the knowledge of Cerrado woody flora. Biota Neotropica, vol. 22, no. 1, pp. e20211306. https://doi.org/10.1590/1676-0611-bn-2021-1306
» https://doi.org/10.1590/1676-0611-bn-2021-1306 - KUHLMANN, M., 2018. Frutos e sementes do Cerrado: espécies atrativas para a fauna Brasília, DF: Edição do Autor, vol. 2, 464 p.
- LORENTE, F.L., BUSO JUNIOR, A.A., OLIVEIRA, P.E.D. and PESSENDA, L.C.R., 2017. Atlas palinológico: laboratório 14C-CENA/USP= Palynological atlas: 14C laboratory-CENA/USP São Paulo: Universidade de São Paulo.
- MARTINS, T., ARAUJO, F.D., SIQUEIRA, K.N., RIBEIRO, A.C.C. and SILVA-NETO, C.M., 2016. Cerrado forests: seasonal forest and cerradão. Agrarian Academy, vol. 3, pp. 1-14.
-
MENDES, L.A., MENESES, M.E.N.S., BEHLING, H. and COSTA, M.L., 2021. Modern and fossil pollen record from region of the middle Araguaia River floodplain, Tocantins State (Brazil). Pesquisas em Geociências, vol. 48, no. 4. https://doi.org/10.22456/1807-9806.111397
» https://doi.org/10.22456/1807-9806.111397 -
MOURA, J.S.S., 2023 [viewed 27 April 2026]. Recursos alimentares dos muriquis (Brachyteles Spix, 1823), primatas endêmicos da Mata Atlântica e criticamente em perigo de extinção [online]. Natal: Universidade Federal do Rio Grande do Norte. Available from: https://repositorio.ufrn.br/handle/123456789/53719
» https://repositorio.ufrn.br/handle/123456789/53719 -
PEREIRA, I.S., CALIL, F.N., MARTINS, T.O., SILVA-NETO, C.D.M., BORGES, J., VENTUROLI, F. and OLIVEIRA, L.H., 2017. Fire effect on the seasonal forest structure in the Cerrado biome. Floresta, vol. 46, no. 4, pp. 499. https://doi.org/10.5380/rf.v46i4.45277
» https://doi.org/10.5380/rf.v46i4.45277 - RESENDE, F.E.C.P., FERNANDES, A.B., RUBIN, J.C.R., BARBERI, M., BICHUETTE, M.E., GALLÃO, J.E., ZEPON, T., SILVA, S.M. and SOUZA, U.F., 2019. Relatório final das ações emergenciais de conservação das pinturas e gravuras rupestres nos sítios de abrigos do Complexo Arqueológico de Serranópolis-GO Brasília: MRS Estudos Ambientais.
- RIBEIRO, J.F. and WALTER, B.M.T., 2008. As principais fitofisionomias do bioma Cerrado. In: S.M. SANO, S.P. ALMEIDA and J.F. RIBEIRO, eds. Cerrado: ecologia e flora Brasília: Embrapa, pp. 151-222.
-
RUBIN, J.C.R., BARBERI, M., LACERDA FILHO, J.V. and SILVA, R.T., 2024. Geociências no contexto arqueológico de Serranópolis, Goiás, Brasil Curitiba: Editora CRV, 66 p. https://doi.org/10.24824/978652516857.9
» https://doi.org/10.24824/978652516857.9 - RUBIN, J.C.R., SILVA, R.T. and BARBERI, M., 2016. Consideraciones al respecto del contexto arqueológico: Serranópolis, Goiás (Brasil). In: MUSEO DEL DESIERTO, ed. El poblamiento temprano en América, Serie Prehistoria en América Ciudad de México: Museo del Desierto, pp. 320-328.
- SALGADO-LABOURIAU, M.L., 1973. Contribuição à palinologia dos Cerrados Rio de Janeiro: Academia Brasileira de Ciências, 291 p.
- SALGADO-LABOURIAU, M.L., 2007. Métodos e técnicas para o Quaternário São Paulo: Edgard Blücher, 283 p.
- SANTOS, R.S., LIMA, P.G.C., SILVA, K.C.L., COELHO-FERREIRA, M. and MAGALHÃES, M.P., 2019. Plantas úteis da Serra de Carajás: espécies associadas a sítios arqueológicos Belém, PA: Museu Paraense Emílio Goeldi.
- SOUSA, D.M.G. and LOBATO, E., 2004. Cerrado: correção do solo e adubação Brasília, DF: Embrapa Informação Tecnológica.
-
SCHEINOST, A.C. and SCHWERTMANN, U., 1999. Color identification of iron oxides and hydroxysulfates: use and limitations. Soil Science Society of America Journal, vol. 63, no. 5, pp. 1463-1471. https://doi.org/10.2136/sssaj1999.6351463x
» https://doi.org/10.2136/sssaj1999.6351463x - SCOLFORO, J.R.S. and MELLO, J.M., 2006. Inventário Florestal Lavras: UFLA/FAEPE, 561 p.
- SILVA JÚNIOR, M.C. and PEREIRA, B.A.S., 2009. +100 árvores do Cerrado: matas de galeria: guia de campo Brasília, DF: Rede de Sementes do Cerrado, 288 p
-
STEWART, A.M., CONCEIÇÃO, A.V., PACHECO, F., CASTRO, F.P., SILVA, G.M. and GONÇALVES, P.R., 2021 [viewed 27 April 2026]. Saberes que vêm de longe: usos tradicionais do fogo no Cerrado e Amazônia [online]. Brasília: Agro é Fogo. Available from: https://acervo.socioambiental.org/sites/default/files/documents/m1d00037.pdf
» https://acervo.socioambiental.org/sites/default/files/documents/m1d00037.pdf -
STRASSBURG, B.B., BROOKS, T., FELTRAN-BARBIERI, R., IRIBARREM, A., CROUZEILLES, R., LOYOLA, R., LATAWIEC, A.E., OLIVEIRA FILHO, F.J.B., SCARAMUZZA, C.A.M., SCARANO, F.R., SOARES-FILHO, B. and BALMFORD, A., 2017. Moment of truth for the Cerrado hotspot. Nature Ecology & Evolution, vol. 1, no. 4, pp. 0099. https://doi.org/10.1038/s41559-017-0099
» https://doi.org/10.1038/s41559-017-0099 -
VENTUROLI, F., CARVALHO, F.A., SILVA NETO, C.M., MORAES, D.C., MARTINS, T.O. and SOUZA, D.M., 2015 [viewed 27 April 2026]. Manejo florestal no bioma Cerrado: uma opção para conservar e lucrar. Scientia Forestalis [online], vol. 43, no. 107, pp. 617-626. Available from: https://www.ipef.br/publicacoes/scientia/nr107/cap13.pdf
» https://www.ipef.br/publicacoes/scientia/nr107/cap13.pdf -
XAVIER, A.C.F., SILVA-NETO, C.M., MARTINS, T.O., FERREIRA, F.G., MONTEIRO, M.M., OLIVEIRA, G.M. and VENTUROLI, F., 2017. Growth and volume of Myracroduon urundeuva Allemão after ten years of silvicultural interventions. Australian Journal of Crop Science, vol. 11, no. 3, pp. 271-276. https://doi.org/10.21475/ajcs.17.11.03.pne360
» https://doi.org/10.21475/ajcs.17.11.03.pne360
*
e-mail: carloskoa@gmail.comEditor:
