|
|
|
|
|
Universidade Federal de Santa Maria
Ci. e Nat., Santa Maria, v. 47, e85058, 2025
DOI: 10.5902/2179460X85058
ISSN 2179-460X
Submitted: 09/12/2023 • Approved: 06/11/2025 • Published: 10/02/2025
Biology-Ecology
Native tree and arborescent species for Agroforestry Systems in the Uruguayan-Brazilian Pampa
Espécies arbóreas e arborescentes nativas para Sistemas Agroflorestais no Pampa Uruguaio-Brasileiro
Guilherme Krahl de VargasI
Adriana Carla Dias TrevisanI
Beatriz Marcela Sosa Calleja I
Paulo BrackII
I Universidade Estadual do Rio Grande do Sul, Santana do Livramento, RS, Brazil
II Universidad de la República, Montevideo, MO, Uruguay
III Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil
ABSTRACT
Tree species are vital to Agroforestry Systems (AFS), and selecting species that are suited to local conditions is essential for achieving this system’s objectives. In this study we characterized the phytophysionomies present in the Uruguayan-Brazilian Pampa Ecoregion in order to identify the most suitable species to include in AFS. In this sense we aimed to answer the following questions: i) Do the a priori defined phytophysionomies have a different species composition compared to the ecoregion? ii) Which environmental variables are related to species composition variation? iii) Which species are indicators of different phytophysiognomies? iv) Which native species are key for AFS implementation in these phytophysiognomies? We analysed floristic and environmental data from 106 sites across five phytophysiognomies using Non-metric Multidimensional Scaling (nMDS) and Analysis of Similarity (ANOSIM) and the Indicator Species Analysis (IndVal). Results revealed significant differences in species composition between the phytophysiognomies, indicating that the predefined formations have distinct species assemblages. Indicator Species Analysis IndVal identified 136 species, of wich 27 were prioritised for AFS use, contributing to a total of 84 priority native species, including those with multipurpose uses. These findings highlight the importance of considering local phytophysiognomies when planning AFS with native species in this ecoregion.
Keywords: Pampa biome; Indicators species; Agroforestry homegardens; Restorative AFS, Silvopasture system
RESUMO
Espécies arbóreas são vitais para os Sistemas Agroflorestais (SAF), e a seleção das espécies adequadas às condições locais é essencial para atingir os objetivos desse sistema. Nesse estudo, as fitofisionomias presentes na Ecorregião do Pampa Uruguaio—Brasileiro foram caracterizadas a fim de identificar as espécies mais adequadas para uso em SAF. Assim, buscamos responder as seguintes questões: i) As fitofisionomias definidas a priori têm uma composição de espécies diferente em comparação com a ecorregião? ii) Quais variáveis ambientais estão relacionadas a variação na composição de espécies? iii) Quais espécies são indicadoras de diferentes fitofisionomias? iv) Quais espécies nativas são fundamentais para a implementação de SAFs nestas fitofisionomias? Analisamos os dados florísticos e ambientais de 106 locais em cinco fitofisionomias utilizando a Escala Multidimensional Não-Métrica (nMDS), a Análise de Similaridade (ANOSIM) e a Análise de Espécies Indicadoras (IndVal). Os resultados revelaram diferenças significativas na composição de espécies entre as fitofisionomias, indicando que as formações definidas a priori têm conjuntos de espécies distintos. A Análise de Espécies Indicadoras (IndVal) identificou 136 espécies, das quais 27 foram priorizadas para o uso do AFS, contribuindo para um total de 84 espécies nativas prioritárias, incluindo aquelas com múltiplos usos. Essas descobertas destacam a importância de considerar as fitofisionomias locais ao planejar o SAF com espécies nativas nessa ecorregião..
Palavras-chave: Bioma Pampa; Espécies indicadoras; Quintais agroflorestais; SAFs restaurativos; Sistema silvipastoril
Agroforestry Systems (AFSs) encompass a wide range of land use practices that integrate ecological and economic objectives in agricultural and livestock productions (Nair, 1993; Ospina, 2003). Among the various AFS approaches, those that deliberately integrate perennial woody species (such as trees, shrubs, palms, and bamboos) with agricultural crops and/or livestock in a given spatial-temporal arrangement are the most widely adopted by the International Centre for Research in Agroforestry (ICRAF) (Lundgreen & Raintree, 1983; Atanga et al., 2014). Considering the three main components managed in an agroforestry system—perennial woody species, agricultural crops, and animals or pasture—Nair (1993) proposed the following classification: (1) Agrosilvopastoral systems, which include all three components; (2) Silvopastoral systems, which combine perennial woody plants with animals/pasture; and (3) Agrisilvicultural systems, which combine perennial woody plants with crops.
Perennial woody species are the essential components in AFSs, acting as catalysts to the system. Thus, the selection of suitable tree species to each AFS type is fundamental to achieve the objectives set when implementing or managing an agroecosystem. There is a recurring interest in the methodological approaches for key-species selection, as they tend to be varied and unsystematized (Fahad, et al., 2022; Lima et al., 2023). In general, these methods form two large groups: those that are participatory and those focused on ecological principles. Participatory methods are primarily based on traditional knowledge and values attributed by the local communities, using interviews, illustrated work sheets (Canosa et al., 2016), drawings and maps (Ferreira, 2014), or secondary data originally obtained by these approaches (Oliveira-Júnior & Cabreira, 2012). Methods based on ecological principles considers species characteristics in relation to ecological succession and forest stratification (Rebello & Sakamoto 2021), species natural distributions and their relationship with environmental filters (Wood & Burley 1991), or by selecting plant species through their functional traits (Moonen & Bàrberi 2008). In practice, a combination of both participatory and ecological principals approaches is common, and, in all cases, basic information about the species is essential.
In regions with limited experiences with AFSs, such as the ecoregion of the Uruguayan-Brazilian Pampa (sense Olson et al., 2001; modified by Sell, 2017), the lack of precise tree species information may limit agroecological transition initiatives. The Uruguayan-Brazilian Pampa, dominated by grasslands, has seen its natural forests being heavily pressured by agrarian systems, transforming vegetation remains into monocultures of soy, rice, and exotic tree species to support the commodity market (Miguel, 2009). Most arboreal monoculture systems found in the Uruguayan-Brazilian Pampa are of Eucalyptus spp., representing the first significant economic contribution of tree species in the ecoregion (De La Torre 2013). Eucalyptus plantations reduces water flow (Farley, et al. 2005), affects soil fertility (Yáñez Díaz et al. 2018), and modifies vegetation richness composition and coverage percentage (Milione, et al. 2024), promoting territorial conflicts that contributes to regional socio-environmental impacts (De La Torre, 2013). Implementing AFS with native species offers an alternative to eucalyptus monocultures. Thus, gathering information on native trees and its relationship with climatic variables is crucial to mitigate the agroecological transitions process, specially in the context of global change (Lima et al., 2023).
Studies that indicate tree species for ecological restoration projects or AFS at the regional scale are frequently restricted within country boundaries (e.g., Guarino et al., 2018). Within the Uruguayan-Brazilian Pampa ecoregion, previous documented forest formations diversities are either restricted to Uruguay (e.g., Paz & Bassagoda, 2002; Grela, 2004; Brazeiro et al., 2020) or to the Rio Grande do Sul state, in Southern Brazil (e.g., Vargas et al., 2022). Oliveira-Filho et al. (2015) is the main study which presented a broader context to this subject, analyzing both the Pampean and the Atlantic Forest. However, focused information on the Uruguayan-Brazilian Pampa Ecoregion is still lacking. Hence, the absence of studies that fully address this ecoregion while also considering its internal heterogeneity hinders integrated planning at the transnational level.
Ecological transition is a gradual process of transforming conventional agricultural systems into more sustainable and ecological practices (Altieri, 1995; Gliessman, 2005), being necessary in the face of the increasing conversion of ecosystems in the Uruguayan-Brazilian Pampa ecosystems. To apply this concept, uplifting technical information is required, guiding practical actions for seed collection, seedling production, and the design and implementation of AFS projects. Herein, we provide information on potentially important species, and their association with environmental variables, for the implementation of AFS projects in the Uruguayan-Brazilian Pampa. Using the information available in the NeoTropTree (NTT) database (Oliveira-Filho, 2017), we performed a general characterization of the studied region, elaborating a comprehensive species list for AFSs in the different phytophysiognomies within it. More specifically, we focused our study on addressing the following questions:
I. Do the forest formations defined a priori at a broader scale have a different species composition from that found in the Uruguayan-Brazilian Pampa Ecoregion?
II. What environmental variables are related to species composition differences within the study region?
III. Which species are the best indicators of the different phytophysiognomies?
IV. Which native species are potentially more important for the implementation of AFSs in the different phytophysiognomies within the Uruguayan-Brazilian Pampa Ecoregion?.
2.1 Study Area
The study area is the Uruguayan-Brazilian Pampa Ecoregion (Figure 1), delimited by the World Wide Fund for Nature (WWF) as the Uruguayan Savannas Ecoregion (sense Olson et al., 2001). This ecoregion is regionally inserted in the bioregion of the Río de la Plata Grasslands (or Pampa). To avoid the problems associated with the concept and delimitation of biomes, we opted to use the bioregion concept proposed by Gudynas (2002) and currently incorporated by WWF. According to this author, bioregions are “geographic spaces where there are homogeneous characteristics from an ecological point of view, with strong links between human populations and complementarities and similarities in the uses that humans make of these ecosystems” (Gudynas, 2002, p. 194). We believe that this is the most appropriate concept and regional classification system for analyzing agroforestry, as it includes cultural that transform the territory.
The Río de la Plata Grasslands (Pastizales del Río de la Plata - sense Soriano et al., 1992) are a complex of ecosystems with a predominance of open formations, found in Argentina, Uruguay and far southern Brazil. The forest formations occurs in mosaic with grasslands, but mainly within the Uruguayan and Brazilian territories. Considering the cultural identity of the people who inhabit the Río de la Plata Grasslands with the word Pampa and the territory delimited by this ecoregion, we adopted the adaptation proposed by Sell (2017), and refer to the Uruguayan Savannas Ecoregion as the Uruguayan-Brazilian Pampa Ecoregion.
Figure 1 – Map of the Uruguayan-Brazilian Pampa Ecoregion with the location of the NeoTropTree (NTT) sites considered in this study
Caption: The embedded legend identifies each site according to their main phytophysiognomy. The bioregion of the Río de la Plata Grasslands is highlighted in the map, including the Uruguayan-Brazilian Pampa, in dark gray, plus four additional ecoregions, in light gray. Data on forest remnants were obtained from Hofmann et al. (2015) for Rio Grande do Sul, base year 2015, and from Proyecto REDD+ Uruguay (2019) for Uruguay, base year 2016. Ecoregions boundaries from WWF (2012)
We used the sites of the NeoTropTree (NTT) database of the Uruguayan-Brazilian Pampa Ecoregion, as delimited by Olson et al. (2001), which approximately coincides with the delimitation of the Pampa Biome proposed by IBGE (2019) in Brazil and the entire territory of Uruguay (Figure 1). For the level of generalization proposed in this paper, we describe bellow the five main phytophysiognomies within our study area — Coastal Sandy Mosaic; Mixed Needle-broadleaved Forest; Savanoid-woodland Vegetation; Semideciduous Seasonal Forest of Atlantic Domain; Semideciduous Seasonal Forest of Pampean Domain — adapted from Oliveira-Filho (2015, 2017). Although two of these are not forest formations per se, they were considered because they include many tree or shrub species with potential usage in AFSs.
Coastal Sandy Mosaic (CSM) — disjunction of the Atlantic Forest lato sensu (Restinga Forest) and associated ecosystems that form a complex of vegetational formations that develop in the sandy substrate of the Coastal Plain. Synonyms or scope: Área de Formações Pioneiras and Floresta de Restinga [Area of Pioneer Formations and Restinga Forest] (IBGE, 2012); Nanofloresta Latifoliada Subtropical Marítima Perenifolia Costeira Arenosa [Subtropical Coastal Sandy Maritime Evergreen Broadleaved Dwarf-forest] (Oliveira-Filho, 2015); Bosque Latifoliado Subtropical de Planicie Costeiro [Subtropical Coastal Broadleaved Dwarf-forest], Bosque Psamófilo [Psammophilous Forest] or Bosque Costero [Coastal Forest] (Brazeiro et al., 2020).
Mixed Needle-broadleaved Forest (MNF) — disjunction of the Atlantic Forest lato sensu in the Pampa that includes species of the Seasonal Semideciduous Forest with the marked presence of Araucaria angustifolia associated with other species of the Antarctic element, as well as of the Neoantarctic and Holarctic elements (sense Waechter, 2002). Synonyms and scope: Floresta Ombrófila Mista [Mixed Ombrophilous Forest] or Floresta de Araucária [Araucaria Forest] (IBGE, 2012); Floresta Mista Lati-aciculifoliada Subtropical Estacional fria [Subtropical Seasonally cold Mixed Needle-broadleaved Forest] (Oliveira-Filho, 2015).
Savanoid-woodland Vegetation (SWV) — Savanna-like formation with the presence of forest and grassland synusia (generically treated in this study as a forest formation), with scattered trees that do not form a canopy and predominance of shrubs and continuous grassland cover. We substitute the word “savanna” with “savanoid” (savanna-like phytophysiognomy), because savanna is not a suitable term for the climate of the study region (Overbeck et al. 2015). Synonyms and scope: Savana-estépica [Savanna-Steppe] or Parque de Espinilho [Espinal Parkland] (IBGE, 2012); Savana Arbórea-arbustiva [Savanna-woodland] (Oliveira-Filho, 2015); Sabana Arbolada Subtropical [Subtropical Savanna-woodland] or Bosque Parque [Parkland Woods] (Brazeiro et al., 2020).
Seasonal Semideciduous Forest of Atlantic Domain (SSFPD) — Disjunction of the Atlantic Forest lato sensu in the Pampa marked by the impoverishment of tropical species with increasing latitude. The delimitation of the Atlantic Domain is that adopted in NeoTropTree (Oliveira-Filho, 2017), and applied by Oliveira-Filho et al. (2015). Synonyms and scope: Floresta Estacional Decidual and Floresta Estacional Semidecidual [Seasonal Deciduous Forest and Seasonal Semideciduous Forest] (IBGE, 2012); Floresta Latifoliada Subtropical Estacional Fria Semideciduifolia [Subtropical Seasonal cold Semideciduous Broadleaved Forest] (Oliveira-Filho, 2015).
Seasonal Semideciduous Forest of Pampean Domain (SSFPD) — Seasonal Forest with a lower contribution of tropical species typical of the Atlantic Domain and a greater presence of species from the Chacoan Domain (subxerophilous). Synonyms and scope: Floresta Estacional Decidual [Seasonal Deciduous Forest] (IBGE, 2012); Floresta Latifoliada Subtropical Estacional Fria Semideciduifolia [Subtropical Seasonal cold Semideciduous Broadleaved Forest] (Oliveira-Filho, 2015); Bosque Latifoliado Subtropical Serrano [Subtropical Lower hills Broadleaved Forest], Bosque Latifoliado Subtropical de Planicie [Subtropical Lowlands Broadleaved Forest], Bosque Ribereño [Riverine Forest], Bosque de Quebrada [Quebrada Forest], Bosque Serrano [Serrano Forest] and Bosque de Cornisa [Cornisa Forest] (Brazeiro et al., 2020).
2.2 Metadata
For the systematization of potential species for AFSs, the NTT database was used (Oliveira-Filho, 2017), which provides information on the occurrence of species (floristic list) and on environmental variables (bioclimatic, edaphic, geographic and phytophysiognomic) in the Neotropical Biogeographic Region. Arboreal and arborescent plant species included in the NTT are those with stems that can reach more than 3 meters in height without supporting themselves on other plants (Eisenlohr & Oliveira-Filho, 2015). This concept is compatible with that of perennial woody species used in AFSs (sense Lundgreen & Raintree, 1983), which includes palms, bamboos, etc. Each NTT site corresponds to a circular area with a radius of 5 km and includes published research information compiled by phytophysiognomy, such as phytosociological studies, taxonomic monographs and herbarium data.
The environmental variables available in the database were generated for the coordinates of the center of each NTT site with the following procedures: extraction of edaphic data from the Harmonized World Soil Database v. 1.1 (Fischer et al., 2009); extraction of bioclimatic data available in WorldClim (Hijmans et al., 2005) for the period from 1970 to 2000; estimation of aridity indices according to Zomer et al. (2006); estimation of periods of water deficit and excess based on Walter’s Diagram (Walter, 1985).
There are 106 NTT sites in the Uruguayan-Brazilian Pampa Ecoregion (Figure 1), with the following sampling distribution among the phytophysiognomies: 46 in the SSFPD, 37 in the SSFAD, 18 in the CSM, 3 in the SWV, and 2 in the MNF. The differences in the number of sampling sites in each forest phytophysiognomy are due to their differential area coverages within the studied ecoregion. Between sites, the mean annual temperature ranges from 16.3 to 21°C and the annual precipitation ranges from 943 to 1887 mm. The list of species obtained in the NTT had the nomenclature updated by the name accepted in the Flora e Funga do Brasil (2022) and in the Flora del Cono Sur (2022).
2.3 Statistical Analyses
To determine if phytophysiognomies (only the forest formations) found in the Uruguayan-Brazilian Pampa Ecoregion have different species compositions (question i), we used nonmetric multidimensional scaling (nMDS), an unrestricted ordination, with the distance measure (D = 1 – SJ), the complement of Jaccard’s similarity index (SJ). By reducing the multidimensionality of the data into two dimensions, we were able to maintain the relationship of the order of distance between the objects, where the most similar sites are the closest and the most dissimilar ones are further apart. A binary asymmetric coefficient was used to avoid the double-zero problem (Legendre & Legendre, 2012). The stress value (standard residuals sum of squares) was used as a measure of fit, ranging from 0 to 1, indicating the adjustment between the original distances (dissimilarities) from the distances of the nMDS (Clarke, 1993). As in Clarke (1993), the value of 0.2 was adopted as the acceptable upper limit of stress. Environmental variables with a significant goodness of fit statistic (p < 0.01) were adjusted to the nMDS to facilitate exploratory analysis (question ii).
The hypothesis of floristic distinction between forest formations (question i) was tested with the analysis of similarities (ANOSIM). Based on the order of distance relationship (dissimilarity) between the sampling units, ANOSIM tests the null hypothesis that there is no difference between the groups of samples defined a priori, assessing differences within and between the groups (Clarke 1993; Anderson 2001; Legendre & Legendre 2012). This method has been widely used in ecological studies to analyze community structures across different forest types. For instance, ANOSIM was applied to compare the basal area among three forest areas in tropical coniferous-forest ecotones (Zhang et al., 2014). Maçaneiro et al. (2019) used ANOSIM in an Atlantic subtropical rainforest to assess the statistical significance of floristic groups identified through NMDS. In tropical montane forests of Ecuador, distinct floristic groups were also identified using ANOSIM (Jadán et al.,2021).
The R statistic of the test ranks dissimilarities between values of -1 and 1, but usually ranges from 0 to 1. R values are close to zero if the null hypothesis is accepted. Positive R values signify dissimilarity between groups and the closer to 1, the more similar sites within a group are to each other than to other sites (Clarke, 1993). Statistical significance was obtained under 9,999 permutations, and post-hoc tests were performed for pairwise group comparisons.
The definition of the indicator species of the different phytophysiognomies (question iii) was made using the indicator species analysis proposed by Dufrene & Legendre (1997), which is expressed by the indication value index (IndVal) and its significance. Indicator species are those that are more characteristic of a group, found mainly in the group of a typology, being present in most of the group sites (Dufrene & Legendre, 1997). This concept expresses the relationship between the fidelity and specificity of a species. Fidelity is the degree to which a species is present in all sites of a group and specificity is the degree to which a species is found only in a group of sites (Legendre, 2013). The index ranges from 0 to 1 (or 0 to 100 %) and has the maximum value when individuals of the species are observed in all sites of only one group of sites (forest phytophysiognomies). The significance of the IndVal of each species was obtained by a random permutation procedure of the sites between groups to simulate and compare patterns expected in the absence of an ecological process (Dufrene & Legendre, 1997; Legendre & Legendre, 2012).
The selection of native species that are potentially important for the implementation of AFSs in the different phytophysiognomies within the study region (question iv) was based on the analysis of indicator species, distribution patterns in the different forest formations, environmental variables related to floristic differentiation between the phytophysiognomies, functional traits of the species, knowledge of the authors, and bibliographic research of studies carried out in the region (Reitz et al., 1983; Backes & Irgang, 2002; Guarino et al., 2018). Three main types of AFSs were considered for the study region: i - the Silvopastoral System, for association with traditional livestock activity; ii - Restorative AFS, considering the need to restore degraded areas; iii - Agroforestry homegardens, for diversity and food security in small properties. For each species, their predominant usages were indicated, considering the most characteristic AFS modalities of each forest phytophysiognomy, for example, for the typical species in Savanoid-woodland Vegetation, the Silvopastoral System was prioritized.
Analyses were performed primarily in the R statistical environment (R Core Team, 2022). The nMDS was performed with the “metaMDS” function and the environmental variables were adjusted by the “envfit” function, both from the “vegan” package (Oksanen et al., 2022). ANOSIM was performed in PAST (Hammer et al., 2001) with 9,999 iterations in the permutation test. IndVal and its significance were estimated by the “indval” function from the package “labdsv” (Roberts, 2022).
We recorded a total of 490 native tree and arborescent species within the 106 NTT sites located in the Uruguayan-Brazilian Pampa Ecoregion (see Table 1 SM – Supplementary Material). This value corresponds to the total species richness of the Uruguayan-Brazilian Pampa Ecoregion, and includes all species that occurred at least once in any of the NTT sites. The mean richness per NTT site was of 94.50 species (Table 1). Across phytophysiognomies, the mean richness ranged from 31 species in the SWV to 137.51 species in the SSFAD, where the later also presented the highest number of rare species (50 spp. with only one record in the 106 sites) and of restricted species (119 spp. found in only one of the forest phytophysiognomies) (Table 1, Figure 2).
Spatial distribution of systematized richness by phytophysiognomy is available in Figure 2. Species richness exhibits a clear decline along a north–south gradient. The highest richness values (exceeding 100 species) are concentrated in the northern portion of the Uruguayan–Brazilian Pampa Ecoregion. Further south, between the Cuareim and Negro Rivers, richness values vary substantially, ranging from 99 to 60 and as low as 20. The lowest values are observed in the southernmost region, between the Río Negro and the Río de la Plata.
Table 1 – Synthesis of the distribution of tree and arborescent species recorded in the 106 NTT sites by phytophysiognomy in the Uruguayan-Brazilian Pampa Ecoregion
|
SWV |
MNF |
SSFAD |
SSFPD |
CSM |
UY-BR |
|
|
Rare species |
0 |
7 |
50 |
10 |
4 |
71 |
|
Restricted species |
0 |
9 |
119 |
24 |
10 |
162 |
|
Indicator species |
17 |
97 |
13 |
3 |
6 |
136 |
|
Mean richness |
31.00 |
122.50 |
137.51 |
77.50 |
57.22 |
94.50 |
|
Standard deviation |
2.00 |
3.54 |
40.21 |
18.45 |
17.97 |
43.58 |
|
Minimum– maximum |
29–33 |
120–125 |
60– 291 |
40–123 |
20– 79 |
20–291 |
Source: Authors (2023). Legend: SWV = Savanoid-woodland Vegetation; MNF = Mixed Needle-broadleaved Forest; SSFAD = Seasonal Semideciduous Forest of Atlantic Domain; SSFPD = Seasonal Semideciduous Forest of Pampean Domain; CSM = Coastal Sandy Mosaic; Pampa UY-BR = Uruguayan-Brazilian Pampa Ecoregion; rare species = only one occurrence in the 106 NTT sites; restricted species = occurrence restricted to a forest physiognomy; indicator species = species that presented a significant IndVal (p < 0.05) for a forest physiognomy; mean richness = mean richness of the sites; standard deviation = standard deviation in relation to the mean richness; minimum – maximum = minimum and maximum richness in the sites
Figure 2 – Distribution of tree and shrub species richness across the Uruguayan-Brazilian Pampa Ecoregion (A), and across different phytophysiognomies (B)
Source: Authors (2023). Legend: SWV = Savanoid-woodland Vegetation; MNF = Mixed Needle-broadleaved Forest; SSFAD = Seasonal Semideciduous Forest of Atlantic Domain; SSFPD = Seasonal Semideciduous Forest of Pampean Domain; CSM = Coastal Sandy Mosaic. Box-plot: horizontal midline represents the median, the horizontal lines at the extremes are the maximum and minimum values. Outliers are plotted as asterisks. The quartiles of 25-75% are the boxes and the violin graph represents the kernel density
The non-Metric Scaling (nMDS) presented an adequate stress value of 0.157. The first two axes of multidimensional scaling (MDS1 and MDS2) demonstrates a clear distinction in species composition between the considered phytophysiognomies, showing no overlap between them (Figure 3). The first axis had a latitudinal gradient related to the Atlantic-Pampean floristic gradient, related to the species richness pattern seen in Figure 2A. Overall, the richest MNF and SSFAD phytophysionomies presented more negative MDS1 values than the SWV, CSM and SSFPD (with more positive values of MDS1, Figure 3). The MDS2 indicates an east-west gradient, where CSM (postive values, located at the east) stands out, showing no overlap in MDS2 values with the other phytophysionomies (negative values, Figure 3).
Figure 3 – Non-metric multidimensional scaling (nMDS) with two dimensions of the 106 NNT sites that occur in the Uruguayan-Brazilian Pampa Ecoregion and environmental variables with a significant goodness of fit statistic (p < 0.01)
Source: Authors (2023). Legend: SWV = Savanoid-woodland Vegetation; MNF = Mixed Needle-broadleaved forest; SSFAD = Seasonal Semideciduous Forest of Atlantic Domain; SSFPD = Seasonal Semideciduous Forest of Pampean Domain; CSM = Coastal Sandy Mosaic; A = altitude; AMT = annual mean temperature; DTR = Mean diurnal temperature range; I = isothermality; TS = temperature seasonality; MTW = maximum temperature of the warmest month; MTC = minimum temperature of the coldest month; ATR = annual temperature range; AP = annual precipitation; PWM= precipitation in the wettest month; PDM = precipitation in the driest month; SP = seasonality of precipitation; DWD= duration of water deficit periods; SWD = severity of water deficit periods; F = frost; S = ranked sand (soil texture); SF = soil fertility class; SS = soil salinity class
The nMDS-adjusted environmental variables (Figure 3) indicates three main sets of environmental gradients related to floristic differentiation between phytophysiognomy groups: 1. a gradient of more sandy and salinized soils, more characteristic of CSM; 2. a gradient with variables related to the greater variations in the southwest region by the effect of continentality and increasing latitude (temperature seasonality, precipitation seasonality, annual temperature range, mean diurnal temperature range, maximum temperature in the warmest month); and 3. another gradient related to the lower latitude and higher altitude regions (which have higher annual precipitation and in the driest and wettest month, a longer period and greater severity of excess water and isothermality).
The similarity analysis rejected the null hypothesis of no difference between the predefined phytophysionomies (R = 0.601 and p = 0.0001). With the exception of the SWV-MNF and SSFAD-MNF pairings, the post-hoc test results show significant differences on species compositions between phytophysiognomies (Table 2). Indicator species analysis resulted in 136 species with a significant IndVal for some of the phytophysiognomies (Table 1 SM – Supplementary Material). Indicator species analysis resulted in 136 species with a significant IndVal for some of the phytophysiognomies (Table 1 SM – Supplementary Material). Most species were indicative of the Mixed Needle-broadleaved Forest, while only a few were indicative of the Seasonal Semideciduous Forest of Pampean Domain. The complete list of indicator species is shown in Table 2 SM - Supplementary Material. From those, 27 priority indicator species were selected for their potential usage in AFSs across the Uruguayan-Brazilian Pampa Ecoregion (Table 3).
Table 2 – Results of post-hoc tests performed to make pairwise comparisons between pairs of groups (phytophysiognomies) in the analysis of similarity (ANOSIM). R statistic is presented above the diagonal, while p-values are shown bellow
|
p R |
MNF |
SSFAD |
CSM |
SSFPD |
SWV |
|
MNF |
0.495 |
0.857 |
0.650 |
1.000 |
|
|
SSFAD |
< 0.05 |
0.827 |
0.437 |
1.000 |
|
|
CSM |
< 0.01 |
< 0.01 |
0.648 |
0.945 |
|
|
SSFPD |
< 0.01 |
< 0.01 |
< 0.01 |
0.869 |
|
|
SWV |
> 0.05 |
< 0.01 |
< 0.01 |
< 0.01 |
Source: Authors (2023)
Legend: MNF = Mixed Needle-broadleaved Forest; SSFAD = Seasonal Semideciduous Forest of Atlantic Domain; SSFPD = Seasonal Semideciduous Forest of Pampean Domain; CSM = Coastal Sandy Mosaic, SWV = Savanoid-woodland Vegetation
Table 3 – List of indicator species of phytophysiognomies in the Uruguayan-Brazilian Pampa Ecoregion with significant indication value index - IndVal (p < 0.05) and potential use (priority) in AFSs
|
Group |
Species |
IndVal |
p |
|
SWV |
Prosopis nigra Hiron. |
0.94 |
0.001 |
|
SWV |
Aspidosperma quebracho-blanco Schltdl. |
0.90 |
0.001 |
|
SWV |
Prosopis affinis Spreng. |
0.84 |
0.001 |
|
SWV |
Parkinsonia aculeata L. |
0.69 |
0.001 |
|
SWV |
Vachellia caven (Molina) Seigler & Ebinger * |
0.54 |
0.001 |
|
SWV |
Vachellia farnesiana (L.) Wight & Arn. |
0.52 |
0.008 |
|
SWV |
Pouteria salicifolia (Spreng.) Radlk. |
0.44 |
0.003 |
|
SWV |
Ruprechtia salicifolia (Cham. & Schltdl.) A.C.Meyer |
0.42 |
0.046 |
|
SWV |
Scutia buxifolia Reissek |
0.30 |
0.021 |
|
MNF |
Araucaria angustifolia (Bertol.) Kuntze |
0.97 |
0.001 |
|
MNF |
Piptocarpha angustifolia Dusén ex Malme |
0.86 |
0.003 |
|
MNF |
Handroanthus albus (Cham.) Mattos |
0.83 |
0.003 |
|
MNF |
Ateleia glazioveana Baill. * |
0.79 |
0.007 |
|
MNF |
Ilex paraguariensis A.St.-Hil. |
0.77 |
0.002 |
|
MNF |
Erythrina falcata Benth. |
0.72 |
0.012 |
|
MNF |
Butia eriospatha (Mart. ex Drude) Becc. |
0.50 |
0.022 |
|
MNF |
Mimosa scabrella Benth. |
0.41 |
0.038 |
|
SSFAD |
Annona sylvatica A.St.-Hil. |
0.71 |
0.001 |
|
SSFAD |
Solanum pseudoquina A.St.-Hil. |
0.53 |
0.001 |
|
SSFAD |
Chrysophyllum gonocarpum (Mart. & Eichler ex Miq.) Engl. |
0.44 |
0.032 |
|
SSFAD |
Nectandra oppositifolia Nees |
0.43 |
0.029 |
|
SSFAD |
Inga marginata Willd. |
0.42 |
0.033 |
|
SSFPD |
Celtis tala Gillies ex Planch. |
0.47 |
0.049 |
|
CSM |
Myrsine parvifolia A.DC. |
0.89 |
0.001 |
|
CSM |
Annona maritima (Záchia) H.Rainer |
0.76 |
0.002 |
|
CSM |
Psidium cattleianum Sabine |
0.50 |
0.001 |
|
CSM |
Ficus cestrifolia Schott ex Spreng. |
0.47 |
0.001 |
Source: Authors (2023). Legend: MNF = Mixed Needle-broadleaved Forest; SSFAD = Seasonal Semideciduous Forest of Atlantic Domain; SSFPD = Seasonal Semideciduous Forest of Pampean Domain; CSM = Coastal Sandy Mosaic; SWV = Savanoid-woodland Vegetation; * = species with use restriction due to the potential to be invasive on grasslands formations
From the 490 species considered in this study, 153 species were selected for use in AFSs (Table 3 SM – Supplementary Material). The selected species include the 27 priority indicator species plus 57 priority non-indicator species, and 69 non-priority and non-indicator species (Table 4). These 69 species are multipurpose, with no established uses yet, but hold significant potential for future applications.
Table 4 – List of priority species for use in AFSs in the Uruguayan-Brazilian Pampa Ecoregion. The percentage value in the phytophysiognomies indicates the frequency of occurrence of the species in the sites
|
Family |
Species |
SWV |
MNF |
SFA |
SFP |
CSM |
LF |
Uses |
|
Anacardiaceae |
Astronium balansae Engl. |
0% |
0% |
14% |
7% |
0% |
Tre |
Sil Res Woo MePo Con EcSu Orn Fir |
|
Anacardiaceae |
Schinus molle L. |
33% |
0% |
78% |
70% |
6% |
Tre |
Sil Res Woo Med MePo EcSu Orn Fir |
|
Anacardiaceae |
Schinus terebinthifolia Raddi |
0% |
100% |
41% |
4% |
17% |
Tre |
Res Foo Woo Med MePo EcSu Orn Fir |
|
Annonaceae |
Annona maritima (Záchia) H.Rainer |
0% |
0% |
0% |
2% |
78% |
Shr Tre |
AgHo Foo |
|
Annonaceae |
Annona neosalicifolia H.Rainer |
0% |
100% |
57% |
37% |
0% |
Tre |
AgHo Foo |
|
Annonaceae |
Annona sylvatica A.St.-Hil. |
0% |
0% |
86% |
20% |
0% |
Tre |
Res AgHo Foo |
|
Apocynaceae |
Aspidosperma quebracho-blanco Schltdl. |
100% |
0% |
0% |
11% |
0% |
Tre |
Sil Woo Con Orn |
|
Aquifoliaceae |
Ilex paraguariensis A.St.-Hil. |
0% |
100% |
22% |
9% |
0% |
Tre |
Sil Res AgHo Foo |
|
Araucariaceae |
Araucaria angustifolia (Bertol.) Kuntze |
0% |
100% |
3% |
0% |
0% |
Tre |
Sil Res AgHo Foo Woo Con Orn |
|
Arecaceae |
Bactris setosa Mart. |
0% |
0% |
22% |
0% |
6% |
Pal |
AgHo Foo MePo Con |
|
Arecaceae |
Butia eriospatha (Mart. ex Drude) Becc. |
0% |
50% |
0% |
0% |
0% |
Pal |
AgHo Foo MePo Cra Con Orn |
|
Arecaceae |
Butia odorata (Barb.Rodr.) Noblick |
0% |
0% |
49% |
33% |
33% |
Pal |
AgHo Foo MePo Cra Con Orn |
|
Arecaceae |
Butia yatay (Mart.) Becc. |
33% |
0% |
11% |
50% |
0% |
Pal |
AgHo Foo MePo Cra Con Orn |
|
Arecaceae |
Geonoma schottiana Mart. |
0% |
0% |
24% |
0% |
39% |
Pal |
AgHo Foo MePo Con |
|
Arecaceae |
Syagrus romanzoffiana (Cham.) Glassman |
0% |
100% |
92% |
61% |
67% |
Pal |
Foo MePo Orn |
|
Asteraceae |
Baccharis dracunculifolia DC. |
33% |
50% |
57% |
50% |
6% |
Shr |
Res MePo Med EcSu |
|
Asteraceae |
Moquiniastrum polymorphum (Less.) G. Sancho |
0% |
100% |
86% |
57% |
0% |
Tre Shr |
Res MePo Woo Med EcSu Orn |
|
Asteraceae |
Piptocarpha angustifolia Dusén ex Malme |
0% |
100% |
16% |
0% |
0% |
Tre Shr |
Res EcSu |
|
Bignoniaceae |
Handroanthus albus (Cham.) Mattos |
0% |
100% |
16% |
4% |
0% |
Tre |
Res Woo Orn |
|
Bignoniaceae |
Handroanthus heptaphyllus (Vell.) Mattos |
0% |
0% |
24% |
26% |
0% |
Tre |
Res Woo Med Orn |
|
Boraginaceae |
Cordia americana (L.) Gottschling & J.S.Mill. |
0% |
100% |
84% |
46% |
6% |
Tre |
Res Woo Orn |
|
Boraginaceae |
Cordia trichotoma (Vell.) Arráb. ex Steud. |
0% |
100% |
57% |
4% |
0% |
Tre |
Res Woo Orn |
|
Cactaceae |
Cereus uruguayanus R. Kiesling |
0% |
0% |
0% |
26% |
56% |
Shr Suc |
AgHo Foo Orn |
|
Cactaceae |
Pereskia nemorosa Rojas Acosta |
0% |
0% |
0% |
20% |
0% |
Shr |
AgHo Foo Orn |
|
Cannabaceae |
Celtis tala Gillies ex Planch. |
0% |
0% |
22% |
63% |
0% |
Tre |
Sil Res Foo Orn EcSu |
|
Cannabaceae |
Trema micrantha (L.) Blume |
0% |
50% |
68% |
15% |
0% |
Tre |
Res EcSu Sil Woo Fir |
|
Caricaceae |
Vasconcellea quercifolia A.St.-Hil. |
0% |
100% |
41% |
11% |
0% |
Shr Tre |
AgHo Foo Med |
|
Celastraceae |
Monteverdia ilicifolia (Mart. ex Reissek) Biral |
67% |
50% |
57% |
78% |
0% |
Shr Tre |
AgHo Med Orn |
|
Euphorbiaceae |
Gymnanthes klotzschiana Müll.Arg. |
0% |
100% |
100% |
91% |
89% |
Tre |
Res EcSu Fir |
|
Euphorbiaceae |
Sapium haematospermum Müll.Arg. |
67% |
0% |
19% |
33% |
0% |
Tre |
Res Orn |
|
Fabaceae |
Apuleia leiocarpa (Vogel) J.F.Macbr. |
0% |
50% |
43% |
7% |
0% |
Tre |
Woo Con Orn |
|
Fabaceae |
Ateleia glazioveana Baill. * |
0% |
100% |
27% |
0% |
0% |
Tre |
Res EcSu Fir |
|
Fabaceae |
Calliandra tweedii Benth. |
0% |
50% |
76% |
74% |
11% |
Shr Tre |
Res Orn |
|
Fabaceae |
Enterolobium contortisiliquum (Vell.) Morong |
0% |
0% |
43% |
37% |
11% |
Tre |
Res Woo EcSu Orn |
|
Fabaceae |
Erythrina crista-galli L. |
67% |
0% |
54% |
65% |
72% |
Tre |
Res MePo Orn |
|
Fabaceae |
Erythrina falcata Benth. |
0% |
100% |
46% |
0% |
6% |
Tre |
Res MePo Orn |
|
Fabaceae |
Inga marginata Willd. |
0% |
0% |
57% |
4% |
0% |
Tre |
Res Foo EcSu Orn Fir |
|
Fabaceae |
Inga vera Willd. |
0% |
0% |
49% |
41% |
11% |
Tre |
Res Foo MePo Riv Orn |
|
Fabaceae |
Mimosa bimucronata (DC.) Kuntze |
0% |
0% |
32% |
11% |
28% |
Shr Tre |
Res EcSu MePo Fir |
|
Fabaceae |
Mimosa scabrella Benth. |
0% |
50% |
8% |
2% |
0% |
Tre |
Res EcSu MePo Woo Fir |
|
Fabaceae |
Myrocarpus frondosus Allemão |
0% |
100% |
46% |
13% |
0% |
Tre |
Res MePo Woo Med Con |
|
Fabaceae |
Parapiptadenia rigida (Benth.) Brenan |
0% |
100% |
0% |
33% |
0% |
Tre |
Res MePo Woo Med Fir |
|
Fabaceae |
Parkinsonia aculeata L. |
100% |
0% |
16% |
39% |
6% |
Shr Tre |
Sil MePo Woo Orn Fir |
|
Fabaceae |
Peltophorum dubium (Spreng.) Taub. |
0% |
50% |
0% |
24% |
0% |
Tre |
Res Woo Orn Fir |
|
Fabaceae |
Prosopis affinis Spreng. |
100% |
0% |
0% |
20% |
0% |
Shr Tre |
Sil MePo Woo Med Foo Fir |
|
Fabaceae |
Prosopis nigra Hiron. |
100% |
0% |
22% |
7% |
0% |
Shr Tre |
Sil MePo Woo Med Con Fir |
|
Fabaceae |
Vachellia caven (Molina) Seigler & Ebinger * |
100% |
0% |
14% |
57% |
6% |
Shr Tre |
Sil MePo Con Fir |
|
Fabaceae |
Vachellia farnesiana (L.) Wight & Arn. |
67% |
0% |
0% |
20% |
0% |
Shr Tre |
Sil MePo Fir |
|
Lamiaceae |
Aegiphila brachiata Vell. |
0% |
50% |
100% |
11% |
0% |
Shr Tre |
Res Foo |
|
Lamiaceae |
Vitex megapotamica (Spreng.) Moldenke |
0% |
100% |
3% |
61% |
67% |
Tre |
Res Woo Med Orn |
|
Lauraceae |
Nectandra oppositifolia Nees |
0% |
0% |
43% |
0% |
0% |
Tre |
Res Foo |
|
Lauraceae |
Ocotea catharinensis Mez |
0% |
0% |
24% |
0% |
6% |
Tre |
Woo Con Orn |
|
Lauraceae |
Ocotea pulchella (Nees & Mart.) Mez |
0% |
100% |
84% |
48% |
72% |
Tre |
Res EcSu |
|
Malvaceae |
Luehea divaricata Mart. & Zucc. |
0% |
100% |
81% |
78% |
39% |
Tre |
Res MePo Woo Med Orn Riv |
|
Meliaceae |
Cedrela fissilis Vell. |
0% |
100% |
62% |
15% |
6% |
Tre |
Res Woo Med Con Orn |
|
Moraceae |
Ficus cestrifolia Schott ex Spreng. |
0% |
0% |
62% |
17% |
89% |
Tre |
Res Woo Con Orn |
|
Myrtaceae |
Campomanesia guazumifolia (Cambess.) O.Berg |
0% |
100% |
95% |
9% |
0% |
Tre |
Res AgHo Foo MePo Woo Orn Fir |
|
Myrtaceae |
Campomanesia xanthocarpa (Mart.) O.Berg |
0% |
100% |
89% |
46% |
11% |
Tre |
Res AgHo Foo MePo Woo Orn Fir |
|
Myrtaceae |
Eugenia involucrata DC. |
0% |
100% |
51% |
28% |
11% |
Tre |
Res AgHo Foo MePo Woo Orn |
|
Myrtaceae |
Eugenia myrcianthes Nied. |
0% |
0% |
32% |
41% |
39% |
Tre |
Res AgHo Foo MePo Woo Orn Fir |
|
Myrtaceae |
Eugenia pyriformis Cambess. |
0% |
50% |
62% |
4% |
0% |
Tre |
Res AgHo Foo MePo Woo Orn Fir |
|
Myrtaceae |
Eugenia rostrifolia D.Legrand |
0% |
50% |
92% |
11% |
0% |
Tre |
Foo MePo Woo Orn Fir |
|
Myrtaceae |
Eugenia uniflora L. |
67% |
50% |
43% |
80% |
61% |
Tre |
Res AgHo Foo MePo Med EcSu Orn Fir |
|
Myrtaceae |
Feijoa sellowiana (O.Berg) O.Berg |
0% |
50% |
81% |
48% |
0% |
Tre Shr |
AgHo Foo MePo Orn Fir |
|
Myrtaceae |
Myrcianthes pungens (O.Berg) D.Legrand |
67% |
100% |
32% |
63% |
6% |
Tre |
Res Foo MePo Woo Fir |
|
Myrtaceae |
Plinia rivularis (Cambess.) Rotman |
0% |
0% |
57% |
33% |
0% |
Tre |
Foo Res Riv Woo Fir |
|
Myrtaceae |
Psidium cattleianum Sabine |
0% |
0% |
38% |
26% |
94% |
Tre Shr |
Res AgHo Foo MePo Woo Orn Fir |
|
Podocarpaceae |
Podocarpus lambertii Klotzsch ex Endl. |
0% |
0% |
8% |
7% |
0% |
Tre |
Res Foo Woo Orn EcSu |
|
Polygonaceae |
Ruprechtia salicifolia (Cham. & Schltdl.) A.C.Meyer |
67% |
0% |
84% |
30% |
0% |
Tre |
Res Riv Woo Fir |
|
Primulaceae |
Myrsine coriacea (Sw.) R.Br. ex Roem. & Schult. |
0% |
50% |
22% |
57% |
17% |
Tre |
Res EcSu |
|
Primulaceae |
Myrsine parvifolia A.DC. |
0% |
0% |
0% |
0% |
89% |
Shr |
Res EcSu |
|
Quillajaceae |
Quillaja lancifolia D.Don |
0% |
0% |
59% |
54% |
6% |
Tre |
Res Med Woo Con EcSu Fir |
|
Rhamnaceae |
Scutia buxifolia Reissek |
100% |
0% |
27% |
87% |
83% |
Tre Shr |
Sil Res Woo Med Fir |
|
Rutaceae |
Helietta apiculata Benth. |
0% |
100% |
97% |
22% |
0% |
Tre |
Res Woo Med EcSu |
|
Salicaceae |
Casearia sylvestris Sw. |
0% |
100% |
27% |
74% |
72% |
Tre |
Res MePo Med |
|
Salicaceae |
Salix humboldtiana Willd. |
0% |
0% |
27% |
48% |
39% |
Tre |
Res Riv MePo Med Orn |
|
Santalaceae |
Acanthosyris spinescens (Mart. & Eichler) Griseb. |
67% |
0% |
100% |
54% |
0% |
Tre |
Sil Foo Orn |
|
Sapindaceae |
Dodonaea viscosa Jacq. |
0% |
0% |
41% |
39% |
67% |
Shr SubShr |
Res EcSu Fir |
|
Sapotaceae |
Chrysophyllum gonocarpum (Mart. & Eichler ex Miq.) Engl. |
0% |
0% |
73% |
37% |
11% |
Shr Tre |
Foo |
|
Sapotaceae |
Pouteria salicifolia (Spreng.) Radlk. |
100% |
0% |
46% |
54% |
28% |
Tre |
Res Riv |
|
Sapotaceae |
Sideroxylon obtusifolium (Roem. & Schult.) T.D.Penn. |
33% |
0% |
76% |
20% |
83% |
Tre Shr |
Foo Med Com |
|
Solanaceae |
Solanum mauritianum Scop. |
0% |
100% |
30% |
46% |
39% |
Tre Shr |
Res EcSu |
|
Solanaceae |
Solanum pseudoquina A.St.-Hil. |
0% |
0% |
78% |
9% |
28% |
Tre |
Res EcSu |
|
Urticaceae |
Cecropia pachystachya Trécul |
0% |
0% |
22% |
2% |
44% |
Tre |
Res EcSu Orn |
Source: Authors (2023). Legend: MNF = Mixed Needle-broadleaved forest; SFA = Seasonal Semideciduous Forest of Atlantic Domain; SFP = Seasonal Semideciduous Forest of Pampean Domain; CSM = Coastal Sandy Mosaic; SWV = Savanoid-woodland Vegetation; LF = life form; Shr = shrub; Tre = tree; Pal = palm; Subshr = subshrub; Suc = succulent; AgHo = Agroforestry homegardens; Res = Restorative AFS; Sil = Silvopastoral AFS; Foo = food; Cra = craft; Con = conservation; Fir = firewood; Woo = wood; Med = medicinal; MePo = melliferous/pollinators; Ole = oleiferous; Orn = ornamental; EcSu = ecological succession (pioneer, early secondary); Riv = riverine; * = species with use restriction due to the potential to be invasive on grassland formations
The results of this study demonstrate the importance of considering the local phytophysiognomies when planning the implementation of AFSs with native species. The five forest phytophysiognomies analyzed have different floristic compositions, which, on a broad scale, are congruent with the latitudinal diversity gradient and the expected effect of continentality on species assemblages (Stevens 1989; de Aledo et al. 2023). Within the Uruguayan-Brazilian Pampa Ecoregion, species assemblages consistently differ between phytophysiognomies, allowing us to select the most adequate native species for AFSs. Through this study, we show that the diversity of native species in the Uruguayan-Brazilian Pampa Ecoregion can potentially be used to our advantage while planning and developing AFSs, keeping agroecosystem functionality while also applying positive conservation actions for maintaining this diversity.
Across the Uruguayan-Brazilian Pampa Ecoregion, this study has shown that native tree and arborescent species richness increases towards the north and northeast. By following a clear gradient, our results implicates on practical solutions to AFS design. In the Savanoid-woodland Vegetation, a AFS designed with 20 species would be a good representative of the local species diversity, while, for other phytophysiognomies, such as the Mixed Needle-broadleaved Forest or the Semideciduous Seasonal Forest of Atlantic Domain, the same number of species would impoverish the local species expected pool. In AFSs, complementary species can increase ecosystem functions in the short term, especially in sites with lower species richness (Moonen & Bàrberi, 2008; Arshad, 2023). Meanwhile, redundant species can increase the resilience and stability of the agroecosystem in the long term, while the possibility remains to reverse functional redundancy over time after environmental changes or disturbances (Moonen & Bàrberi, 2008). Hence, by observing the local pool of species prior to their selection for AFS is an important tool for the functionality of agroecosystems.
The floristic differentiation presented between the different phytophysiognomies demonstrates the importance of recognizing the heterogeneity within the Uruguayan-Brazilian Pampa Ecoregion. The native forests in the Pampa are generally neglected in relation to their economic potential or diversity, and are generically treated as Riverine, Alluvial, Pampa Forests, etc. (IBGE, 2012; Oliveira-Filho et al. 2015). Due to scale-dependent processes, the different classification systems available (biogeographic provinces, domains, etc.) often implicate on underestimated differences in vegetation composition for the ecoregion, altering our perception of the occurrence and classification of forests in the Pampa (Vargas & Brack, 2021). The ecoregion concept has the advantage of not being strongly influenced by the classification of vegetation, while it includes the history of human occupation and transformation of the territory. Thus, we avoided a discussion on biome boundaries and adopted a classification system that allows the occurrence of distinct forest phytophysiognomies by highlighting the cultural identity of the people within the territory.
The distinct floristic compositions across the phytophysiognomies and their relationship with sets of environmental variables allowed us to establish different strategies for the design of AFSs in the Uruguayan-Brazilian Pampa Ecoregion. In the regions where the Savanoid-woodland Vegetation occur, greater variation in temperature and rainfall variables is expected, benefiting the predominance of livestock activity, and, hence, the main implementation of Silvopastoral Systems. The tree and shrub species that occur naturally in consortium with the grassland matrix in such environments can bring greater thermal comfort to cattle, benefiting pasture composition, while also providing sufficient wood quality for fence posts and firewood in a region with limited supply (Patt & Aian, 2008; Nascimento et al., 2018). In the Mixed Needle-broadleaved Forest region, implementing AFS for the production of pine nuts, yerba mate, native fruits, wood and firewood can also be associated with livestock in a Silvopastoral System similar to the Caívas (Hanisch, 2018). Although MNF occurs only in the extreme north of the Uruguayan-Brazilian Pampa Ecoregion, tree species typical to the MNF, such as Araucaria angustifolia (Bertol.) Kuntze and Ilex paraguariensis A.St.-Hil., are also present in the Serra do Sudeste plateau in Rio Grande do Sul (Carlucci et al., 2011), and may be used in future AFS. The same cannot be said to other species, such as Mimosa scabrella Benth. and Butia eriospatha (Mart. ex Drude) Becc., that are not found in the Serra do Sudeste.
Herein, the use of the NTT database (Oliveira-Filho, 2017) and the profile of the Uruguayan-Brazilian Pampa Ecoregion allowed us to address the heterogeneity of Pampa phytophysiognomies in order to select species for use in AFSs. By applying a metadata approach to a broad region, we acknowledge a certain level of generalization that was added to our results, which does not necessarily translates to some characteristic formations such as the Butiazais (Marchiori et al., 1995), the Pau-ferro woodlands (Longhi et al., 1987) and the Mixed forests with Araucaria in the Serra do Sudeste (Carlucci et al., 2011) or with Podocarpus (Longhi et al., 1992). Still, the species occurring in these formations were included in the sites located in their surroundings, where a reinterpretation of our results would sufficiently be applicable during AFS planning.
Our study emphasizes the importance of considering the different phytophysiognomic contexts while designing future AFSs. Several tree and shrub species currently or historically used in agroforestry systems are among the most widespread and ecologically damaging plant invaders, highlighting the importance of selecting native species to prevent long-term ecological disruption (Richardson et al. 2004). These authors suggested that even when native species may offer lower agroforestry performance compared to exotic alternatives, their use is preferable due to their lower risk of invasiveness and greater ecological compatibility. Building on this principle, the present study not only compiles a list of native species but also evaluates their functional roles within agroforestry systems (AFSs), thereby supporting more informed and ecologically responsible decision-making. This approach allows practitioners to balance functionality with ecological fit when selecting species for AFS implementation in the Uruguayan–Brazilian Pampa Ecoregion. In this context, greater attention should be paid to species with a strong potential to colonize native fields, such as Vachellia caven (Molina) Seigler & Ebinger and Ateleia glazioveana Baill (Carvalho 2003). In such cases, our list of priority species can be used as an important tool while implementing or managing AFSs in the Uruguayan-Brazilian Pampa Ecoregion. In contrast to other available lists, our study included species that are important indicatives of local phytophysiognomies, that have consolidated or potential use, and/or have multiple purposes. Hence, it foresees working with a reduced set of species that sufficiently represent the local biodiversity and are already adequate to local climatic conditions. We expect our list not only will help maintaining the local floristic characteristics, but it will also facilitate and diminishes the expected costs of AFS implementation in the Uruguayan-Brazilian Pampa Ecoregion.
This study is the first to foster a transnational discussion on AFS planning based on the shared environmental characteristics between southern Brazil and Uruguay. The references for the approach adopted included the patterns of species richness and composition in the study region, allowing us to establish an overview of its distribution. Adding to the detailed information reported in previous studies (e.g., Brazeiro et al., 2020; Guarino et al., 2018), we provide novel views on how to approach AFS while considering the heterogeneity within the Uruguayan-Brazilian Pampa Ecoregion. Moving forward in this ongoing discussion, we propose the adoption of an intermediate level-approach between generalization and specification. To this aim, integrating finer-scaled and experiment-based data between the two countries, while using our proposal as a base, will be an important path to pursue while applying ecologically-based AFS strategies in the ecoregion.
We thank the Binational Specialization in Agroecology between the Universidade Estadual do Rio Grande do Sul and Universidad de la República Uruguay.
Ali, A. (2023). Biodiversity–ecosystem functioning research: Brief history, major trends and perspectives. Biological Conservation, 285, 110210.
Altieri, M. A. (1995). Agroecology: The Science of Sustainable Agriculture. CRC Press. A foundational book that establishes agroecology as a scientific approach to sustainable agriculture.
Andrade, C. M. S., Salman, A. K. D., Gama, M. D., Oliveira, L. C., Oliveira, T. K., Mendes, A., & Assis, G. M. L. (2012). Método de seleção de espécies arbóreas para sistema silvipastoris. VII Congresso Latinoamericano de sistemas agroflorestais para a produção pecuária sustentável. UFPA.
Atanga, A., Khasa, D., Chang, S., & Degrande, A. (2014). Tropical Agroforestry. Springer.
Backes, P., & Irgang, B. (2002). Árvores do Sul: guia de identificação e interesse ecológico. Instituto Souza Cruz, 332 p.
Boccardo, B. L., Bernardi, L., Miguel, C., Olivera, J., Penego, C., & Rama, G. (2019). Ordenamiento Territorial y Medio Ambiente. Ministerio de Ganadería, Agricultura y Pesca - Ministerio de Vivienda. https://www.gub.uy/ministerio-ganaderia-agricultura-pesca/monitoreo-bosques
Brazeiro, A. (2020). Bosques nativos de Uruguay: distribución, diversidad & propuesta de clasificación. Plantae, v. 3, 19-36.
Canosa, G. A., Cassarino, J. P., & Leandrini, J. (2016). Uso de fichas ilustradas para seleção de espécies arbóreas nativas no planejamento participativo de sistemas agroflorestais com famílias agricultoras do Núcleo Luta Camponesa da Rede Ecovida de Agroecologia, PR. Desenvolvimento e Meio Ambiente, v. 39, p. 133-157.
Carvalho, G. (2020). Flora: Tools for interacting with the Brazilian Flora 2020. R package version 0.3.4. https://cran.r-project.org/package=flora
Carvalho, P. E. (2003). Espécies Arbóreas Brasileiras. Brasília: Embrapa Informação Tecnológica, 1039 p.
Cheng, K., Zang, R., & Ding, Y. (2014). Changes in floristic composition, community structure, and species diversity across a tropical coniferous-broadleaved forest ecotone. Tropical Conservation Science, v. 7, n. 1, p. 104–118. https://doi.org/10.1177/194008291400700104
Clarke, K. R. (1993). Non‐parametric multivariate analyses of changes in community structure. Australian journal of ecology, v. 18, n. 1) p.117-143.
de Aledo, J. G., Paneghel, M., Cayuela, L., Matas-Granados, L., Saadi, C. B., Salinas, N., La Torre-Cuadros, M. d. l. Á., García-Villacorta, R., & Macía, M. J. (2023). Floristic diversity, composition and dominance across Amazonian forest types respond differently to latitude. Journal of Biogeography, v. 50, p. 673–698. https://doi.org/10.1111/jbi.14561
De La Torre, W. G. (2013). Monocultivos de eucalipto e florestas: conceitos inconciliáveis. Revista Eletrônica da Associação dos Geógrafos Brasileiros, v. 17, p. 116-128.
Dufrêne, M., & Legendre, P. (1997). Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecological monographs, v. 67, n 3, p. 345-366.
Eisenlohr, P. V. & Oliveira-Filho, A. T. (2015). Obtenção e estruturação de metadados para trabalhos fitogeográficos de síntese e o banco de dados NeoTropTree como caso de estudo. In: Eisenlohr, P. V., Felfili, J. M., Melo, M. M. R. F., Andrade, L. A. & Meira Neto, J. A. A. (Eds.). Fitossociologia no Brasil: métodos e estudos de casos. v. 2, Editora UFV. p. 387-411.
Fahad, S., Chavan, SB, Chichaghare, AR, Uthappa, AR, Kumar, M., Kakade, V., Pradhan, A., Jinger, D., Radale, G., Yadav, DK, Kumar, V., Farooq, TH, Ali, B., Sawant, AV, Saud, S., Chen, S., & Poczai, P. (2022). Sistemas Agroflorestais para Melhoria e Manutenção da Saúde do Solo. Sustentabilidade, v. 14, n. 22, 14877.
Farley, K. A.; Jobbágy, E. G.; Jackson, R. B. (2005). Effects of afforestation on water yield: A global synthesis with implications for policy. Global Change Biology. v. 11, n10, p. 1565– 1576.
Ferreira, L. R. (2014). As agroflorestas como expressões de desenvolvimento rural no Rio Grande do Sul: uma análise a partir da produção de novidades. [Dissertação de Mestrado, Universidade Federal do Rio Grande do Sul]. Repositório da UFRGS. https://lume.ufrgs.br/handle/10183/109257
Fischer, G., Nachtergaele, F. O., Prieler, S., Teixeira, E., Tóth, G., Velthuizen, H. V., Velrest, L., & Wiberg, D. (2008). Global Agro-ecological Zones Assessment for Agriculture (GAEZ 2008). FAO.
Flora Del Cono Sur. Instituto de Botánica Darwinion. http://conosur.floraargentina.edu.ar
Flora e Funga do Brasil. Jardim Botânico do Rio de Janeiro. http://floradobrasil.jbrj.gov.br
Fundação Instituto Brasileiro de Geografia e Estatística (2012). Manual técnico da vegetação brasileira. IBGE.
Fundação Instituto Brasileiro de Geografia e Estatística. (2019). Biomas e sistema costeiro-marinho do Brasil: compatível com a escala 1:250.000. IBGE.
Gliessman, S. R. (2007). Agroecology: The Ecology of Sustainable Food Systems (2nd ed.). CRC Press.
Grela, I. A. (2004). Geografía florística de las especies arbóreas de Uruguay: propuesta para la delimitación de dendrofloras. [Tese de Doutorado, Universidad de la República].
Guarino, E. S. G., Overbeck, G. E., Müller, S. C., & Rovedder, A. (2018). Espécies de Plantas Prioritárias para Projetos de Restauração Ecológica em Diferentes Formações Vegetais no Bioma Pampa: Primeira Aproximação. Embrapa Clima Temperado.
Gudynas, E. (2002). El concepto de Regionalismo Autónomo y el desarrollo sustentable en el Cono Sur. In: Gudynas, E. (Ed.). Sustentabilidad y regionalismo en el Cono Sur. p. 177-211. Coscoroba Ediciones.
Hammer, Ø. (1991). PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica. V. 4, n 1, p. 1- 9.
Hanisch, A. L. (2018). Intensificação do uso da pastagem em Sistema Silvipastoril tradicional (Caívas) como estratégia de uso sustentável de remanescentes de Floresta Ombrófila Mista. [Tese de Doutorado, Universidade Federal do Paraná].
Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G., & Jarvis, A. (2005). Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology: A Journal of the Royal Meteorological Society, v. 25, n. 15, p.1965-1978.
Hofmann, G. S., Weber, E. J., Hasenack. (2015). Uso e cobertura vegetal do Estado do Rio Grande do Sul - situação em 2015. UFRGS IB Centro de Ecologia. http://www.ufrgs.br/labgeo
Jadán, O., Donoso, D. A., Cedillo, H., Bermúdez, F., & Cabrera, O. (2021).
Floristic groups, and changes in diversity and structure of trees, in tropical montane forests in the southern Andes of Ecuador. Diversity, v.13, n. 9. https://doi.org/10.3390/d13090400
Legendre, P., & Legendre, L. (2012). Numerical Ecology. Elsevier.
Legendre, P. (2013). Indicator species: computation. In: Levin, S. A. (Ed.). Encyclopedia of Biodiversity. 2a ed., v. 4, p. 264-268. Elsevier – Academic Press.
Lima, V. P., Joner, F., Raes, N., & Siddique, I. (2023). Integrating climate change into agroforestry conservation: A case study on native plant species in the Brazilian Atlantic Forest. Journal of Applied Ecology, v. 60, n. 9, p. 1977-1994. https://doi.org/10.1111/1365-2664.14464
Longhi, S. J. (1987). Aspectos fitossociológicos de uma floresta natural de Astronium balansae Engl., no Rio Grande do Sul. Revista do Centro de Ciências Rurais, v. 17, n 1.
Longhi, S. J., Selle, G. L., Ragagnin, L. I. M., & Damiani, J. E. (1992). Composição florística e estrutura fitossociológica de um “capão” de Podocarpus lambertii Klotz., no Rio Grande do Sul. Ciência Florestal, v. 2, p. 9-26.
Lundgreen, B. O, & Raintree, J. B. (1983). Sustained agroforestry. In: Nestel, B. (Ed.). Agricultural research for development: potentials and challenges in Asia. p. 37-49. ISNAR.
Maçaneiro, J. P., Liebsch, D., de Gasper, A. L., Galvão, F., & Schorn, L. A. (2019). Structural and floristic variations in an Atlantic subtropical rainforest in southern Brazil. Floresta e Ambiente, v. 26, n. 1, e20160101.
Marchiori, J. N. C., & da Silva Alves, F. (1995). O palmar de Coatepe. Ciência & Ambiente, v. 11, p. 93-102.
Miguel, L. A. (2009). Dinâmica e Diferenciação de Sistemas Agrários. Editora da UFRGS.
Milione, G. M., Scaramuzzino, R. L., Goyenetche, J. M., Lara, B. D., Mujica, C. R., Laddaga, J. E., & Ramos, E. J. (2024). Reemplazo de pastizales con forestaciones en el Sistema Serrano de Tandilia: efecto sobre la riqueza, composición y cobertura vegetal resultante. Quebracho, v. 32, n. 1, 2, p. 138-151.
Moonen, A. C., & Bàrberi, P. (2008). Functional biodiversity: an agroecosystem approach. Agriculture, Ecosystems and Environment, v. 125, p. 7-21.
Nair, R. (1993). An Introduction to Agroforestry. Kluwer Academic Publishers.
Nascimento, S. G. (2018). Sistema Silvipastoril na bacia hidrográfica do rio Santa Maria: uso do inhanduvá. Biodiversidade, v. 17, n. 3, p. 130-144.
Oksanen, J. (2022). Vegan: Community Ecology Package. R package version 2.6-4. Available from: https://github.com/vegandevs/vegan
Oliveira-Filho, A. T. (2015). Um sistema de classificação fisionômico-ecológico da vegetação neotropical: segunda aproximação. In: Eisenlohr, P. V., Felfili, J. M., Melo, M. M. R. F., Andrade, L. A. & Meira Neto, J. A. A. (Eds.). Fitossociologia no Brasil: métodos e estudos de casos. v. 2, Editora UFV, p. 452-473.
Oliveira-Filho, A. T., Budke, J. C., Jarenkow, J. A., Eisenlohr, P. V., & Neves, D. R. (2015). Delving into the variations in tree species composition and richness across South American subtropical Atlantic and Pampean forests. Journal of plant ecology, v. 8, n 3, p. 242-260.
Oliveira-Filho, A. T. (2017). NeoTropTree: Flora arbórea da Região Neotropical: Um banco de dados envolvendo biogeografia, diversidade e conservação. <http://www.neotroptree.info/
Oliveira Júnior, C. J. F., & Cabreira, P. P. (2012). Sistemas agroflorestais: potencial econômico da biodiversidade vegetal a partir do conhecimento tradicional ou local. Revista Verde de Agroecologia e Desenvolvimento Sustentável, v. 7, n. 1, p. 2012-224.
Olson, D. M., Dinerstein, E., Wikramanayake, E. D., Burgess, N. D., Powell, G. V., Underwood, E. C., ... & Kassem, K. R. (2001). Terrestrial Ecoregions of the World: A New Map of Life on Earth: A new global map of terrestrial ecoregions provides an innovative tool for conserving biodiversity. BioScience, v. 51, n 11, p. 933-938.
Ospina, A. A. (2003). Agroforestería: aportes conceptuales, metodológicos y prácticos para el estudio agroforestal. ACASOC. 209 p.
Overbeck, G. E., Boldrini, I. I., Carmo, M. D., Garcia, E. N., Moro, R. S., Pinto, C. E., & Zannin, A. (2015). Fisionomia dos campos. Os campos do Sul, v. 21, n. 12, p. 31-39.
Patt, G. S., & Ayan, H. F. (2005). Influencia del dosel de Aspidosperma quebracho-blanco schltdl. sobre el pastizal del Chaco Árido. Revista Científica Agropecuaria, v. 9, n. 2, p. 181-186.
Paz, E. A., & Bassagoda, M. J. (2002). Aspectos fitogeográficos y diversidad biológica de las formaciones boscosas del Uruguay. Ciência & Ambiente, v. 24(, n 1, p. 35-50.
Proyecto REDD+ Uruguay (2019). Metodología aplicada en la elaboración del mapa de cobertura de bosque nativo de Uruguay para el año 2016 con imágenes Landsat 8. Bernardi, L., Boccardo, A., Miguel. C., Olivera, J., Penengo, C. y Rama, G., Serafini, J., Kindgard, A. Ministerio de Ganadería, Agricultura y Pesca - Ministerio de Vivienda, Ordenamiento Territorial y Medio Ambiente. Montevideo.
R Core Team (2022). R: a language and environment for statistical computing. R Foundation for Statistical Computing. Available from: https://www.R-project.org/
Rebelo, J. F., & Sakamoto, D. G. (2021). Agricultura sintrópica segundo Ernst Götsch. Reviver. 156 p.
Reitz, R., Klein, R. M., & Reis, A. (1983). Projeto madeira do Rio Grande do Sul. Corag. 528 p.
Richardson, David M., Bingeeli, Pierre, and Schroth, Gotz. (2004). Invasive Agroforestry Trees: Problems and Solutions. In Agroforestry and Biodiveristy Conservation in Tropical Landscapes. Schroth, Gotz, da Fonseca, Gustavo A. B., Harvey, Celia A., Gascon, Claude, Vasconcelos, Heraldo L., and Izac, Anne-Marie N., editors. 371 p. Island Press.
Roberts, D. W. (2022). labdsv: Ordination and Multivariate Analysis for Ecology. R package version 2.0-1. https://CRAN.R-project.org/package=labdsv
Sell, J. C. (2017). Estradas paisagísticas: estratégia de promoção e conservação do patrimônio paisagístico do Pampa Brasil-Uruguai. [Tese de Doutorado, Universidade Federal de Santa Maria]. Repositório UFSM. https://repositorio.ufsm.br/handle/1/13265
Sobral, M., Jarenkow, J. A., Brack, P., Irgang, B., Larocca, J., & Rodrigues, R. S. (2013). Flora arbórea e arborescente do Rio Grande do Sul, Brasil. RiMa. 362 p.
Soriano, A. (1992). Río de la Plata grasslands. In: Coupland, R. T. (Ed.). Ecosystems of the world p. 367-407. Elsevier.
Stevens, G. C. (1989). The latitudinal gradient in geographical range: how so many species coexist in the tropics. The American Naturalist, v. 133, n. 2, p. 240–256.
Vargas, G. K., & Brack, P. (2021). A problemática ambiental na gestão do Bioma Mata Atlântica no Rio Grande do Sul. Bio Diverso, v. n. 1.
Vargas, G. K. D., Frangipani, M. A., Müller, S. C., & Jarenkow, J. A. (2022). Estrutura, diversidade e contingentes fitogeográficos do componente arbóreo de uma floresta ribeirinha no Planalto da Campanha, bioma Pampa. Iheringia. série botânica. v. 77, e2022015.
Waechter, J. L. (2002). Padrões geográficos na flora atual do Rio Grande do Sul. Ciência & Ambiente, v. 24, n 1, p. 93-108.
Walter, H. (1985). Vegetation of the earth and ecological systems of the geo-biosphere. Springer-Verlag.
Wood, P. J.; Burley, J. (1991). A tree for all reasons: the introduction and evaluation of multipurpose trees for agroforestry. International Centre for Research in Agroforestry.
World Wide Fund for Nature. (2012). Terrestrial Ecoregions of the World. https://www.worldwildlife.org/publications/terrestrial-ecoregions-of-the-world
Yáñez Díaz, M. I., Cantú Silva, I., & González Rodríguez, H. (2018). Efecto del cambio de uso de suelo en las propiedades químicas de un vertisol. Terra Latinoamericana, v. 36, n 4, p. 369-379.
Zhang, J., Cheng, K., Zang, R., & Ding, Y. (2014). Changes in floristic composition, community structure, and species diversity across a tropical coniferous-broadleaved forest ecotone. Tropical Conservation Science, v. 7, n. 1, p. 104–118.
Zomer, R., Trabucco, A., van Straaten, O., & Bossio, D. (2006). Carbon, land and water: hydrologic dimensions of climate change mitigation through afforestation and reforestation. International Water Management Institute.
Authorship Contributions
1 – Guilherme Krahl de Vargas
Master’s degree in Botany from the Federal University of Rio Grande do Sul
https://orcid.org/0000-0002-7028-469X • guilhermekvargas@gmail.com
Contribution: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing - original draft, Writing - review & editing
2 – Adriana Carla Dias Trevisan
Post-Doctorate in Agroforestry Systems from the Federal University of Santa Catarina
https://orcid.org/0000-0002-5192-6431 • adriana-trevisan@uergs.edu.br
Contribution: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing
3 – Beatriz Marcela Sosa Calleja
PhD in Biological Sciences from the Basic Sciences Development Program/UdelaR
https://orcid.org/0000-0002-9259-0516 • beatriz@fcien.edu.uy
Contribution: Funding acquisition, Methodology, Validation, Writing - review & editing
4 – Paulo Brack
PhD in Ecology and Natural Resources from the Federal University of São Carlos
https://orcid.org/0009-0005-2232-6903 • paulo.brack@ufrgs.br
Contribution: Validation, Writing - review & editing
How to quote this article
Vargas, G. K. de, Trevisan, A. C. D., Sosa Calleja, B. M., & Brack, P. (2025). Native tree and arborescent species for Agroforestry Systems in the Uruguayan-Brazilian Pampa. Ciência e Natura, Santa Maria, 47, e85058. DOI: https://doi.org/10.5902/2179460X85058. Available in: https://doi.org/10.5902/2179460X85058