Introduction
Plant biodiversity plays a strategic role in ecosystemic provision of goods and services for human health and well-being (21). Agriculture causes changes in these ecosystems involving both productive activities and consequent effects on the socioeconomic and biophysical environment. Agriculture requires ecosystem services (water, soil and genetic diversity) while generating food, fiber and fuel, soil fertility, habitats for wildlife, landscape and culture, among others (29). However, the dominant models of agricultural intensification affect natural resources by generating pollution, fragmentation and degradation, associated with loss of ecosystem services (33).
Fragmentation of natural environments consists of disintegrating a continuous landscape into smaller fractions, separated from one another by a cover matrix resulting from human activity (9). Replacement of natural ecosystems for agroecosystems, urbanization, and other infrastructures increases land vulnerability. Landscape fragmentation levels could be considered an indicator of soil degradation (28) generally related to inappropriate agricultural practices, overgrazing, unplanned urbanization and expansion of agricultura frontier.
In Argentina, 66% of the territory corresponds to arid, semiarid and dry sub-humid areas, where water and wind erosion affect two million hectares a year (16). In turn, about 47% of agricultural production comes from drylands under desertification processes (2). These ecosystems have low capacity for natural regeneration. Degradation associated with reduced biological and economic productivity is hard to reverse. Consequently, water and food supplies are affected, threatening investment and technical development, and causing poverty and rural desertion, feeding back the originating processes (27).
Mendoza plains, where the most important agricultural activities take place, belongs to the Monte Phytogeographic Province, an extensive vegetation unit that occupies great part of Argentina’s drylands (7). Given water deficit is the major factor limiting productivity, agriculture is intensive and irrigated. Productive activities follow an agro-industrial model in the large irrigated oases of Mendoza, already expanded over non-irrigated areas. In San Martín (east Mendoza), more than 2200 farms comprise 94000 ha of private producers, most of them landowners (14). In Montecaseros, a district of San Martín, viticulture occupies most agricultural land, where clearing has resulted in a mosaic of landscapes. Productive crops merge with abandoned crops (with virtually bare soils, evidence of degradation and low floristic diversity) and fragments of natural fields (with appropriate native flora coverage preventing erosion processes). Vegetation is an important component of landscape affected by geomorphology, soil, climate, and even land use history (35). Physiognomically, the area reflects environmental heterogeneity. Prosopis flexuosa DC. woods and other areas representative of the Monte biodiversity in Montecaseros provide fundamental ecosystem services, where native plants might constitute reservoirs for pest predators and crop pollinators.
At local level, no publications addressing fragmentation in agricultural environments or floristic-ecological studies are available. Related references in the region involve ecology, management and conservation of Prosopis woodland (36), flora and vegetation in protected areas (25), vegetation assessment of peripheral oasis lands (11), and vegetation coverage classification in San Martín (19).
Considering background information scarcity in San Martín and the importance of viticulture for economic and social development, this study aimed to characterize plant communities in natural and intervened environments, in a representative area of the fragmentation processes caused by agriculture in Montecaseros district.
Materials and methods
Study area
The area comprises 250 ha in Montecaseros district, General San Martín Department, Mendoza, Argentina (Figure 1), located on the eastern fluvial-eolian plain of Mendoza (1), concentrating the greatest agricultural activity of the province. Altitude ranges between 598 and 644 m a. s. l. Climate is desertic (17), with mean annual temperature of 15°C, mean annual rainfall reaching 200 mm under a spring-summer regime, and prevalence of SE winds, frequently warm and dry (föehn type or “Zonda” winds).
The vegetation corresponds to the Monte Phytogeographic Province (7), typical of the Northern District and Central Sub-district of the Monte Desert (26), recently considered within the Septentrional District according to the latest classification (3). In natural environments, evergreen shrub-steppe and scrubland are dominant, represented by species of the genera Larrea, Atriplex and Lycium, and small relics of Prosopis flexuosa and Geoffroea decorticans woodland (Gillies ex Hook. & Arn.) Burkart. Grasses develop on open sites with low tree and shrub coverage. Fixed and semi-fixed sand dunes are frequent, with bare-root growth and psammophilous plants. Crops have fragmented these natural vegetation areas, creating ideal conditions for the development of native and exotic weed species.
Montecaseros soils are poorly developed Entisols (23,32) with strong moisture déficit most of the year. Sandy textures prevail, with surface or subsurface sites with finer soil layers, suitable for cropping given groundwater availability. The entire area is affected by land degradation processes, accelerated by natural water and wind erosion events.
Methodology
Analysis of plant communities
Areas were selected by satellite imagery (Google Earth, 2012) and available maps, representing regions affected by fragmentation processes (Figure 1, page 48). Research activities were conducted between December 2011 and June 2012. Maps were made using free access QGIS software, version 3.22.6 (2022).
Plant communities were studied by the phytosociological method (5), based on criteria of floristically, physiognomically and ecologically homogeneous sites. Each species was assigned abundance-dominance and sociability values in every location (24). In total, 87 surveys were conducted, and the study area was delimited in the field according to community composition and physiognomy. All plots were 49 m2 except at crop sites (30 m2) and in P. flexuosa woodland, where surveys were done over a larger area (225 m2). Additionally, each survey included vegetation stratification, geographic and altitude coordinates (GPS) and visual assessment of slope decline, water of dynamic and wind processes, solar exposure and evidence of human activity. Data were first ordered and analyzed on a comparative table of surveys, and summarized in a synthetic comparative table (Table 1, page 50 and 51), grouping those species best adjusted to each environment.
BF: biological forms (SP: succulent phanerophytes, G: geophytes, H: hemicryptophytes, M: microphanerophytes, N: nanophanerophytes and T: therophytes). P: species degree of presence (V: presence>80% of surveys; IV: 60.1-80%; III: 40.1-60%; II: 20.1-40%; I: <20%). Cov.: global coverage value (sum of mean species covers divided by total surveys per community, multiplied by 100). Annex 2 details species with low values of presence and/or coverage in each community. * Possible hybrid between P. flexuosa and P. chilensis.
BF: formas biológicas (SP: fanerófitas suculentas, G: geófitas, H: hemicriptófitas, M: microfanerófitas, N: nanofanerófitas y T: terófitas). P: grados de presencia de cada especie (V: presencia >80% de los relevamientos; IV: 60.1-80%; III: 40.1-60%; II: 20.1-40%; I: <20%). Cov.: valor global de cobertura (suma de las coberturas medias de cada especie dividido el número total de los relevamientos de cada comunidad por 100). El Anexo 2 detalla las especies con valores escasos de presencia y/o cobertura en cada comunidad. * Posible híbrido entre P. flexuosa y P. chilensis.
Categories describing each group of species (24) were 1- Exclusive characteristic species (strongly linked to a community, practically not observed outside) or Preferential characteristic species (with a wider ecological range but finding an optimum in one community), 2- Accompanying species (wide ecological range, not related to a particular community, indifferent) and 3- Accidental species (intruders in a community). Floristic compositions of non-cleared sites (scrubland) were interpreted as reference to the original natural vegetation. Table 2 (page 52) synthesizes total mean coverage of each community, floristic richness and dominant processes.
Total mean coverage (M.C.) expresses mean percentage of soil covered with vegetation. Richness (R) represents mean number of species per survey. Superindices and sub-indices show maximum and minimum observed values, respectively.
La cobertura total media (M.C.) expresa el porcentaje promedio de suelo cubierto por vegetación. La riqueza (R) representa el número medio de especies por relevamiento. Superíndices y subíndices muestran los valores máximos y mínimos observados, respectivamente.
Life form was analyzed for each species according to perpetuation organ position (22) as succulent phanerophytes or cacti (SP, plants with thick juicy stems), geophytes (G, plants with underground survival organs), hemicryptophytes (H, herbs and perennial grasses with renewal buds at ground level), microphanerophytes (M, trees with renewal shoots from 2 to 8 m), nanophanerophytes (N, shrubs with renewal buds at less than 2 m height), and therophytes (T, annual plants). In each community, relative contributions of every biological form were estimated according to coverage and frequency, and expressed in histograms or biological spectra (Figure 2, page 52). Coverage spectrums were determined after mean coverage values per abundance-dominance interval of each species (24). A frequency spectrum considered the number of species of each biological form over total species in the community.
Taxonomic determinations
Representative samples of species found in the area were herborized and taxonomically identified based on reference specimens from the Mendoza Ruiz Leal Herbarium (MERL, Argentine Institute for Dryland Research), and monographic works and publications on Argentina’s regional floras. Regarding nomenclature, we consulted the Catalog of Vascular Plants of Argentina (38) and the database of Argentine Flora (n. d.), the International Plant Names Index (n. d.), Tropics organization (n. d.), and the International Code of Nomenclature for algae, fungi, and plants (2018). Recently, a disintegration of the genus Prosopis L. based on molecular phylogeny has been proposed, followed by a new nomenclatural combination for American species of the genera Neltuma and Strombocarpa (12). Nevertheless, we keep Burkart’s traditional classification of Prosopis (6) for nomenclatural stability, prevailing current use and disadvantageous changes (31, articles 14, 34 and 56), until acceptance of nomenclatural changes by the International Committee of Nomenclature.
Soil analysis
Soil samples supplemented environmental descriptions of different communities. After establishing 16 sites, three samples were removed from a pit dug in each site, achieving a maximum depth of 1.10 m, totalling 48 samples and including data from the intermediate layer (0.20 to 0.50 m), where most root activity is concentrated (Table 3, Annex 1). Texture, fertility (nitrogen, phosphorus, potassium and organic matter), salinity (electric conductivity and sodium adsorption ratio), acidity (pH), ion complex (sodium, calcium, magnesium, carbonate, bicarbonate, sulfate and chloride) and humidity were determined for each simple (4). Results were analyzed with descriptive statistics by InfoStat software version 2020 (8), graphed on star charts for every community, where each variable is represented by a radius whose length is the mean value of samples taken from each community (Figure 3).
Each graph shows mean values of N: nitrogen, P: phosphorus, K: potassium, OM: organic matter, Na: sodium, Ca: calcium, Mg: magnesium, EC: electric conductivity, SO4: sulfates, Cl: chlorides and pH: hydrogen potential.
En cada uno de los gráficos valores medios de N: nitrógeno, P: fósforo, K: potasio, OM: materia orgánica, Na: sodio, Ca: calcio, Mg: magnesio, EC: conductividad eléctrica, SO4: sulfatos, Cl: cloruros y pH: potencial hidrógeno.
Results and discussion
Six plant communities were identified as sand dunes, scrubland, G. decorticans woodland, P. flexuosa woodland, shrubland in recovery after crop abandonment, and crop (Photo 1; Table 1, page 50 and 51; Figure 1, page 48). Species with wider ecological range (accompanying species) were L. crinita, T. terrestris, P. flexuosa, G. decorticans, A. lampa and P. philippianum.
Poor soils in organic matter and nitrogen had good contents of phosphorus and potassium, and different levels of salinity. Most of the samples showed mildly alkaline and non-sodic characteristics (Table 3, Annex 1).
Scrubland of T. usillo and L. divaricata with M. ephedroides and P. urvilleanum on sand dunes
Sand dunes had predominant scrub vegetation and small isolated trees, distributed in three vegetation-height layers as trees dominated by P. flexuosa and G. decorticans (average height of 3 m), shrubs dominated by T. usillo and L. divaricata (1.5 m), and an herbaceous layer dominated by L. crinita and P. urvilleanum (0.5 m). Heterogeneous sun exposure of this natural topography (from 5 to 10 m in height) was evidenced in vegetation growth. Thus, P. urvilleanum forms compact colonies and is occasionally dominant in dry sunny áreas (fixing and stabilizing more exposed dune sectors), whereas Z. jamesonii grows mostly in shady places, reflecting the micro-environmental diversity of dunes and being one of the greatest floristic resources of the study area (Table 2, page 52).
Exclusive characteristic species of this community resulted P. urvilleanum, S. goldsackii, P. globosum, I. parodii and C. castilloniana, preferential characteristic species are T. usillo, M. ephedroides, L. divaricata, P. grandiflora, G. mendocina, X. americana and T. terrestris, and accompanying species are represented by A. lampa, L. chilense, M. aspera, B. spinosa, S. divaricata, G. bracteata, S. leucopila, B. barbata, S. arnottii and E. leucantha.
Species coverage and presence spectra reflected a community dominated by nanophanerophytes (distinctive plants in a sand dune physiognomy), therophytes and hemicryptophytes. Microphanerophytes, although not frequent, seemed to contribute considerable coverage in contrast to the numerous succulent phanerophytes, of insignificant relative coverage (Figure 2, page 52).
Sand dunes are generally not saline due to soil permeability. The low water content of soil surface layers hinders organic matter decomposition and nutrient fixing, resulting in low fertility. Nevertheless, higher nitrogen values were found on shady sides of the dune, closest to crops and coinciding with a higher density of P. flexuosa trees, harbouring atmospheric nitrogen-fixing bacteria in their roots (34). Dunes pH was moderately to strongly alkaline, consistent with higher carbonate and bicarbonate values.
Dense scrubland of L. tenuispinosum and A. argentina with P. flexuosa
This community, on sites with no evidence of clearing, is mainly composed of tall shrubs and trees, isolated or in small groups (Photo 2). It presented three height layers, tres dominated by P. flexuosa (average height of 3 m), shrubs dominated by L. tenuispinosum, L. cuneifolia and A. argentina (1.8 m), and an herbaceous layer dominated by L. crinita (0.5 m). These scrublands on saline soils occupy the greatest land expanse in the study area, fragmented by vineyards. Tall and dense scrubland is physiognomically predominant in the landscape, alternating with sparse shrubland, sandy deposit sites, ground surface wáter accumulation, or evidence of laminar runoff in natural drainage areas. This scrubland would be considered a reference of typical natural vegetation in the area, helping understand plant plot recovery after crop abandonment, since both communities show similar edaphic and floristic features.
Exclusive characteristic species are P. tetracantha, G. boliviana and M. ovata in Sandy soils, and A. spegazzinii in saline, finer-textured soils. All preferential characteristic species like A. argentina, C. genistoides, S. mirabilis and S. phleoides are typical of saline soils. Accompanying species are P. flexuosa, A. emarginata and T. usillo in sandy areas, and L. tenuispinosum, L. crinita, L. cuneifolia and S. divaricata in saline or floodable lands. A. hybridus and F. bidentis appear as accidental species after nearby crop influence.
Scrubland physiognomy was evidenced by dominance of nanophanerophytes, followed by microphanerophytes. Therophytes and hemicryptophytes species were abundant. All succulent phanerophyte species surveyed in this study are represented in this community. Scrubland soil varied depending on the area surveyed, with sandy to loamy textures, low fertility, on slightly to excessively saline sites. In the latter case, the highest soluble sodium salt content (chlorides and sulfates) indicated strongly sodic soils.
Closed dune-base woodland of G. decorticans with L. tenuispinosum and A. emarginata
Closed G. decorticans woodland with three height layers included trees dominated by G. decorticans with some P. flexuosa trees (heights from 1.5 to 5 m), shrubs dominated by L. tenuispinosum and A. emarginata (average height of 1 m), and an herbaceous layer with low coverage values, dominated by L. crinita (0.80 m) and A. incarnata (0.1 m). Geoffroea decorticans woodland, longwise developed, was found at the base of a sand dune, spreading by gemmiferous roots, facilitating colonization of the abandoned crop area in recovery process.
Solanum triflorum was found exclusive to this community, beneath P. flexuosa canopies. Preferential characteristic species are G. decorticans, L. tenuispinosum and O. sulphurea (tolerant to floodable, clay or saline soils); A. emarginata and T. brunonis (representative of P. flexuosa woodland), and A. incarnata and C. asperata (linked to sandy soils by proximity to the sand dune of G. decorticans woodland). Accompanying species observed included P. flexuosa, L. cuneifolia and L. chilense, notoriously affecting total coverage and presence, where 95% of the relative cover is contributed by nanophanerophytes and microphanerophytes, and 4% by hemicryptophytes and therophytes. This is the only community comprising all life forms, showing a more equitable distribution of presence values.
Soil is sandy-loam up to 0.70 m deep, and sandy at greater depths. Fertility is low and the sand dune base determines good water drainage, reflected in the low-salinity soil profile.
Woodland of P. flexuosa with A. emarginata and A. patens
Relics of P. flexuosa woodland, fragmented by crops or cleared areas, were surveyed on 12 plots of 225 m2 each showing open P. flexuosa groves (trees between 2.5 and 5 m tall, whose canopies rarely touch), or closed and semi-closed groves (trees up to 6.5 m tall, whose canopies frequently touch). Three height layers were distinguished with tres dominated by P. flexuosa, sometimes accompanied by G. decorticans and B. retama (average height of 3 m), shrubs dominated by L. tenuispinosum and/or A. emarginata (1 m), and an herbaceous layer dominated by L. crinita (0.80 m). We observed P. flexuosa trees with a single trunk and multiple stems and mean tree coverage ranging between 22.5% and 62.5% (value explained by the great number of saplings with main shoot diameter under 5 cm: 130/survey, on average).
A. patens, P. strombulifera, A. odorata, V. astringens and V. aroma are exclusive characteristic species in this community. P. flexuosa reaches the highest relative coverage in the area, as well as A. mendocina and S. aphylla, identified as preferential characteristic species. P. flexuosa canopies prevent light from reaching the ground, limiting development of succulent phanerophytes, geophytes and therophytes with low relative coverage values.
Soil samples were collected from the grove adjoining the crop (A4, figure 1), finding non-saline soils of sandy loam texture up to 0.20 m deep and sandy at greater depths. Nitrogen and organic matter contents were among the highest recorded at all sites in the study area and, just as phosphorus and potassium contents, reached higher values in surface layers, probably induced by humus beneath tree canopies (34).
Low and open steppe of L. crinita and P. flexuosa with L. cuneifolia and B. retama in abandoned crops
Open steppe (52% of mean total coverage) showed shrubs and grasses with young, isolated, small, multi-stemmed trees (Photo 3, page 57).
Three height layers were distinguished by trees dominated by P. flexuosa and B. retama (average height of 2 m), shrubs dominated by L. cuneifolia (1 m), and an herbaceous layer of L. crinita and P. philippianum (0.80 m). This community corresponds to cleared areas levelled in 1972 for crop establishment (grapes, apricots, alfalfa and vegetables), labored for as long as 15 years and abandoned in 1987 due to poor performance and yield. Agricultural tillage altered soil characteristics in this area during 25 years between abandonment and present surveys. The surrounding natural vegetation has recolonized this site with numerous species, typical of fine-textured, saline or floodable soils. Adjoining crops contribute to species surviving the edaphic limitations of these abandoned plots.
Flaveria haumanii and P. cuneato-ovata behave as exclusive characteristic species; L. crinita, P. philippianum, L. cuneifolia, B. retama, S. subulatus and D. californica as preferential characteristic species, depending on site propensity for salt and water accumulation, or surface runoff after rainfall (25). P. flexuosa, as an accompanying species, contributes important coverage and presence, and C. dactylon and P. glauca, as accidental species, denote close crop-area influence.
Hemicryptophytes (primarily L. crinita) establishment increases in sunny areas with almost non-competition, defining the vegetation physiognomy together with nanophanerophytes and microphanerophytes. Scarce succulent phanerophytes and geophytes were evidenced by these plant limitation to propagate from seeds or in agamic reproduction.
Soils resulted sandy loam, with nitrogen and organic matter contents similar to non-cleared sites. Salinity increased with depth. A gypsum layer detected in the soil profile was analytically confirmed by higher calcium sulfate concentrations. This pattern may be due to machinery effects during years of tillage, generating a deep compacted layer that favoured salt concentration in the soil profile after periodic irrigation was suspended.
In general terms, these results are consistent with other studies associating site use history, particularly productive plots abandonment with higher soil erodibility and changes in landscape physiognomy of the Monte native vegetation (35,37).
Vitis vinifera crop with M. albus and F. bidentis
A 31-ha vineyard with different grapevine cultivars, conducted both vertically and horizontally under drip irrigation, presented two height layers, one shrub layer dominated by V. vinifera (average height of 2 m) and a herbaceous layer dominated by M. albus, F. bidentis, S. verticillata and B. catharticus (0.5 m). The vineyard was surrounded by cleared areas with spontaneously recovered vegetation, and by other areas with undisturbed natural vegetation, a sand dune, and P. flexuosa groves.
Exclusive characteristic species with higher coverage values were V. vinifera, M. albus, B. catharticus, C. album, C. bonariensis, S. asper and T. officinale. Overall, they were widely distributed (normally adventitious or introduced). Despite great multiplying capacities, these species require more water than other drought-tolerant ones, such as H. glauca, P. glauca, C. dactylon, F. bidentis, A. hybridus, S. verticillata, A. officinalis and S. euacanthum, all preferential characteristic vineyard species, although capable of surviving in less favorable conditions. L. crinita (more abundant at sunlit crop sites as row ends, borders, plots with young vines, with less coverage or lower height) and T. terrestris behave as accompanying species, whereas P. flexuosa and G. decorticans are accidental, with small presence, coming from adjoining scrublands.
Dominant in biological spectra are nanophanerophytes (primarily grapevines, contributing nearly half the coverage despite low presence) and therophytes (generally small and abundant). Geophytes reach the highest coverage and frequency values, probably because tillage does not remove rhizomes, bulbils or other organs of asexual reproduction. Low microphanerophytes frequency is explained by spontaneous P. flexuosa and G. decorticans individuals.
Soils are sandy on the surface and intermediate layers and sandy -loam at depths greater than 0.5 m, with low nitrogen and organic matter contents, non to slightly saline.
Conclusions
Scrubland is the predominant physiognomy in the area, typical for the Monte Province, with species of the genera Larrea, Atriplex and Lycium. According to this study, the prevailing vegetation before crop establishment was halophilic or salt-tolerant. Species like A. lampa, A. spegazzinii, A. argentina, P. tetracantha, G. bracteata and S. divaricata are frequent and grow in almost all communities identified.
Prosopis flexuosa groves grow where trees can benefit from groundwater, bordering the steppe or in inter-dune areas. As woodland canopy density increases, lower-height vegetation layers tend to diminish and even disappear due to sunlight shortage, except for sciophilous species like A. emarginata. G. decorticans is abundant on sites with finer-textured soil, at the base of sand dunes or in depressed areas, growing in well-defined groves. Through powerful gemmiferous roots, this species is one main colonizer of abandoned crop fields.
In sand dunes and their surroundings, psammophilous vegetation is well represented by a great diversity of species, given the micro-environmental features of these formations. P. urvilleanum is especially frequent in sand dunes destabilized by anthropic action. Adventitious or introduced species, dependent on adequate water supply (M. albus, C. album, S. asper, T. officinale, etc.) coexist in combination with more rustic species able to develop under water and/or saline stress (F. bidentis, P. glauca, A. hybridus, H. glauca, S. verticillata, C. dactylon, etc.). Thus, a crop area is the source of propagules dispersal for these last species being introduced into neighbouring sites.
In shrubland under recovery, clearing, tillage and later crop abandonment emphasized edaphic constraints, analytically determined by signs of soil erosion, compaction or crusting. Spontaneous re-vegetation led to lower floristic richness compared to the adjacent natural areas taken as reference sites, after 25 years of abandonment, with strong presence of tolerant species or species indicating fine-textured soils, floodable areas, or with surface runoff and saline characteristics. Original shrubland has turned into a steppe of bushes (halophilic in part) and grasses with small young trees. Several species present in the surroundings (O. sulphurea, S. divaricata, S. mirabilis, C. genistoides and P. tetracantha, among others) had difficult establishments on the plot, despite being halophilic or salt-tolerant. This result supports the hypothesis that the main problem for plot recovery would not be saline content, but soil structure alteration by tillage and subsequent crop abandonment, reinforcing uncertainty about biodiversity restoration levels to be sought in degraded soils, and recovery time span. This study demonstrated that a survey of natural environments with proper assessment methodologies is fundamental before land clearing for agriculture. Determination of plant communities in fragmented areas, and valuation of species indicative of edaphic limitations restricting crop establishment and productivity, are essential in defining land uses.
Flora components and species distribution are adequate indicators for monitoring environmental processes in plant communities. Incorporating the private sector into environmental protection systems towards the establishment of isolated preserved sites is necessary but not sufficient in disjointed landscapes. This first analysis of natural áreas fragmented by agriculture in Montecaseros, shows that typical communities of the Monte subsist, demonstrating how diverse flora is preserved in privately owned lands. Changes in floristic composition of plant communities may have occurred in the study area during the time elapsed between this survey and the current situation. The results obtained constitute an important baseline for monitoring sampling sites and vegetation dynamics in fragmented areas of Montecaseros. Cultivable land is a vulnerable and scarce resource, particularly in drylands. A proper interpretation recognizes limiting territory factors while approaching systemic criteria highlighting complementarity in management between crops and native vegetation areas. Studies of vegetation dynamism in private lands will contribute to plan and implement sustainable land management, tending to recover and conserve biodiversity in degraded soils, considering all associated ecosystem services.