Highlights:

Introduction

The arid zones of northern Mexico represent 65 % of the national territory and are characterized by water scarcity and extreme temperatures (^{Jiménez et al., 2015}). However, in such areas, water bodies can be found that are fed, mainly, by subway runoff giving rise to oases (^{Cariño-Olvera & Castillo-Maldonado, 2017}), those humid environments create relict biota habitats contrasting with the vegetation of the surrounding dry areas (^{Tenza et al., 2015}).

A total of 184 oases have been identified in the Baja California peninsula, 171 of which are located in the state of Baja California Sur. Of these, 48 % are considered typical subcalifornian oases because they show a freshwater body with associated palm grove vegetation (^{Cariño-Olvera & Castillo-Maldonado, 2017}). Although these ecosystems represent less than 1 % of the state surface, the oases concentrate great biological diversity, host endemic and introduced species and are used by native desert fauna, therefore, they are considered unique environments in Mexico (^{Martínez-Ballesté & Ezcurra, 2018}).

In Baja California Sur, the San Ignacio oasis, with an area of 2.7 km², is one of the most extensive oases and is derived from the presence of a spring surrounded by mesophytic vegetation contrasting with the adjacent xerophytic vegetation. The oasis hosts a great variety of animal species that use it as a watering hole, has feeding and breeding sites, as well as resting stations for migratory species (Ruíz-Campos et al., 2014). However, this ecosystem has been modified by pests and diseases, forest fires, introduction of exotic species, and economic activities allowing subsistence for the inhabitants of the surrounding village (^{Cariño & Ortega-Santos, 2014}; ^{Velázquez-Miranda & Ruiz-Campos, 2011}). Although the natural resources of this humid zone have been widely studied under production and conservation approaches (^{Cariño-Olvera & Monteforte-Sánchez, 2018}; ^{Diaz & Troyo, 1997}), very little has been studied in the characterization of vegetation diversity to assess its current status and the changes that have occurred.

The indicators of specific richness, proportional abundance, dominance, homogeneity and tree structure provide important information about the conditions of species diversity in a region and can be used in the evaluation of spatial or temporal changes due to natural or induced causes (^{Mora-Donjuán et al., 2014}). Therefore, the purpose of this study was to compare richness and diversity in tree, shrub and herbaceous strata, and to estimate the importance value of the species in the San Ignacio oasis. This study represents a baseline of indicators that can be used as a reference in the evaluation of changes generated by natural or anthropogenic events within the area of influence of the San Ignacio oasis.

Materials and Methods

Study area

The study area is located inside the San Ignacio oasis, municipality of Mulegé, Baja California Sur; it is part of the San Ignacio-La Laguna region, with geographic coordinates of 27° 17´ 55’’ N and 11° 52´ 30” W, at 135 m elevation (Figure 1). This surface covers an area of 1 237 ha of which 672.8 are under management and are part of the area of influence of the most important water body for the community of San Ignacio.

Figure 1 Location of the San Ignacio oasis, municipality of Mulegé, Baja California Sur. 

According to the Comisión Nacional del Agua (^{CONAGUA, 2015}), the oasis is located in the Lake San Ignacio-Arroyo San Raymundo watershed in Hydrological Region No. 2 "West Central Baja California (Vizcaino)", the characteristic climates are BWhw(x`) and BW(h`)hs(x) (^{Garcia, 2004}). The predominant soils are Fluvisol, Leptosol and Regosol (^{Instituto Nacional de Estadística y Geografía [INEGI], 2013}). Four types of land use and vegetation are located in the oasis area: a) riparian vegetation b) annual irrigated agriculture, c) sarcocaulescent scrub and d) urban sector.

Results

A total of 53 species distributed in 17 orders, 25 families and 47 genera were recorded in the study area. Based on the physical-biological characteristics, 11 herbaceous, 21 shrub and 21 tree species were identified. The Cactaceae family, represented by eight species, was the most abundant, followed by the Fabaceae family with seven species, Asteraceae with six species and Poaceae with five species (Table 1).

A total of 13 of the 21 succulent and woody species, typical of the upper stratum, are present in both the middle and lower stratum, which guarantees their survival into the habitat. However, some species are only present in the lower or middle stratum. Species in the upper stratum that have presence in the lower (Mimosa xantii A. Gray) or middle stratum (Parkinsonia microphylla Torr. and Stenocereus thurberi [Engelm.] Buxb.) do not seem to have survival problems; in contrast, those that do not have presence in either of these two strata (Mammillaria dioica [R. T. Craig bis] D. R. R. Hunt, Yucca valida Brandegee, Lophocereus schottii [Engelm.] Britton & Rose, Lysiloma candidum T. S. Brandegee and Olneya tesota A. Gray) seem to have regeneration or reproductive problems (Table 1).

Also, from the 21 species typical of the shrub stratum, 16 are found in the lower stratum and five of them are not (Phragmites australis [Cav.] Steud, Atriplex canescens (Pursh) Nutt., Callaeum macropterum [Moc. & Sesee ex DC] D. M. Johnson, Baccharis salicifolia [Ruíz & Pavón] Pers. and Cardiospermum corindum L.), suggesting that these also have problems with regeneration and establishment, limiting their continuity.

A total of 41 species were recorded in the lower stratum (of which 11 are herbaceous), with the families Cactaceae and Asteraceae being the most represented with five species each. A total of 36 species belonging to 33 genera were recorded in the middle stratum. As in the lower stratum, the Cactaceae family was the most represented with six species. Also, from the 21 species of the high stratum reported in the floristic inventory, seven belong to the Cactaceae family and six to the Fabaceae, which were the most represented families.

According to the species listed in Table 1, barrel cactus (Ferocactus peninsulae [F. A. C. Weber] Britton & Rose), ragweed (Ambrosia bryantii [Curran] Payne), Mexican fan palm (Washingtonia robusta Wendl.) and dagger cholla (Corynopuntia invicta Brandegee F. M. Kunth) are considered endemic, while spiny gourd (M. dioica), senita cactus (L. schottii) and desert ironwood (O. tesota) re listed in NOM-059-SEMARNAT-2010 as subject to special protection. Giant cane (Arundo donax L.), augustine grass (Stenotaphrum secundatum [Walter] Kuntze), tamarisk (Tamarix ramosissima Ledeb.), ice plant (Mesembryanthemum crystallinum L.), crowfoot grass (Dactyloctenium aegyptium [L.] Willd.), buffelgrass (Pennisetum ciliare [L.] Link) and palma datilera (Phoenix dactylifera L.) are introduced species; among them, the first four are considered aggressive invaders.

Rarefaction analysis indicates that species richness, diversity and dominance, evaluated with the Shannon-Wiener (H') and Simpson (λ) indices, are different between strata, while homogeneity in number of individuals (J') per species of the lower stratum is different (P = 0.05) compared to the middle and upper strata. The number and diversity of species are higher in the lower stratum, followed by the middle and upper stratum (Table 2).

On the other hand, Table 3 shows the diversity indices according to plant life form. The species richness of shrub plants (S = 21) is equal to that of tree type (S = 21), but lower than the species richness of herbaceous type (S = 11), while species diversity, estimated by the Shannon-Wiener index, presents differences (P = 0.05) among the three types of vegetation. Pielou indices indicate that the homogeneity in the number of individuals of herbaceous and shrub species is significantly equal, but greater than in tree species.

Table 4 shows that the lower stratum is populated by both herbaceous and shrub and tree species in regenerative state. The sum of the importance values of the species Prosopis articulata S. Watson, Cylindropuntia alcahes (F. A. B. Weber) F. M. Kunth, Jatropha cinerea (Ortega) Müll. Arg., Jatropha cuneata Wiggins & Rollins and W. robusta indicate that the lower stratum is dominated by species with tree-like biological growth.

At the middle stratum, shrub species and the regrowth of tree species C. alcahes, P. articulata, J. cinerea, W. robusta, Larrea tridentata (Moç. & Seseé ex DC.) Coville, J. cuneata, Stenocereus gummosus (Engelm.) Gibson & Horak and Fouquieria diguetii (Tiegh.) I. M. Johnst. account for more than two thirds of the total ecological importance value (Table 5).

The species with the highest IVI in the tree layer was P. dactylifera, an induced exotic species, followed by W. robusta, P. articulata and Pachycereus pringlei (S. Watson) Britton & Rose. The rest of the species showed IVI lower than 10 (Table 6).

Discussion

The diversity of flora in the oases of Baja California Sur is high in relation to that recorded in the state, despite the small area they occupy. After evaluating eight oases, ^{Arriaga et al. (1997}) reported 146 species related to wetland vegetation, of which 53 are found in the San Ignacio oasis. Furthermore, the number of species present in this oasis, although not all shared, represent about 50 % of the species recorded in the San José del Cabo oasis (109 species), which is located to the south of the peninsula (^{León-de la Luz et al., 2004}).

The Shannon-Weiner diversity indices estimated in the low (H'= 3.209), middle (H'= 2.598) and high (H'= 2.097) strata indicate higher species diversity than those estimated in other arid and semiarid areas of Mexico. For example, ^{Solís-Garza et al. (2017}), when studying the structure and tree and shrub riparian vegetation in arid areas of the Bacanuchi and Sonora rivers, estimated diversity indices H'= 1.61 and H'= 0.76 for tree and shrub vegetation, respectively. ^{Mora-Donjuán et al. (2014}) estimated H' = 1.87 for the desert scrub of northwestern Nuevo León, while ^{Gelviz-Gelvez and Pavón-Hernández (2013}) reported that true species diversity in a semiarid zone in the state of Hidalgo ranged between 7.8 and 1.6, equivalent values of H' = 2.054 and H' = 0.47, respectively.

It is reasonable to infer that the number of species is an important factor that explains the difference in diversity estimators; however, the level of homogeneity in the number of individuals per species is also a determining factor in diversity estimators (^{Sanches-Melo, 2008}). In this study, the superiority of the Shannon-Weiner diversity index of the low stratum in relation to that of the middle and high strata is attributed to the lack of dominance of some species (1- λ = 0.941) and, therefore, to a greater homogeneity in the number of individuals per species (J´= 0.864). Regarding the middle and high strata, Simpson's index estimators (1- λ = 0.873 and 0.858, respectively) suggest that there is no dominance of some species; however, Pielou's indices (J'= 0.719 and 0.618, respectively) indicate heterogeneity in the number of individuals per species (^{Medrano-Meraz et al., 2017}).

The difference in species richness between strata is attributed to the fact that the lower stratum includes shrub and tree species and the middle stratum includes tree species that have not exceeded the height growth that characterizes each stratum. It is important to mention that the four endemic species registered in the oasis are present in the lower stratum, with W. robusta as the one with the highest IVI, followed by A. bryantii, C. invicta and F. peninsulae; in addition, it is emphasized that all the induced exotic species of high risk for the ecosystem are found in this stratum. According to ^{Campos and Herrera (2009}), S. secundatum, whose relative frequency (Fr) was 5.76, is an herbaceous species with a high probability of expansion that tends to have a negative effect on native species by competing strongly for water, light, nutrients and space, as well as producing allelopathic substances that inhibit the growth and development of grasses. On the other hand, M. crystallinum, a succulent perennial shrub with Fr = 2.88, has the ability to invade coastal areas and strongly affect native species; its high capacity to absorb soil moisture, accumulation of salt from the roots to the sprouts and nitrate below them generate adverse conditions for the establishment and survival of native plants (^{CONABIO, 2016a}, ²⁰¹⁷). A. donax (Fr = 1.44) is a riparian herbaceous species, ranked among the most aggressive and difficult to eradicate. This species produces large patches of roots that alter hydrology and nutrient cycling; in addition, it reproduces mainly vegetatively via rhizomes and requires abundant light and water during the photosynthesis process (^{CONABIO, 2015}). Meanwhile, T. ramosissima (Fr = 0.72) is considered among the 100 most harmful invasive exotic species, as it increases soil salinity, alters the hydrological regime and affects the proliferation of fires, harming ecosystem management and diversity conservation (^{CONABIO, 2016b}).

Considering the relative frequency criterion, none of the species of the middle stratum has a wide distribution range. In this case, the native species C. alcahes, P. articulata and J. cinerea had the widest distribution range (Fr = 9.40 %, 6.27 % and 5.28 %, respectively). The rest of the species, with Fr between 0.006 % and 1.8 %, are confined to restricted areas. This is the case of the exotic species T. ramosissima and P. dactylifera; both were recorded in 0.41 % of the sampled area.

Regarding species sustainability, ^{Norden (2014}) mentions that regeneration is the most important process in the life cycle of plants, because the continuity of species depends on the ability of plants to regenerate or reproduce. Environmental conditions and germplasm (seeds or reproductive material) limit the regeneration and development of species. ^{Granados-Sánchez et al. (2011}) mention that the seasonality of precipitation affects the establishment and development of vegetation in arid zones, and explain that, while some species flower and bear fruit before the onset of rainfall, others survive by taking advantage of the water stored in the subsoil, so that precipitation and temperature are important environmental factors in the production and germination process of seeds.

Conclusions

Species richness and diversity was higher in the herbaceous stratum, due to the inclusion of shrub and tree species in this stratum. The importance value indicates that each of the species found in the middle stratum have a small distribution range. The endemic species Washingtonia robusta, Ambrosia bryantii, Corynopuntia invicta and Ferocactus peninsulae and the exotic species Stenotaphrum secundatum, Masembryanthemum crystallinum, Arundo donax and Tamarix ramosissima are found in the lower stratum. The indicators of diversity and structure reported in this study could be useful for the proposal of management alternatives that promote the sustainability of the oasis; in addition, this information establishes a baseline for the evaluation of possible changes in the vegetation composition of the oasis.