RESUMEN

Microbial communities respond to changes in environmental conditions; however, how compositional shifts affect ecosystem processes is still not well-understood and it is often assumed that different microbial communities will function equally under similar environmental conditions. We evaluated this assumption of functional redundancy using biological soil crusts (BSCs) from two arid ecosystems in Mexico with contrasting climate conditions (hot and cold deserts) following an experimental approach both in the field (reciprocal transplants) and in laboratory conditions (common garden), focusing on the community's composition and potential for nitrogen fixation. Potential of nitrogen fixation was assessed through the acetylene reduction assay. Community composition and diversity was determined with T-RFLPs of nifH gene, high throughput sequencing of 16S rRNA gene amplicons and metagenomic libraries. BSCs tended to show higher potential nitrogen fixation rates when experiencing temperatures more similar to their native environment. Moreover, changes in potential nitrogen fixation, taxonomic and functional community composition, and diversity often depended on an interactive effect of origin of the communities and the environment they experienced. We interpret our results as legacy effects that result from ecological specialization of the BSC communities to their native environment. Overall, we present evidence of nonfunctional redundancy of BSCs in terms of nitrogen fixation.

Asunto(s)

RESUMEN

BACKGROUND: In arid and semiarid shrublands, water availability directly influences ecosystem properties. However, few empirical tests have determined the association between particular soil and hydrology traits with biodiversity and ecosystem biomass at the local scale. METHODS: We tested if plant species richness (S) and aboveground biomass (AGB) were associated with soil and topographic properties on 36 plots (ca. 12.5 m2) in 17 hectares of chaparral in the Mediterranean-climate of Valle de Guadalupe, Baja California, México. We used close-to-the-ground aerial photography to quantify sky-view cover per species, including all growth forms. We derived an elevation model (5 cm) from other aerial imagery. We estimated six soil properties (soil water potential, organic matter content, water content, pH, total dissolved solids concentration, and texture) and four landscape metrics (slope, aspect, elevation, and topographic index) for the 36 plots. We quantified the biomass of stems, leaves, and reproductive structures, per species. RESULTS: 86% of AGB was in stems, while non-woody species represented 0.7% of AGB but comprised 38% of S (29 species). Aboveground biomass and species richness were unrelated across the landscape. S was correlated with aspect and elevation (R = 0.53, aspect P = 0.035, elevation P = 0.05), while AGB (0.006-9.17 Kg m-2) increased with soil water potential and clay content (R = 0.51, P = 0.02, and P = 0.04). Only three species (11% of total S) occupied 65% of the total plant cover, and the remaining 26 represented only 35%. Cover was negatively correlated with S (R = -0.38, P = 0.02). 75% of AGB was concentrated in 30% of the 36 plots, and 96% of AGB corresponded to only 20% of 29 species. DISCUSSION: At the scale of small plots in our studied Mediterranean-climate shrubland in Baja California, AGB was most affected by soil water storage. AGB and cover were dominated by a few species, and only cover was negatively related to S. S was comprised mostly by uncommon species and tended to increase as plant cover decreased.

Asunto(s)

RESUMEN

Vegetation greenness (normalized difference vegetation index, NDVI) showed significant temporal and spatial correlations with precipitation and topography-derived features within the context of slope aspect (South- (SFS) and North-facing slopes (NFS) and an intermountain valley (IMV)) in a semi-arid Mediterranean-climate watershed in northwestern Baja California, México. Rank correlation with annual precipitation (1986-2016) showed a strong positive relationship with wet season NDVI at SFS (Rs = 0.82), IMV (Rs = 0.79), and NFS (Rs = 0.65) but moderate relation and only on hillslopes in the dry season (SFS, Rs = 0.47; NFS, Rs = 0.39). Thus, the vegetation on the more xeric SFS sites was more responsive to intra-annual and inter-annual precipitation than on either IMV or NFS. The correlation of NDVI with six topography-derived environmental attributes (elevation, slope gradient, curvature, drainage density, topographic wetness index, solar radiation) was weak to moderate, varied in degree and significance between years with exceptionally high or low NDVI, and often differed in sign between SFS, NFS, and IMV. Results showed that precipitation controlled vegetation greenness, under the three aspect conditions, more closely than did the other topography-derived features, and the sparse deciduous vegetation of SFS showed stronger associations with precipitation than IMV or NFS. The measurement of these relationships should be continued and complemented by other studies to improve the overall model, because they are important to modeling ecohydrology and productivity, and may be of use for projecting and hindcasting vegetation dynamics.

Asunto(s)

RESUMEN

Soil respiration (Rs) has been usually measured during daylight hours using manual chambers. This approach assumes that measurements made during a typical time interval (e.g., 9 to 11 am) represent the mean daily value; locally, this may not always be correct and could result in systematic bias of daily and annual Rs budgets. We propose a simple method, based on the temporal stability concept, to determine the most appropriate time of the day for manual measurements to capture a representative mean daily Rs value. We introduce a correction factor to adjust for biases due to non-optimally timed sampling. This approach was tested in a semiarid shrubland using 24 hr campaigns using two treatments: trenched plots and plots with shrubs. In general, we found optimum times were at night and potential biases ranged from -29 to + 40% in relation to the 24 hr mean of Rs, especially in trenched plots. The degree of bias varied between treatments and seasons, having a greater influence during the wet season when efflux was high than during the dry season when efflux was low. This study proposes a framework for improving local Rs estimates that informs how to decrease temporal uncertainties in upscaling to the annual total.

RESUMEN

Remotely sensed imageries were used to analyze the response of desert vegetation to physiographic factors and accumulated precipitation in drier and wetter years within a region of >16,500 km(2) sampled with 5,000 random pixels of 30 m. Vegetation development was indexed by the annual maximum values for greenness (SAVI) and canopy water content (NDII). Precipitation was interpolated from the 0.25° grid of the Tropical Rainfall Measurement Mission satellite-based estimates, showing a regional average of ∼55 mm in the wetter year. The vegetation indices were only weakly related to total precipitation, often in a negative sense. Terrain factors that most often affected the vegetation indices, in multiple regression models, were Topographic Wetness Index, elevation, and slope gradient; these often had different signs for SAVI and for NDII. Models for NDII on intrusive igneous rocks gave better results than on extrusive igneous rocks. The strongest patterns in vegetation development were the contrast among Pacific coast, Cordillera, and Gulf coast subregions and the generally stronger results for NDII than SAVI.

Asunto(s)