Aims: Across tropical regions, large forest areas have been converted to different agricultural land uses. These uses impose ecological disturbances affecting forest regeneration potential after field abandonment. Finding ways to identify those agricultural land uses limiting forest regeneration is a critical issue for conserving biodiversity in human-modified landscapes. Here, we developed a fast and inexpensive index, useful for quantifying ecological disturbance regimes associated with agricultural land uses, and tested its power to predict forest regeneration potential.Location: Municipality of Marqu es de Comillas, southeast Mexico.Methods: Interviews were conducted with local farmers to quantify disturbance components (size, duration and severity) associated with agricultural land uses. The scaled values of these disturbance components were added in a simple ecological disturbance index (EDI). In each one of nine recently abandoned fields representing a wide range of EDI values, two 10-m 2 plots, one close to and one far from nearby forest remnants, were established. On each plot, all woody plants of 10-100 m in height were counted, identified and measured in four 1m 2 subplots, at the time of field abandonment and 2 yr later. In addition, at each plot, 18 site condition (microclimate and soil) attributes were quantified at the time of abandonment. Plant density, biomass, species richness and species diversity were used as regeneration variables, and EDI and site condition attributes as independent ones.Results: Two years after abandonment, most regeneration variables declined exponentially with EDI. Biomass was not explained by EDI but changed positively with light availability. EDI was strongly correlated to vapour pressure deficit, which also predicted regeneration potential (except biomass).Conclusions: EDI is a cheap and easy tool for quantifying the ecological disturbance produced by a wide range of agricultural land uses. The index predicted several regeneration variables as well as or better than direct measurements of the site condition at the time of abandonment. EDI can be used to identify biodiversity-friendly agricultural land uses in human-modified landscapes.

In human-modified tropical landscapes (HMLs) the conservation of biodiversity, functions and services of forest ecosystems depends on persistence of old growth forest remnants, forest regeneration in abandoned agricultural fields, and restoration of degraded lands. Understanding the impacts of agricultural land uses (ALUs) on forest regeneration is critical for biodiversity conservation in HMLs. Here, we develop a conceptual framework that considers the availability of propagules and the environment prevailing after field abandonment as two major determinants of forest regeneration in HMLs. The framework proposes that regeneration potential decreases with size, duration and severity of agricultural disturbance, reducing propagule availability and creating ill-suited environmental conditions for regeneration. We used studies from Southern Mexico to assess this framework. First, we identify regeneration bottlenecks that trees face during transit from seed to follow-up life stages, using demographic analysis of dominant pioneer species in recently abandoned fields. Then, we explore effects of ALUs on forest regeneration at the field and landscape scales, addressing major legacies. Finally, we integrate agricultural disturbance with landscape composition to predict attributes of successful second growth forests in HMLs, and provide indicators useful to select tree native species for active restoration. An indicator of disturbance inflicted by ALUs, based on farmers' information, predicted better regeneration potential than measurements of soil and microclimate conditions at time of abandonment. Cover of cattle pastures in the landscape was a stronger indicator of forest regenerating attributes than cover of old growth forest remnants. To conclude, we offer recommendations to promote forest regeneration and biodiversity conservation in HMLs.Abstract in Spanish is available with online material.

No abstract

Questions What is the magnitude of the differences in environment and chronic human disturbance between contrasting slope aspects and topographic positions in a seasonally dry tropical forest? What is the effect of such topography‐related differences on composition, structure and diversity attributes of the tree community of this forest? Location Tziritzícuaro, Michoacán State, southern Mexico. Methods Vegetation was sampled in 36 100‐m2 plots evenly distributed among three topographic positions (lower, middle and upper parts of a slope) and two slope aspects (N‐ and S‐facing). Environment at these sites was described through modelling incoming solar radiation and in situ recording of temperature during 1 yr. Disturbance was visually assessed in the field to calculate a Chronic Disturbance Index. Vegetation structure and diversity were compared among the resulting combinations of slope and topographic position. PERMANOVA and CCA were used to examine the multivariate relationship among vegetation, topography and disturbance. Results Slope aspects and topographic positions differed in terms of annual mean temperature, potential energy income and evapotranspiration. Conversely, disturbance was not so clearly related to topography. Regarding vegetation structure, significant differences were only found for individual sizes and abundance; these values increased towards the upper portion of S‐facing slopes, but decreased with elevation in N‐facing slopes. Species diversity (S, Jacknife 1 and Fisherʼs α) was higher in S‐facing slopes and increased from lower to upper topographic positions. PERMANOVA showed that vegetation structure and diversity were influenced by topographic position (12.9%) and the interaction between soil moisture and chronic disturbance with slope aspect (8.2% and 8.3%, respectively). CCA showed that a modest proportion (21%) of variation in species composition is explained by the combination of environmental and disturbance variables. Conclusions Slope aspect and topographic position represent axes of environmental and disturbance differentiation. Although vegetation attributes respond to these ecological factors, they do not show homogeneous responses. Floristic composition is clearly linked to environmental heterogeneity, while structural attributes and α‐diversity appear to be more closely related to human disturbance and soil moisture, particularly on S‐facing slopes. Integrating environmental heterogeneity and human disturbance with topographic variability enhances our understanding of large variation in tree community attributes in seasonal dry tropical forests.

No abstract