Microclimate and seed bank under Pithecellobium keyense (Fabaceae) in a coastal scrubland

This study was conducted on a sandy beach of Yucatán, Mexico. We collected soil samples and measured microclimatic variables under 30 adult individuals of the shrub Pithecellobium keyense at three different positions under the canopy of every studied individual: “Open” or completely sunny, “Half shaded”, placed at the border of the canopy and “Fully shaded” placed under the canopy, next to the stem of the shrub. Seeds in the soil samples were collected, counted and tested for viablity. The shade of Pithecellobium keyense buffers for extreme values of environmental variables such as noon temperature, PPFD (Photosynthetic Photon Flux Density) and air relative humidity, this effect is stronger in the dry season. During the dry season, the probability of seed viability decreased with increasing PPFD. During the rainy season, the probability of seed viability was larger in the Half shade position and lower in Full shade and in the Open (high light) positions. The seed bank was larger during the dry season than during the rainy season.

Methods
To characterize the microclimate associated with the shade under P. keyense, 20 out of the 30 original P. keyense shrubs were randomly chosen. For each of these shrubs, ambient light (Photosynthetic Photon Flux Density or PPFD), air temperature (noon and midnight), soil water potential (predawn and noon) and air relative humidity (noon and midnight), were measured in three different positions beneath the canopy of each one of the shrubs: (1) Open: an area with no shrub cover, less than 1 m of distance of the corresponding shrub; (2) Half shaded: placed at the outer boundary of the shrub canopy; and (3) Full shaded, placed next to the plant stems. the total ambient light was measured with a quantum sensor (LI-190S, LI-COR, Lincoln, Nebraska, USA) placed 10 cm above the ground; air temperature and air relative humidity were measured with data recorders (Extech, SD800). Those data were recorded every 2 s and data was stored as a 10-minute average on a memory card (SM192, Campbell Scientific) connected to a datalogger (CR1000, Campbell Scientific). Soil relative water content was determined gravimetrically before dawn and at noon.  All microenvironmental variables were measured during the rainy (September 2015) and the dry seasons (May 2016) during 48 consecutive hours in each season. To characterize soil seed bank composition associated with the 30 P. keyense shrubs, three soil samples were taken at the same three positions under the canopy used for microclimatic measurements (Open, Half shade and Full shade). To determine seed viability, seeds were tested with tetrazolium salt.

Two way ANOVAs were carried out for every variable taking into account season and position under the canopy as factors.