Depth matters: effects of precipitation regime on soil microbial activity upon rewetting of a plant-soil system

Abstract

Changes in frequency and amplitude of rain events, that is, precipitation patterns, result in different water conditions with soil depth, and likely affect plant growth and shape plant and soil microbial activity. Here, we used ¹⁸O stable isotope probing (SIP) to investigate bacterial and fungal communities that actively grew or not upon rewetting, at three different depths in soil mesocosms previously subjected to frequent or infrequent watering for 12 weeks (equal total water input). Phylogenetic marker genes for bacteria and fungi were sequenced after rewetting, and plant-soil microbial coupling documented by plant ¹³C-CO₂ labeling. Soil depth, rather than precipitation pattern, was most influential in shaping microbial response to rewetting, and had differential effects on active and inactive bacterial and fungal communities. After rewetting, active bacterial communities were less rich, more even and phylogenetically related than the inactive, and reactivated throughout the soil profile. Active fungal communities after rewetting were less abundant and rich than the inactive. The coupling between plants and soil microbes decreased under infrequent watering in the top soil layer. We suggest that differences in fungal and bacterial abundance and relative activity could result in large effects on subsequent soil biogeochemical cycling.

Fig. 1

Dynamics of soil water content in the experimental treatments (infrequent and frequent water input, dotted and full curves, respectively) over the duration of the experiment. Lines and shaded polygons around them indicate mean ± standard error (n = 5)

Fig. 2

Root biomass after 12 weeks under contrasting precipitation pattern treatments (open bars: infrequent water input, closed bars: frequent water input), in three soil depth layers (top: 0–5 cm, middle: 10–15 cm, bottom: 30–35 cm). Bars indicate mean ± standard error (n = 5). Significance between treatments in each layer: * 0.05>p>0.01, ** 0.01>p>0.001, *** 0.001>p.

Fig. 3

Principal coordinates analysis (PCoA) of the UniFrac pairwise dissimilarity of the relative abundance of bacterial sequences based on 16S rRNA gene a and of the Bray–Curtis distance of the relative abundance of fungal sequences based on ITS rRNA region b. The active (red symbols) and the inactive (blue symbols) communities were determined in the infrequent (open symbols) and frequent (closed symbols) water input treatments, in three soil depth layers (top: 0–5 cm, triangles; middle: 10–15 cm, squares, bottom: 30–35 cm, circles)

Fig. 4

Bacterial a and fungal b observed species richness and evenness (determined by the Simpson reciprocal index) in the active (red symbols) and inactive (blue symbols) communities at three soil depths (top: 0–5 cm, triangles; middle: 10–15 cm, squares, bottom: 30–35 cm, circles). Bars indicate mean ± standard error

Fig. 5

Relative abundance of the OTUs that responded significantly to soil depth in the active a and inactive b bacterial communities (phylum assignment provided, class provided for Proteobacteria). Based on their relative abundance, the OTUs clustered by top and bottom groups

Fig. 6

Relative abundance of the OTUs that responded significantly to soil depth in the active a and inactive b fungal community (order provided). Based on their relative abundance, the OTUs clustered by top and bottom groups