INTRODUCTION

Biological soil crusts are a community of organisms living in the soil surface composed mainly of cyanobacteria, lichens, and/or mosses, which have key roles for the health and functionality of the ecosystems, improving soil structure and stability, influencing the local hydrologic cycles, enhancing soil fertility, or increasing the biodiversity of soil microbial community (^{Belnap and Lange, 2013}). After a wildfire, mosses are described as fast colonizers before the establishment of vascular vegetation (^{Esposito, 1999}), suggesting a positive role in early post-fire stages. In the absence of vegetation after a wildfire, the presence of a moss biocrust stabilizes the soil surface protecting it effectively against water erosion (^{Silva et al., 2019}). After wildfires in semi-arid conditions, soils covered by mosses reach higher soil fertility, boosting the ecosystem recovery (^{García-Carmona et al., 2020}). However, apart from erosion control, little is known about the role for soil recovery of the early emergence of mosses, even less in climates under oceanic influence as in Portugal.

Post-fire management is an important factor that will determine the capacity of soil to recover from degradation. Salvage logging, the most common practice in Portugal, under certain circumstances can trigger erosion processes (^{Malvar et al., 2017}), alter nutrient cycling (^{Pereg et al., 2018}), and induce detrimental consequences for microbial diversity (^{García-Carmona et al., 2021}). On the other hand, effective mitigation measures such as mulching reduce the negative impacts of post-fire erosion providing a surface cover to soils before the vegetation regrowth (^{Girona-García et al., 2021}). Although the treatments’ effectiveness at mitigating soil erosion is widely documented, their medium-term implications on soil physico-chemical properties recovery is partially unknown.

In this work, we studied soils under two contrasting forest managements, salvage logging and mulch application, six years after a wildfire in eucalypt plantations. Our aim was to determine the medium-term effects of fire and forest management on soil physico-chemical properties in combination with the presence of burgeoning moss biocrust patches that emerged after the fire.

MATERIAL AND METHODS

The study area is located in Miranda do Corvo, central Portugal, affected by a moderate severity wildfire during August 2015 that burned 715 ha mainly of Eucalyptus globulus Labill. plantations. The climate is Mediterranean with oceanic influence characterized by mild winters and warm, dry summers, with mean annual temperatures of 12ºC and 851 mm of precipitation. Soils were characterized as an association of Epileptic Umbrisol and Cambic Umbrisol with a depth of 30 cm and loam texture. Six years after the wildfire, a developed moss biocrust covered the topsoil at several patches in the fire-affected area.

After the wildfire, two contrasting post-fire managements were applied in separate areas. One was salvage logging, without slash treatment or removal, for which a forwarder tractor made two passes, causing skid trails. The other was mulching, performed by homogeneously applying chopped eucalypt residues from the logging operations. Mulch was applied at a “standard” rate of 8.0 Mg ha^-1, effective in earlier field studies in the region (^{Keizer et al., 2018}), and a “low rate” of 2.6 Mg ha^-1, effective under laboratory conditions of simulated rainfall (^{Prats et al., 2017}).

Soil sampling was conducted in July 2021. For each treatment, 10 samples were collected from 0-2.5 cm depth, five replicates in soils underneath moss biocrusts and five in uncrusted soils (n=50).

Soil pH and electrical conductivity were measured in 1:2.5 and 1:5 (w/v) aqueous extracts. Soil organic carbon was determined by the potassium dichromate oxidation following the Walkley-Black method; nitrogen was analyzed following the Kjeldahl method; and available phosphorus was extracted and measured following the Olsen method. The aggregate stability and the total content of macro-aggregates were examined after an artificial rainfall of known energy (279 J min⁻¹ m⁻¹) (^{Roldán et al., 1994}). Soil water repellency was assessed by the Water Drop Penetration Time (WDPT) test.

RESULTS AND DISCUSSION

Six years after the wildfire, contrasting effects were found in relation to post-fire management and the presence of mosses in soils. Soils affected by salvage logging did not reflect negative consequences due to the management; indeed, an increment in available P was registered. Those results suggest the way the management was performed had no long-lasting effects on soil properties, results that are in accordance with Fernández and Vega, who did not find detrimental effects after logging under similar climate conditions and organic carbon content. On the other hand, the lower organic carbon and nitrogen content in soils underneath mosses show the opposite trend than in drylands, where an improvement in soil fertility and nutrient cycling is expected with the presence of a moss-biocrust (^{Ferrenberg et al., 2022}), even in post-fire environments, creating fertility islands (^{Muñoz-Rojas et al., 2021}). A possible hypothesis may be mosses preferred to avoid establishing where the most recalcitrant organic matter concentrates after the wildfire, which would be in accordance with the higher soil hydrophobicity registered in the uncrusted soils (^{Doerr et al., 2009}). The rapid recovery of vascular vegetation in the area together with the naturally high levels of organic matter apparently counteracts the effects of a moss biocrust on soil fertility in a medium-term. Nevertheless, the presence of mosses preserved soils, with a higher total content of macro-aggregates in all cases, providing better conditions to soils compared to the uncrusted ones, more exposed to degradation (^{Chamizo et al., 2012}).

Mulching increased nutrients and soil organic carbon contents, although the increment was detected only in soils under the highest rate of mulch and without moss development. The direct contact of the mulch residues with the topsoil may have accelerated its decomposition, thus the increment in soil fertility, while mosses could be slowing down the decomposition process (Figure 1, Table 1).

Figure 1 Soil organic carbon, nitrogen and available phosphorous measured in the two contrasting forest managements, salvage logging (untreated and logging) and mulch application (untreated, low rate and standard rate), in soils under mosses and uncrusted soils. 

CONCLUSIONS

The studied post-fire forest managements showed contrasting effects on soil properties in the medium-term. Whereas salvage logging did not negatively affect soils, the mulching at a standard rate increased soil fertility six years after the fire. The moss biocrust emerged after the wildfire preserved soil structure, thus decreasing the risk of soil erosion.