Determining the potential impacts of fire and different land uses on splash erosion in the margins of drylands

Highlights

• Wildfires alter soils and remove vegetation, which can lead to soil erosion.

• We estimate the splash erosion during the first months after a wildfire.

• Results showed that the highest soil losses occurred in the burned scrublands.

• The most influential factor was the presence of bare soil.

• Vegetation ecosystem restoration is the key issue after a wildfire.

Abstract

This research aimed to estimate the splash erosion and its evolution during the first months in specific land uses after a forest fire. The study area was located in Congosto (North-West Spain), in the margins of Spanish drylands, after a wildfire occurred in May 2012, which burned 15.56 ha of scrubland and Pinus reforestation. Two different burned land uses were selected and compared to control areas: i) burned pine forest; and, scrublands. Rainfall intensity and the number, sizes and speed of raindrops were measured by an optical disdrometer and soil loss by funnels. Moreover, infiltration, soil moisture content, aggregate stability, water repellence, pH and organic matter were also measured. Results showed that the highest soil losses occurred in the burned areas, especially in the scrubland plots. The most influential factors were the presence of bare soil and the low vegetation recovery rate. Changes in soil properties did not significantly influence splash erosion, although an increase in the presence of smaller classes of aggregates could promote erosion in the scrubland. We conclude that the vegetation ecosystem restoration is the key issue to be considered after a wildfire, especially, in those types of land uses which are severely affected by the fire in the margins of drylands.

Introduction

Fires are one of the most aggressive and powerful agents of land degradation in forests because it removes the vegetation cover that protects the soil and enhances runoff activation and, subsequently, soil erosion (Alcañiz et al., 2020; Chen, 2006; Pausas et al., 2009). To date, several challenges must be further studied considering the relationship between soils, humans and fire (e.g. Bento-Gonçalves et al., 2012; Rodrigo-Comino et al., 2020; Santín and Doerr, 2016). It is stated by several authors that splash erosion is the first stage of soil erosion activation, producing the collapse of unstable aggregates (Jomaa et al., 2012; Sadeghi et al., 2017). This fact is due to the raindrop impact and consequent detachment of soil particles that directly affects, especially, sites with lack of vegetation (Kavian et al., 2019; Liu et al., 2015). Splash erosion is determined by the characteristics of the soil surface it falls on, and intensity (Angulo-Martínez et al., 2012; Dunne et al., 2010). It is also determinant the size of drops and the speed, which can be modified by vegetation interception during their fall (Belmonte Serrato and Romero Diaz, 1998; Šraj et al., 2008). This can be very variable depending on the weather type conditions influenced by the wind origin, speed and intensity, and affecting the final soil erosion results (Nadal-Romero et al., 2015; Rodrigo-Comino et al., 2019). When the individual raindrops overcome the kinetic energy threshold, they can detach and transport soil particles (Salles et al., 2000, 2002). This threshold will also depend on the size and weight of the soil particles and organo-mineralogical composition of the soil aggregates (Furbish et al., 2007).

In Spain, the scientific community is paying more and more attention to this issue because the arson has significantly risen at the central plateau of the Iberian Peninsula after 2000 (Vecín-Arias et al., 2016). Since the last century, they are focusing on the relation between wildfires, ecosystem recovery, soil restoration and water resources (Cerdà, 1998). Immediately after the fire, the absorbing properties of the ash layer can reduce the potential activation of runoff to a negligible rate, decreasing the soil water repellency (Cerdà and Doerr, 2008). However, recent investigations also affirm that the ash layer soon loses this biological ability and is drastically exposed to immediate raindrop impact (Oliveira-Filho et al., 2018). As a result, after an initial stage, forest fires trigger higher soil erosion rates due to the absence of vegetation as other land uses such as agricultural fields, increasing in water repellency and reducing the water storage capacity (Bodí et al., 2012; Úbeda et al., 2006). Using an experimental set up rather exhaustive would permit to bring additional quantitative information on the effect of fire on soil erosion processes, which is scarce. Understanding the splash effect would give new insights to develop more specific soil erosion control measures in burned lands.

Fires change some key soil properties such as aggregate stability (Fox et al., 2007) or soil water repellency (Keizer et al., 2008), which also will deeply affect the splash effect (Zavala et al., 2009). Also, fire may produce changes in much stable physical and chemical characteristics of the soil such as compromising the mean weight diameter of aggregates, the distribution of aggregates, pH, organic matter content and aggregate stability (Fernández et al., 2016), which highly increase the potential splash erosion (Saedi et al., 2016). But the changes related to splash erosion that the fire produces in the soil also depends on the fire intensity and severity of fire (Jordán et al., 2011, 2016), which depends on several factors such as the previous ecosystem quality. Jordán et al. (2010) stated that high soil moisture minimizes the intensity of the fire, and, conversely, a fire produces a decrease of moisture content. For instance, phenomena such as soil compaction, building terraces in steep slopes (Hammad et al., 2006), tillage or engineering constructions (Abrantes et al., 2018; Awal et al., 2019) may influence the mean weight diameter of the soil aggregates and also decrease the infiltration capacity (Freebairn et al., 1991; Moldenhauer and Kemper, 1969). This would generate a surface water layer that may protect the soil from direct raindrop impacts, decreasing, subsequently, the splash erosion. Also, the released material can compact the soil surface and fill pores creating an impermeable seal that may leave soil particles exposed and ready to be washed away (Di Prima et al., 2018; Morin and Van Winkel, 1996).

The changes produced by the heating and combustion during the fire also affects the plant recovery (Bodí et al., 2012). High-intensity fires produce high seed mortality in the soil seed bank and the roots system decreasing the regeneration capacity of seeder species (Trabaud, 1998). On the other hand, under moderate intensity fires, the capacity of regeneration is higher, not showing symptoms of damage even after repeatedly burned (Schaffhauser et al., 2012). It is reported that some land uses need about 25 years to naturally return to their previous state, after a moderate fire, for example in Pinus spp areas or scrublands (Tang et al., 2013). The capacity of regeneration depends on the species of plants and the frequency of the time elapsing between consecutive wildfires because fires can make the soil nutrient status poorer, producing a transition from mature ecosystems to scrublands (Keesstra et al., 2017). Therefore, the main goal of this research is to determine the potential impacts of fire and different land uses on splash erosion, comparing two types of ecosystems: scrubland and Pinus reforestation. We investigated: i) how splash erosion behaves relating to the rainfall characteristics during the first months after fire; ii) which type of land uses depending on their capacity of recovery registers a clearer reduction of soil erosion; and, iii) which soil physical and chemical changes are produced by the fire in the margins of drylands.