**1. Introduction**

Wildfires a ffect agricultural and forest land covers more than any other land cover type [1]. Prescribed fires involve basically controlled burning of forests as a mean to reduce fire fuel levels. This practice minimises the extent, severity and danger of potential wildfires [2]. In Europe, prescribed fire use has disappeared from many countries due to the intensification of agriculture and socio-economical changes [3]. In most European countries prescribed burning is prohibited. Where permitted (e.g., in UK, Germany, France, Spain, Slovenia), there are strict controls on the timing and location of the burning extension. European fire managemen<sup>t</sup> policies are still aimed at fire suppression instead of using prescribed fires as a preventative measure. At present, only 10,000 ha year<sup>−</sup><sup>1</sup> of the Mediterranean basin is prescribed burnt; for comparison, this is about 3% of the wildfire extent in Spain, France and Portugal alone [4,5].

The use of prescribed fires in Europe is mostly an 'unnatural' event for both biotic and abiotic systems. Ecosystem alterations can arise as a consequence of fires, such as niche alteration for native species [6], invasion of alien species [7], and drastic changes to forest soil systems properties [2]. Climate change is an additional threat to wildfires frequency, due to projected higher mean temperatures, increase in the number of summer days and the decline of average precipitation. Climate change models sugges<sup>t</sup> that by 2100, temperatures in central Europe will increase up to about 3 ◦C [8]. This increase in temperatures will cause soils to become drier for longer periods and therefore would extend the wildfire season globally. Southern Europe has already experienced record summer temperatures in 2017. The Intergovernmental Panel on Climate Change's (IPCC) Fifth Report (AR5) on forest fire risk shows as a general trend that each region, below the 55th parallel north, will either increase in forest fire risk or stay at the same level [9]. Temperate forests are particularly sensitive to the e ffects of climate change due to their long lifespan preventing rapid adaptation. The large amount of C in forest soils means that any deviation of the level will have a significant e ffect on global C balance and climate change [10]. The future IPCC projections support that the e ffects of wildfires will be an increasing concern for both the scientific community and the governmental agencies.

A major factor influencing the e ffects of fires on soil properties is fire severity, a parameter primarily controlled by combustion and site factors. Combustion factors include quantity, moisture and type of fuel, whereas site factors include topography, wind direction, air temperature and humidity [2]. There are two key components of fire behaviour that influence fire severity: intensity and duration. Intensity is defined as the rate of thermal production in the fire, whereas duration is the length of time soil is exposed to fire. The most varied change to forest soils during burning is the loss of soil organic matter (OM) and plant available phosphorous and nitrogen in mineral soils. Fire severity plays a significant role in the e ffect on this important carbon pool. Substantial consumption of OM occurs between 200 and 250 ◦C and is complete at ca. 460 ◦C [11]. Fire does not maintain a consistent temperature in the soil. Some parts of the burn area might experience over 460 ◦C, whereas other areas may only experience less than 200 ◦C. This means that the OM may only experience minor volatilisation at low temperatures compared to complete oxidation at high temperatures. There is consensus about combustion causing reduction or total removal of the forest floor [12]. The reduction of OM is not uniform with depth: for example, some studies have found that concentrations of OM decreases significantly by the heating e ffect at 1 cm depth, but not at 2 and 3 cm depth in a pine forest ecosystem, North-East Spain [13].

Studies have found di fferences between prescribed burning and wildfire e ffects on soil 10 years following the fire event: i.e., lower carbon content following prescribed fire and higher soil C content following wildfire, due to the accumulation of charcoal and the encroachment of post-fire N-fixing vegetation [14]. Studies support these findings, describing, e.g., OM returning to normal after one year [13]. They also showed that fire has a fertilising e ffect, due to the dissolution of ashes and the mineralisation of charcoal enriching the soil.

The key nutrients within forest soils, mainly contained in its organic matter component (OM) are: nitrogen (N), available phosphorus (P), exchangeable potassium (K), exchangeable calcium (Ca) and exchangeable magnesium (Mg). Each nutrient reacts di fferently to fire, depending on its individual volatilisation threshold [15]. The combustion of nutrients bound in vegetation and soil organic matter add inorganic forms of K, Ca, Mg, P and N to the soil [16]. Each nutrient has its own response to burning, e.g., early studies have found that concentrations of K, Ca and Mg ions can increase, whereas N and S often decrease [17]. Fire intensity is directly linked to the temperature an object experiences during fire, which in turn impacts the type of nutrient and the amount volatilised. Fire acts as a rapid mineralising agen<sup>t</sup> [18]. This mineralisation has been studied extensively for N and P, because they are the two key macronutrients for plants. Soil nutrients that have low volatilisation thresholds are the first to be mineralised. There is an immediate reduction in soil organic N due to volatilisation [19]. Substantial proportions of soil organic N survive low intensity fires, however, moderate to high intensity fires convert most soil organic nitrogen to inorganic forms [2]. The effects of fire on properties of forest soils have been extensively reviewed and found that generally, losses of P through volatilisation are low [2]. However, the combustion of vegetation and litter cause major modifications on the P cycle. Micronutrients are also affected immediately after burning; e.g., total content and reducible forms of Mn increase significantly following fire due to the ash produced [20].

Soil pH has been also found to increase post fire. This increase has been attributed to the addition of base cations in ash, and to organic acid denaturing due to the heating of the soil [2]. However, significant increases only occur at temperatures >450 ◦C [21]; this temperature links with the total combustion of fuel as the ash produced has the capacity to neutralise soil acidity. Through analysing mixed forest soils, topsoil pH could increase as much as three units immediately after burning due to the production of K and Na oxides [22]. This significant increase can only be found in non-calcareous soils; calcareous soils are already alkaline, and in them the pH increases induced by fire are often negligible.

Currently, very little is known about the effect fires have on forest soils in Central Europe. Most research in Europe comes from Mediterranean forest studies, while there are few investigations about the areas that, according to the IPCC climate change scenarios, will have in the future climate similar to that of the Mediterranean region [9]. Soil in Mediterranean forests have low moisture and available nutrients, and plants are adapted by co-existing to prevent competition in resource scare environments [23]. Field research on fires, to our knowledge, is limited in Hungary in particular. Primary research is generally on the effects to biodiversity, mainly in grasslands and based on questionnaires to fire departments [24,25]. In particular, little research has been done on the effects of fire on Haplic Luvisols in Central Europe. Luvisols occupy ca. 5% of the total continental land area on Earth, and are found mainly in west-central Russia, the United States, Central Europe, the Mediterranean and southern Australia [26].

In purview of the above, the present study objectives were to: (1) characterise the effect of different fire intensities in a burn experiment on the organic matter, pH, potassium, phosphorus, calcium and magnesium of forest Haplic Luvisol through laboratory analysis; (2) examine the effectiveness of the experimental burning design at measuring changes in soil properties, and, (3) provide recommendations for fire authorities in Hungary about potential risks to the dominant soil Haplic Luvisol from both prescribed fires and wildfires. For this purpose, a field experiment of three fire treatments on forest Haplic Luvisol in Damak forest, NE Hungary was implemented.
