Comparison on effect of mowing and spring burning on the occurrence of grassland plants in Western Siberia (preliminary results)

. This article provides data on changes in the occurrence of plant species in grassland communities after the spring fire and the subsequent cessation of mowing in comparison with regularly mowed communities. It has been established that burning more often leads to a decrease in the occurrence of the species than to an increase. At the same time, for most species values are restored within 1 to 3 growing seasons


Introduction
Secondary grassland communities that form in the boreal forest zone in place of cleared forests are considered as the initial stage of succession that lead to forest reproduction.Therefore, their long-term existence is usually possible only under regular anthropogenic impact that prevent regrowth of trees and shrubs.The most common anthropogenic factors in this situation are grazing, mowing, cutting of undergrowth and periodic fires.
Due to the high agriculture value of grassland communities, applied management regime includes periodically repeated actions, mainly mowing and/or prescribed burning.Many researches has been devoted to the impact of these factors on the biodiversity of grassland communities; for example, for temperate grasslands many experimental studies were carried out in Scandinavia [1][2][3].As a result, it was found that mowing is a more preferable management option because of increased species richness of plants, while annual fires significantly reduce species richness and lead to specific successional pathways.However, prescribed burning is widely used to control grassland communities in many countries because treeless communities are considered to be well adapted to fire [4].At the same time, to achieve the greatest biodiversity with the traditional two-year burning period, it is necessary to preserve patches in the cultivated area with longer intervals between burning [5].
Despite the high frequency of fires in the Siberian part of Russia, the attention of researchers is primarily focused on forest fires, and the pyrogenic dynamics of meadow communities in this region requires further studies.As part of our research, in addition to the current article focused on the occurrence of individual species in grassland sites under pyrogenic impact, we also planned to publish the results of a study of the pyrogenic dynamics of the surface soil layer [6] and the pyrogenic dynamics of plant communities in general (to be published).

Materials and methods
The basis for this study was data collected in sites of controlled spring burning of different ages in the south-eastern part of Tomsk city on the territory of the Basic Experimental Complex of the V.E.Zuev Institute of Atmospheric Optics SB RAS (hereinafter referred to as BEC).The study area is located on the upper terrace of the right bank of the Ushaika River (right tributary of the Tom river, Ob basin) with absolute elevations of 152-160 m and dominant terrain slopes of 0.3-3.0°.The climate of the territory according to the Köppen-Geiger climate classification is Dbf (snow fully humid with warm summer) [7].The parent rocks are loess-like loams, the soils are chernozems [6].The territory of the BEC is subjected to mowing 1-2 times a year with periodic cutting of undergrowth in certain parts, and its vegetation cover can be characterized as polydominant forb-grass meadows with a predominance of Poa pratensis, Festuca pratensis, Dactylis glomerata, Trifolium pratense , Taraxacum officinale, Galium mollugo, etc. in various areas.
Controlled fires were carried out in areas of various sizes in late April-early May (after the snow melted).Each area was subjected to fire once and mowing was excluded after burning until the end of the experiment.In plots burned from 2019 to 2022, data on the species composition of vascular plants and species participation (as a percentage of the plot area) were collected in 1x1 m plots during three field seasons (July 2021, 2022 and 2023).Similar information was collected for background areas of the BEC that were not affected by pyrogenic effects.
The obtained geobotanical data (172 plots with a fire age from 1 to 5 years and 76 unburnt plots) in this study was analyzed according to the criterion of plant species occurrence.Occurrence values are defined as the proportion of sites where a species is found out of all sites in this group, expressed in percents.

Results
The calculation results are shown in Table 1.The effect of a fire can be most clearly seen in the first three years after the fire, since then the species composition and structure of the community begins to be significantly influenced by the factor of undergrowth developing in the absence of mowing.In addition, it should be taken into account that the occurrence values can be influenced by other factors (phytocyclicity of species, differences in environmental microconditions, etc.), which are only partially compensated by the data sampling in our experiment.
Of the 60 species of vascular plants recorded on the studied sites, it is possible to make an assumption about the positive or negative impact of fire only for 42 species, since the rest are too scattered or do not show pronounced trends.For 20 species, a lower occurrence was noted after burning, while a higher occurrence under the same conditions was noted for only 16 species.0.0 0.0 0.0 0.0 0.0 5.0 --Note.The "Effect of fire" column shows the characteristics of the change in occurrence: -"negative" -after burning, the occurrence of the species decreased compared to the background communities; -"positive" -after burning, the occurrence of the species increased compared to the background communities; -"neutral" -after burning, the occurrence of the species approximately corresponds to the background communities; -"-" -it is impossible to determine the effect of burning (the status is not obvious, or data on occurrence are not enough).For the positive and negative effect of fire, the column "Time of restoration, years" indicates how many growing seasons the corresponding trend is observed.

Discussion
Our assumptions about the positive or negative impact of fire are consistent with existing data in the literature about a slight decrease in biodiversity as a result of grassland burning E3S Web of Conferences 463, 02017 (2023) EESTE2023 https://doi.org/10.1051/e3sconf/202346302017 in comparison with mowing and statements about haymaking as the better management regime for grasslands [1][2][3].
A significant increase in occurrence after the fire was noted for tall short-rhizomatous and loosely-tufted grasses (Dactylis glomerata, Festuca pratensis, Phleum pratense), some meadow-margin herbs (Veronica chamaedrys, Equisetum sylvaticum, Taraxacum officinale), and the trend is most pronounced for Linaria vulgaris.In turn, a decrease in the species occurrence after a fire is typical for the tree undergrowth (Betula pendula, Pinus sylvestris, Populus tremula, Salix caprea), some meadow species (Achillea millefolium, Trifolium pratense, Vicia cracca, Galium mollugo, Hypericum perforatum), etc.Moreover, for the majority of species experiencing a positive impact of fire on occurrence, the effect is observed only during one season (median value for the array -1, average -1.8), while the negative impact of fire on average is observed longer (median value -3, average -2.35).It can be assumed that the positive effect of the fire is associated with a short-term decrease in competition for environmental resources (due to a decrease in total projective cover observed in the first growing season after fire, and the entry of mineral nutrition elements into the soil from the ash, which also has a noticeable effect only in the first growing season [6].In turn, the negative effect of fire can be expressed in a significant decrease in the participation of the species up to its absence in the first seasons after burning (for example, for Dianthus deltoides, Trifolium hybridum).In this case, restoration of occurrence takes a significantly longer period due to different probability of diaspore entry/resprouting of the species, competition between different species, etc.
In general, however, it should be concluded that a single fire impact on a limited area of the experimental sites did not lead to significant changes in the composition of the community, and 2/3 of plant species regained their occurence already in the second growing season.At the same time, all tree species show a decrease in occurrence until the 2nd -3rd growing season, which confirms the effectiveness of prescribed burning to control undergrowth and correlates with the two-year burning period most often used for prescribed fires [5].At the same time, the fact that a number of herbaceous plant species show lower occurrence in the third growing season after the fire is consistent with the assumption that it is necessary to preserve areas with longer periods of burning to preserve biodiversity [5].This is also supported by evidence that many treeless communities show recovery after fire in 3-4 years [8][9][10].

Conclusion
As a result of our experiment, it was established that a single spring fall on the territory of the BEC with forb-grass vegetation does not lead to significant restructuring of plant communities, and the occurrence of the vast majority of plant species is restored within up to 3 growing seasons.It is also indirectly confirmed that, in terms of plant biodiversity preservation, mowing is more preferable as a form of management in grassland areas in comparison with periodic burnout.