Response of Siberian River Discharge to Disturbances of Forests Caused by Wildfires

: The objective of this work was to perform a quantitative analysis of the correlation between the forest burning index and abnormal decrease in river discharge under conditions of the cryolithozone of Siberia. We analyzed the long-term and seasonal variation of river discharge in Central Siberia (Nizhnyaya Tunguska and Podkamennaya Tunguska rivers) and in Eastern Siberia (Aldan, Viluy rivers) together with the forest burning dynamics within the river basins. The data on river discharge were obtained from the archive of The Global Runoff Data Centre for 1939–2015. The relative burned area (RBA) index was calculated from wildfire databases collected using satellite technique for 1996–2017. RBA was evaluated as the ratio of the annual burned area within the river basin to the total area of the river basin. RBA values of 2.5–6.1% per year were considered as extremely high. The analysis of available chronologies of extreme fire events in Central and Eastern Siberia showed high correlation (r > − 0.55) with long-term data on the runoff minima. Abnormally low levels of discharge were 68–78% of the averaged annual rate. The most significant response of river discharge to the wildfire effect was shown for the summer–autumn period of the season after extreme burning in mid-summer. to significance is as by the of and post-fire in the permafrost soil


Introduction
In Siberia, significant and long-term post-fire effects are observed in the permafrost zone [1][2][3][4]. In particular, these are changes in and degradation of the near-surface layers of permafrost, and shortterm and long-term anomalies of temperature and water balance [5][6][7][8][9]. This affects the flow regime of small and medium rivers of Siberia, the supply of which is determined by groundwater (10-25% of total) [10]. Interannual fluctuations and trends in river discharge are mainly associated with climatic processes in Siberia [11][12][13][14], and the results of analyses of wildfire impact on river flows are have not been thoroughly discussed in the literature.
In this work, we determined the degree of connection between intra-and interseasonal variations in river runoff with the relative burned area (RBA) of forests in the river basins of Siberia. The following aspects of the issue were considered: (i) The correlation of long-term data on river flow anomalies and RBA; (ii) the intraseasonal variations of the river flow under post-fire conditions; and (iii) features of post-fire river runoff dynamics.
The study of this issue allows us to predict the long-term response of the boreal ecosystems to the fire impact, one of the most significant factors under current climate conditions and fire regimes [3,13].

Results
Data on fires and river discharge are presented in Table 1.
In some seasons, we fixed the level of runoff at 68-78% of the average annual rate. When analyzing the available chronologies of extreme fire events in Central and Eastern Siberia [15][16][17][18], it was possible to compare the discharge minima with extreme fire events (Figure 1). The frequency of extremely low runoffs, ranging from 18 to 25 years, is consistent with the reported data on the variability of the width of the tree rings in larch forests of Central Siberia [19], which is determined by the temperature and the moisture regimes of weather. Thus, the phase coincidence of the flow anomalies and extreme fire events associated with the precipitation deficit is expected.  To solve the problem of quantitative descriptions of the relationship, we jointly analyzed data on the forest fire in the borders of river basins (γ, %) and runoff anomalies for the first half of the growing season (March-July) for 2002-2015 ( Figure 2). The results of the correlation analysis of the relationship between the intraseasonal dynamics of the discharge and the RBA are presented in (Table 2).
All figures and tables should be cited in the main text as Figure 1, Table 1, etc.

Discussion
In Ref. 14, it was shown that the data on the moisture content in the soil are an effective indicator of the prediction of forest burning in the permafrost zone of Siberia. At the same time, it is noted that the moisture reserves in the soil in the current season determines the degree of fire danger of this and the subsequent season. To use these results in fire monitoring practice requires a wide network of data collection points, which is not always possible to implement in remote areas of Central and Eastern Siberia. An alternative solution is the development of methods for remote monitoring of water mass dynamics according to gravimetric survey data [1,20]. In our work, a qualitatively similar result was obtained, while the data on anomalies of river discharge are a more effective criterion for predicting the fire regime within the river basins.
The response to the fire impact was also recorded in the territories of the considered basins of the rivers in Central Siberia, which expressed an abnormal low discharge in the post-fire summer-autumn period (r > −0.55). At the same time, the level of relation is lower for the basins of rivers in Eastern Siberia/Yakutia (r < −0.45).
The revealed differences can be a consequence of the post-fire condition of permafrost soils, which determines the share of groundwater in the formation of the total river flow. The post-fire transformation of vegetation and on-ground cover can be the cause of heat and water balance anomalies [17,21], changes in the depth of seasonally thawed layers of soils, or changes in water permeability of soil horizons [22,23]. Thus, if we do not take into account seasonal variations in the precipitation regime, the features of the post-fire discharge anomalies are determined by the interdependency of "fire effect", "ground cover and vegetation", and "soil". The influence of wildfires is significant only for the seasonally thawed layer that is active in the summer-autumn period.
A more detailed study of post-fire effects on river discharge anomalies is important for predicting the response of boreal ecosystems to the fires effects, which currently tends to increase [1,4,24].

Materials and Methods
The area of interest is the territory of Siberia within the boundaries of 57-67 N, 85-110 E. The total area is more than 110 million hectares. The studies were performed for four river basins of Central Siberia and Yakutia (Figure 3), such as Lower Tunguska, Podkamennaya Tunguska (Basin District of Yenisei River), and Aldan, Viluy (Basin District of Lena River). We determined the average annual value of the discharge ( ) and analyzed deviations ( * ) from the average statistical norm (discharge anomalies) for the each month (i) as * = ( ) × 100%. (1) We determined the relative burned area (RBA) of forests within the river basins on the basis of satellite fire monitoring data of the Sukachev Institute of Forest (Federal Research Center KSC SB RAS, Krasnoyarsk, Russia). The data were presented in the format of a geoinformation (Geographic Information System, GIS) wildfire database for 1996-2015 [28]. RBA (γ) was defined for each month as the ratio of the total area of fires (Sburned) to the total area of the river basin (S): Data processing was performed using geospatial pre-processing of vector data layers using GIS tools, correlation and statistical analysis, and a method for the optimization of regression coefficients.

Conclusions
For the current river basins, the scale of wildfire impact is up to 2.5-6.1% of the total area per year. Its effects are strong on the forest ecosystems of the permafrost zone. Within the river basins of Central Siberia, the response to pyrogenic (post-fire) impact is expressed in anomalies of discharge in the post-fire summer-autumn period (r > −0.52). For river basins in Eastern Siberia, the correlation is lower. The level of significance is determined as highly likely by the state of and post-fire changes in the permafrost soil conditions. Author Contributions: Conceptualization, E.I.P. and T.V.P.; methodology, E.I.P.; writing-original draft preparation, E.I.P. and T.V.P.; writing-review and editing, E.I.P.; funding acquisition, E.I.P., T.V.P.
Acknowledgments: This research was funded by the Russian Foundation for Basic Research (RFBR) grant number 17-04-00589 "Radiometric method for quantitative analysis of soils structure organization", RFBR and Government of the Krasnoyarsk krai, and Krasnoyarsk krai Foundation for Research and Development Support, grant number 18-41-242003 "Modeling and satellite monitoring of effects from thermal anomalies of he underlying surface in the seasonally thawed soil layer of the permafrost zone of Siberia". River discharge data were provided by The Global Runoff Data Centre, 56068 Koblenz, Germany. The satellite data-receiving equipment used was provided by the Regional Center for Remote Sensing, Federal Research Center "Krasnoyarsk Science Center, Siberian Branch of Russian Academy of Sciences", Krasnoyarsk, Russia.

Conflicts of Interest:
The authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript: RBA Relative burned area RIMS Rapid Integrated Monitoring System RD river discharge KSC SB RAS Krasnoyarsk Science Center of Siberian Branch of Russian Academy of Sciences GIS Geographic Information System