Water Regulating in Kenozero Taiga: Excess or Lack of Water and Where Does It Go?

Round 1

Reviewer 1 Report

The purpose of this article is to compare the water regulating ES by forest under different land management scenarios, further the currently understudied ecosystem function of recycling atmospheric moisture and precipitation. This paper is interesting and necessary. However, the authors need to do much more to improve the paper and make a good scientific story. Kindly, see below for detailed comments:

 

1 In the abstract, the research results should be introduced in detail. For example, water yield regulation due to evapotranspiration of forests; prevention of soil erosion, and soil flushing into water bodies.

2 Materials and Methods, the environmental characteristics (e.g., soil characteristics, rainfall, temperature) should be introduced in detail.

3 Figure 2.1., provide a longitude and latitude grid, scale unit, and just the landcover map of the study area.

4 Materials and Methods, how to solve the problem of data resolution?

5 The formula of water yield regulation and prevention of soil erosion should be added.

6 Materials and Methods, I doubt whether the value of Kc and the land use factor (C) for each type of land cover is scientific.

7 Table3.1 the sentence ‘Change in a few years after total felling (compared to 2021)’ statement is inaccurate.

8 Lines 215-216 our estimates show that the runoff from the study area is about 160 mm/year (Table1, Figure 3.2.1 a), which is 26% of precipitation (605 mm/year). Where did 160 come from? Table1 should be Table 3.1.

9 The statistical chart of each land use of ES should be added.

 

10 The Conclusion was too long, please highlighting the significance and guiding role of this study.

Author Response

Dear reviewer, thank you for the helpful pointers on the shortcomings of our manuscript. We tried to fix them. We also made some corrections to the English language. All corrected text passages are highlighted in yellow in the manuscript.

Comment 1: In the abstract, the research results should be introduced in detail. For example, water yield regulation due to evapotranspiration of forests; prevention of soil erosion, and soil flushing into water bodies.

Response: The main results of the ES quantification were presented in abstract. The following text was added (lines 19-24):

“Water yield reduction due to ecosystem evapotranspiration was estimated as 125 mm/year and soil erosion prevention was estimated as 9.6 t/ha/year. Forest felling in the study area from 2007 to 2021 led to an increase in runoff and soil loss by 6 mm/year and 0.03 t/ha/year respectively. Hypothetical total instantaneous forest loss can lead to an increase in runoff and soil loss by 71 mm/year and 2.4 t/ha/year respectively”.

Comment 2: Materials and Methods, the environmental characteristics (e.g., soil characteristics, rainfall, temperature) should be introduced in detail.

Response: Environmental characteristics were described in more detail in the Section 2.1 (lines 102-138):

"The study area is located in the northern moraine part of the Russian Plain, within Fennoscandia (territory center coordinates 62͑॰15’N, 38॰15’E). It belongs to Dfc climate type (continental boreal climate with cold summer and without dry season) according to the Koppen-Geiger classification scheme. The climate is formed under the influence of the Arctic seas and intensive transfer of atmospheric masses from the west. The average annual air temperature is about +2°C. The average duration of the frost-free period is 108 days. Snow cover usually appears in October, lasts 175 days on average and melts until May 1st. The growing season lasts from May to October. The average annual rainfall is 605 mm/year. The minimum precipitation falls in February, the maximum in June-September.

The study area includes the catchment of the lake Kenozero (620 thousand ha) belonging to the White Sea basin and the territory of Kenozero National Park (Figure 1). The catchment of the lake Kenozero includes most of the territory of the Kenozero National Park (140 thousand ha). Two areas in the southern part of the park belong to the basin of the lake Lache (White Sea basin) and the basin of the Baltic Sea. Maselga watershed ridge between the basins of the Atlantic and Arctic oceans is a unique geographic object in the territory of the park. From it, two lakes belonging to the basins of different oceans are visible.

The relief is flat with pronounced fluvioglacial forms: elongated ridges, rounded hills and outwash plains. The ancient crystalline foundation is covered with thick layers of poorly permeable sedimentary rocks. Slow runoff processes in the flat area and shallow groundwater table led to the development of swamps. The largest bog massifs are located in extensive depressions with glacial sandy and sandy loam deposits. 

The study area is located in the zone of boreal forests of middle taiga.  Forests and swamps together occupy 90% of the area (Figure 1) and are typical for the subzones of the northern and middle taiga of the European part of Russia. Spruce forests are the primary zonal formation. Pine forests are also common. Deciduous and mixed forests are located in areas that were previously used for agriculture or were cut down for timber. 6% of the area is occupied by water bodies. Few grasslands (less than 0.1%) are of predominantly anthropogenic origin.

Soils of podzolic, bog-podzolic, bog types are common in the territory. Gley-podzolic soils lie under spruce forests. Typical podzolic soils are common on hillsides in well-drained watersheds under coniferous forests. Illuvial-humus and ferruginous podzols are developed on sands under lichen pine forests. More fertile soddy-podzolic soils are typical in well-drained areas under deciduous and mixed forests. Poor waterlogged bog-podzolic soils (peaty-podzolic) are common on flat poorly drained loam watersheds, under spruce or spruce-pine forests. Bog soils are developed in low areas with constantly excessive moisture".

Comment 3: Figure 2.1., provide a longitude and latitude grid, scale unit, and just the landcover map of the study area.

Response: Figure 2.1 was corrected. Longitude and latitude grid, and scale unit were shown.

Comment 4: Materials and Methods, how to solve the problem of data resolution?

Response: The solution to this problem was explained in the following paragraph in the Section 2.2 (lines 161-164):

"All the above and below mentioned raster spatial datasets used in this study were reprojected into the WGS 84 / UTM zone 37N projection and resampled to a minimum spatial resolution equal to the size of one pixel of the vegetation map (i.e., 30 m). The calculations were carried out in the ArcGIS 10.5 software".

Comment 5: The formula of water yield regulation and prevention of soil erosion should be added.

Response: Formulas for the ES of water yield were inserted into the Section 2.3 (lines 185-203): 

“The ES of water yield regulation was estimated and mapped using the “Annual Water Yield” module of InVEST. The annual water yield Y(x) was calculated as the difference between the precipitation in the study area and the actual evapotranspiration by the formula (1):

Y(x) = P(x) - AET(x)                                                                           (1),

where P(x) is the annual precipitation and AET(x) is the annual actual evapotranspiration in pixel ?. 

For built-up areas and wetlands AET was calculated directly from the reference evapotranspiration ??0(?) and has an upper limit determined by precipitation (2):

AET(x) = Min (PET(x), P(x)) = Min (??(ℓ?)??0(?), ?(?))                          (2), 

where PET(x) is the potential evapotranspiration, ET0(x) is reference evapotranspiration, P(x) is the annual precipitation in pixel ? and Kc(lx) is the evapotranspiration coefficient for each landcover type.

For vegetation, the formulas for AET calculating was as follows (3, 4):

 (please, see the attached file)             

where N(x) is the number of rain events per year, PAWC(x) is plant available water content, h(x) is root bounding layer depth defined as (5):

h(x) = Min (Root restricting layer depth; Root depth)                                  (5).

Formulas for the ES of prevention of soil erosion were inserted into the Section 2.3 (lines 261-270): 

The ES of prevention of soil erosion and soil flushing into water bodies was evaluated using the “Sediment Delivery Ratio” module. The calculation of potential soil erosion from bare soil USLEbare in the InVEST program is based on the universal soil loss equation (USLE) by formula (6):

USLEbare = RKLS                                                                              (6),

where R is rainfall erosivity factor, K is soil erodibility factor, LS is terrain factor. 

Actual soil erosion was calculated by formula (7):

USLE(l(x)) =USLEbare C(l(x)) P(l(x))                                              (7),

where C is the cover management factor, and P is the support practice factor for each landcover type.

Comment 6: Materials and Methods, I doubt whether the value of Kc and the land use factor (C) for each type of land cover is scientific.

Response: The choice of Kc values was explained in the Section 2.3 (lines 222-256):

“Evapotranspiration coefficient (Kc) for all forest types was set as 1.00, for croplands and grasslands as 0.67, for sphagnum bogs as 1.00, for sedge and grass bogs as 1.10, for wet swamps and bogs as 1.20, for bare ground and fresh clearings as 0.50, and for built-up areas as 0.56.

• The coefficient for forests was calculated based on the Leaf Area Index (LAI) for the growing season according to 300-m resolution data of the European Union Earth Observation Program Copernicus [40]. LAI values exceeding 3.0, which is typical for forests on study area, were equated to Kc = 1.0 in the InVest model.
• In the study area, areas defined as croplands on the land cover map are represented by perennial grasses, therefore, the same value of Kc (clover hay, averaged cutting effects) was used for grasslands and croplands from the FAO Guidelines for computing crop water requirements [FAO, 1998].
• The growing season was considered from May to October [41]. The beginning of the initial stage was determined by the average date of the resumption of the vegetation of winter rye and by the date of transition of the average daily air temperature through 5 degrees. In both cases it is April 30. The initial stage falls in May and June. The beginning of the middle stage, i.e., full summer, was determined by the date of earring of spring wheat (July 15) and the date of blueberry ripening (July 25). This stage falls in July and August. The end stage begins when the air temperature passes through 10 degrees (September 5), that is, it falls in September and October.
• Kc values for wetlands was determined based on the principle that the more water saturated the ecosystem, the greater the value. InVest suggests that Kc for wetlands can be assumed in the range of 1.0 to 1.2. Therefore, we used the value 1.0 for sphagnum bogs, 1.1 for sedge and grass bogs, and 1.2 for wet swamps and bogs.
• Fresh clearcuts (less than 3 years) were merged into one class with bare ground, because destroyed vegetation in the middle taiga recovers slowly.
• In accordance with InVest, Kc for built-up areas was calculated as Kc = 0.1f + 0.6(1-f) where f is the fraction of impervious surface in the area. There are no paved roads in the study area, so the only impervious surfaces are roofs of houses. For example, in the Pershlakhta village (the area of the village 37374 m2), the total area of buildings is 8,5% (3176 m2) (Figure 2), which is close to average value. Thus, the share of impermeable surface for the built-up land cover class was set as 8.5%.

The cover management factor C was set in accordance with global and European reviews of C factor values for natural and arable lands (Ebabu et al., 2022; Panagos et al., 2015).

Comment 7: Table 1: the sentence ‘Change in a few years after total felling (compared to 2021)’ statement is inaccurate.

Response: The inaccurate writing of the indicator was replaced by the following "Change in 3 years after total felling (compared to 2021)”

Comment 8: Lines 215-216 our estimates show that the runoff from the study area is about 160 mm/year (Table1, Figure 3.2.1 a), which is 26% of precipitation (605 mm/year). Where did 160 come from? Table1 should be Table 3.1.

Response: This error has been corrected in the text as follows (lines 310-313): “Our estimates show that the runoff from the study area varies from 159 mm/year in 2007 to 164 mm/year in 2021 (Table 1, Figure 3 a), that is 26% of precipitation (605 mm/year)”.

Comment 9: The statistical chart of each land use of ES should be added.

 Response: Statistical charts of each land use of ES were added to the manuscript (Figures 4 and 7). Short explanations of these charts were added to the Sections 3.2 and 3.3.

Comment 10: The Conclusion was too long, please highlighting the significance and guiding role of this study.

Response: The Сonclusion was completely rewritten and shortened (lines 599-613):

Our estimation of two water regulating ES (water yield regulation due to evapotranspiration and water quality assurance due to prevention of soil erosion) revealed the trade-off between them. Forest harvesting leads to water yield increase due to evapotranspiration weakening and water quality decrease due to soil erosion. Formally, this tradeoff makes it difficult to choose between harvesting and conservation forest management strategies, while the profit from timber increases the likelihood of choosing harvesting strategy.

However, the use of a more complete set of ES (including regulation of water table and precipitation recycling), even at the level of approximate expert estimates, increases the likelihood of choosing a conservation strategy instead of a harvesting one in forest management. Given the projected future climate change, the most important forest ES, which should be additionally included in the decision-making process, is recycling and redistribution of air moisture and precipitation. Incorporating this ES into decision-making requires extending the spatial scale from local watersheds to precipitation-sheds of subcontinental and even continental scale.

 

 

 

 

 

 

Author Response File: Author Response.pdf

Reviewer 2 Report

General comment

The study is interesting in the sense that it performs an ecosystem valuation focused on the management and recovery of ecosystems of conservation interest. However, the document requires more finesse, style correction and lexicographical precision. 

 

Specific comments

Introduction

It is necessary that the authors include the hypothesis of their research.

 

Materials and methods

A better characterization of the study area is required, including geographic coordinates, climate, soils and the general floristic component of the zone.

The hyperparameters considered in the modeling must be better specified using InVEST model

Explain in more detail each of the environmental parameters taken from different climatological data sources. Do the same with erosion parameters

Results

All maps require the location of the north and the scale worked (1:25.000: 1:100.000?).

The numbering of tables and figures does not correspond to instructions of Earth journal.

Homogenize the unit and decimal systems throughout the document. 

 

Discussion

Do not underline sentence lines

what is the relevance of placing figure 4.2.1 and table 4.3 in the discussion?

It is necessary to strengthen the discussion in terms of using other proven models for similar studies and compare with their findings.

What is the implication of your study in terms of vulnerability to climate variability and climate change? 

What impact does your study have in terms of conservation, sustainable forest management and restoration?

Conclusions

Conclusions should be adjusted in relation to the objectives of the work, rather than a checklist of the manuscript.

Author Response

Dear reviewer, thank you for the helpful pointers on the shortcomings of our manuscript. We tried to fix them all. We also made some corrections to the English language. All corrected text passages are highlighted in yellow in the manuscript.

Comment: Introduction. It is necessary that the authors include the hypothesis of their research.

Response: Dear reviewer, thank you for pointing out this important shortcoming. For a clear formulation of the objectives of the article and the hypotheses to be tested, the introduction was restructured and partially rewritten (the changed text is highlighted in yellow in the manuscript). Justification and formulation of hypotheses are presented in the following two paragraphs (lines 76-87):

"Different possibilities for evaluating different ES can lead to a bias in favor of easier-to-estimate ES, while difficult-to-estimate ES can be underestimated. Such biases in ES assessment can lead to management errors [4, 5]. Another difficulty for decision-making are trade-offs between different ES, when strengthening of one ES leads to weakening of the other. Among water regulating ES, trade-off is usually found be-tween water yield and water quality [3, 5].

The purpose of this article is to compare the management interpretation of a partial ES assessment by two water regulating ES (i.e., ensuring yield and quality of water) with assessment of a wider ES range using the watershed of the lake Kenozero as an example. The tested and discussed hypotheses are as follows: a) the presence of a trade-off between the ES of ensuring water yield and water quality; b) increasing the likelihood of choosing a conservation strategy instead of harvesting one in forest management".

Comment: Materials and methods. A better characterization of the study area is required, including geographic coordinates, climate, soils and the general floristic component of the zone.

Response: The territory was characterized in more detail (lines 102-138):

"The study area is located in the northern moraine part of the Russian Plain, within Fennoscandia (territory center coordinates 62͑॰15’ N, 38॰15’ E). It belongs to Dfc climate (continental boreal climate with cold summer and without dry season) according to the Koppen-Geiger classification scheme. The climate is formed under the influence of the Arctic seas and intensive transfer of atmospheric masses from the west. The average annual air temperature is about +2°C. The growing season lasts from May to October. The average annual rainfall is 605 mm/year.

The study area includes the catchment of Lake Kenozero (620 thousand ha) belonging to the White Sea basin and the territory of Kenozero National Park (Figure 1). The catchment of Lake Kenozero includes most of the territory of the Kenozero National Park (140 thousand ha). Maselga watershed ridge between the basins of the Atlantic and Arctic oceans is a unique geographic object in the territory of the park. From it, two lakes belonging to the basins of different oceans are visible.

The relief is flat with pronounced fluvioglacial forms: elongated ridges, rounded hills and outwash plains. The ancient crystalline foundation is covered with thick layers of poorly permeable sedimentary rocks. Slow runoff processes in the flat area, as well as groundwater close to the surface, led to the development of swamps. The largest bog massifs are located in extensive depressions with glacial sandy and sandy loam deposits.

The study area is located in the zone of middle taiga forests.  Almost all the area is covered with forests and swamps which together occupy 90% of the area (Figure 1). Forests and swamps are typical for the subzones of the northern and middle taiga of the European part of Russia. Spruce forests are the primary zonal formation. Pine forests are also common. Deciduous and mixed forests are located in areas that were previously used for agriculture or were cut down for timber. 6% of the area is occupied by water bodies. Few grasslands (less than 0.1%) are of predominantly anthropogenic origin.

Soils of podzolic, bog-podzolic, bog types are common in the territory. Gley-podzolic soils lie under spruce forests. Typical podzolic soils are common on hillsides in well-drained watersheds under coniferous forests. Illuvial-humus and ferruginous podzols are developed on sands under lichen pine forests. More fertile soddy-podzolic soils are typical in well-drained areas under deciduous and mixed forests. Poor waterlogged bog-podzolic soils (peaty-podzolic) are common on flat poorly drained loam watersheds, under spruce or spruce-pine forests. Bog soils are developed in low areas with constantly excessive moisture".

Comment: The hyperparameters considered in the modeling must be better specified using InVEST model

and

Comment: Explain in more detail each of the environmental parameters taken from different climatological data sources. Do the same with erosion parameters.

Response: Environmental parameters used for estimation and mapping water yield and erosion were explained in more detail in the Section 2.3 (lines 204-256 and 274-299):

The following parameters were used for the ES calculation and mapping.

• Average annual precipitation from WorldClim global dataset [36] expressed in mm/year.
• Average annual reference evapotranspiration ET0 from Global Aridity and PET Database, CGIAR [37]. Alfalfa or sward are considered as a reference vegetation cover. Based on ET0, the potential evapotranspiration PET of each land cover type was calculated using the evapotranspiration coefficient Kc (see below).
• Root restricting layer depth from the Harmonized World Soil Database, HWSD [38]. The “Roots” field was used, which expresses the presence of obstacles to the roots at various depths of the soil profile. We used the average values for the depth of obstacles for the roots: class 1 - 800 mm, class 2 - 700 mm, class 3 - 500 mm, class 4 - 300 mm, class 5 - 200 mm, class 6 - 100 mm.
• Plant available water content PAWC calculated in SPAW (Soil-Plant-Air-Water) software based on soil texture data (texture class, the percentage of sand and clay) from HWSD.
• Root depth containing 90% of root biomass according to [39]. This value is 1500 mm for taiga forests and 100 mm for herbaceous communities (i.e., grasslands and croplands on the land cover map of the study area).
• Evapotranspiration coefficient (Kc) for all forest types was set as 1.0, for croplands and grasslands as 0.67, for sphagnum bogs as 1.0, for sedge and grass bogs as 1.1, for wet swamps and bogs as 1.2, for bare ground and fresh clearings as 0.50, and for built-up areas as 0.56.
• The coefficient for forests was calculated based on the Leaf Area Index (LAI) for the growing season according to 300-m resolution data of the European Union Earth Observation Program Copernicus [40]. LAI values exceeding 3.0, which is typical for forests on study area, were equated to Kc = 1.0 in the InVest model. 
• In the study area, areas defined as croplands on the land cover map are represented by perennial grasses, therefore, the same value of Kc (clover hay, averaged cutting effects) was used for grasslands and croplands from the FAO Guidelines for computing crop water requirements [FAO, 1998].
• The growing season was considered from May to October [41]. The beginning of the initial stage was determined by the average date of the resumption of the vegetation of winter rye and by the date of transition of the average daily air temperature through 5 degrees. In both cases it is April 30. The initial stage falls in May and June. The beginning of the middle stage, i.e., full summer, was determined by the date of earring of spring wheat (July 15) and the date of blueberry ripening (July 25). This stage falls in July and August. The end stage begins when the air temperature passes through 10 degrees (September 5), that is, it falls in September and October. 
• Kc values for wetlands was determined based on the principle that the more water saturated the ecosystem, the greater the value. InVest suggests that Kc for wetlands can be assumed in the range of 1.0 to 1.2. Therefore, we used the value 1.0 for sphagnum bogs, 1.1 for sedge and grass bogs, and 1.2 for wet swamps and bogs.
• Fresh clearcuts (less than 3 years) were merged into one class with bare ground, because destroyed vegetation in the middle taiga recovers slowly.
• In accordance with InVest, Kc for built-up areas was calculated as Kc = 0.1f + 0.6(1-f) where f is the fraction of impervious surface in the area. There are no paved roads in the study area, so the only impervious surfaces are roofs of houses. For example, in the Pershlakhta village (village area is 37374 m2), the total area of buildings is 8,5% (3176 m2) (Figure 2), which is close to average value. Thus, the share of impermeable surface for the built-up land cover class was set as 8.5%.

The following parameters were used to calculate potential soil erosion:

• Rainfall erosivity factor R was determined from 1 km resolution data of the Join Research Center of the European Soil Data Center (ESDAC) [44]. On the study area, this parameter varies from 280 to 335 ?? ??/ℎ?/ℎ?.
• The soil erodibility factor K was determined based on the soil types of the Harmonized World Soil Database HWSDB and the methodology of the Ontario Department of the ministry of Agriculture, Food and Rural Affairs [45]. One of the most important properties which determine soil erodibility is the content of clay, sand, silt, and coarse-grained particles. For soils in the study area, K value varies from 0.005 to 0.05.
• The terrain factor LS was calculated based on a digital elevation model of the study area by the method of Desmet and Gowers (1996) for a two-dimensional surface. It considers, in addition to the classical parameters of the length L and steepness S of the slope, also the exposure of the slope and the flow accumulation area at the inlet to the grid cell.
• The cover management factor C was set as the average values for land cover categories according to the materials [46, 75]: 001 for swamps, 0.003 for forests, 0.31 for clearcuts (as for degraded land), 0.27 for unvegetated anthropogenic areas.  The coefficient for croplands and grasslands was calculated using above-mentioned methodology for the state of Ontario [45]. Two parameters were considered: crop type that is grass in the study area (coefficient is 0.02) and the absence of specialized processing equipment of tillage (coefficient is 0.25). Multiplying these two coefficients, C factor for croplands and grasslands turns out to be extremely low (0.005) that is, of the same order of magnitude, for swamps.
• The support practice factor P is the coefficient of soil loss as a result of the presence of certain soil-protective structures (field-protective and snow-retaining forest belts, tree and shrub shafts, shrub thickets and meadows in erosional landforms). It was set to 1.0, as no such structures are presented in the study area.

Comment: Results. All maps require the location of the north and the scale worked (1:25.000: 1:100.000?).

Response: North indication and scale (1:500 000) added to all maps.

Comment: The numbering of tables and figures does not correspond to instructions of Earth journal.

Response: The numbering of figures and tables is made in accordance with the requirements of Earth journal

Comment: Homogenize the unit and decimal systems throughout the document. 

Response: The units and decimal systems (using decimal point) were homogenized throughout the document

Comment: Discussion. Do not underline sentence lines

Response: Underlines have been removed from everywhere in the text

Comment: what is the relevance of placing figure 4.2.1 and table 4.3 in the discussion?

Response: Dear reviewer, thank you for pointing out that the figure 4.2.1 and table 4.3 explain almost nothing in addition to the text. These figure and table have been removed from the text.

Comment: It is necessary to strengthen the discussion in terms of using other proven models for similar studies and compare with their findings.

Response: Unfortunately, we are not aware of studies that have applied other models to assess water regulating ES in the study area or in adjacent regions with similar conditions. Comparison with the results of InVEST application in other regions is presented in the Sections 4.1 and 4.2.

Comment: What is the implication of your study in terms of vulnerability to climate variability and climate change? 

Response: Dear reviewer, thank you for pointing out the insufficient attention to this aspect, which is extremely relevant in the current conditions of global change. The implications of our analysis for the resilience of taiga ecosystems under climate change were reflected in the Section 4.3 (lines 527-541):

The ES of recycling and redistribution of atmospheric moisture and precipitation appears to be the most important for the region. As mentioned above, in the boreal forest zone a third to half of the precipitation can be the result of this ES [9]. In the context of climate change it can be of key importance for ecosystem stability and climate regulation. For example, a realistic reforestation can increase summer precipitation by 7.6 ± 6.7% on average over Europe, potentially offsetting a substantial part of the projected precipitation decrease from climate change (Meier et al., 2021). The function of forests to enhance the movement of moist air masses from the oceans into the continents [18, 19] enhances forest importance for future climate regulation. Precipitation reduction is not predicted for our study area. In recent years there has been a slight increase in precipitation in the European taiga zone (5% over 10 years). However, the simultaneous increase in temperature (0.6 degrees over 10 years) makes the problem of maintaining the water balance of ecosystems very important. These trends are predicted to continue in the future (Report… 2017; Report…, 2022) along with an increase in the risk of natural fires and an increase in the fire hazard period [67].

This issue is also emphasized in the Conclusion (lines 609-611):

“Given the projected future climate change, the most important forest ES, which should be additionally included in the decision-making process, is recycling and redistribution of air moisture and precipitation”.

Comment: What impact does your study have in terms of conservation, sustainable forest management and restoration?

Response: The following paragraphs have been added to the text on this issue (lines 583-597):

Thus, our results and discussions highlighted the following issues in the use of assessments of water management services in forest management:

- Evaluation of only one selected water regulating ES, especially such as regulation of water yield due to evapotranspiration, can lead to incorrect management conclusions about the optimal intensity of forest exploitation.

- Considering a broader range of ES strengthens arguments for a forest conservation strategy instead of a forest exploitation.

- The inclusion of the ES of precipitation recycling in decision-making is necessary in the context of climate change. It requires a transition from the scale of a local catchments to a regional or even continental scale the development of interregional and international markets of water regulating ES of forests.

- Local assessments and mapping of individual ES are useful for territorial planning of forest protection and use within certain areas, as they allow to identify the importance of different types of vegetation, features of soils and topography in maintaining ES (Sections 3.2 and 3.3).

Comment: Conclusions should be adjusted in relation to the objectives of the work, rather than a checklist of the manuscript.

Response: The Сonclusion was completely rewritten and shortened (lines 599-613):

Our estimation of two water regulating ES (water yield regulation due to evapotranspiration and water quality assurance due to prevention of soil erosion) revealed the trade-off between them. Forest harvesting leads to water yield increase due to evapotranspiration weakening and water quality decrease due to soil erosion. Formally, this tradeoff makes it difficult to choose between harvesting and conservation forest management strategies, while the profit from timber increases the likelihood of choosing harvesting strategy.

However, the use of a more complete set of ES (including regulation of water table and precipitation recycling), even at the level of approximate expert estimates, increases the likelihood of choosing a conservation strategy instead of a harvesting one in forest management. Given the projected future climate change, the most important forest ES, which should be additionally included in the decision-making process, is  recycling and redistribution of air moisture and precipitation. Incorporating this ES into decision-making requires extending the spatial scale from local watersheds to precipitation-sheds of subcontinental and even continental scale.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

All comments were correctly addressed by the authors, so that the manuscript was substantially improved for publication.