Effect of strip clear-cutting on the natural regeneration of Pinus tabuliformis plantations in northeastern China

Study area

This research was conducted in the Liaoning ecological and experimental farm located in the semiarid area of the western part of the Liaoning Province in northeastern China (120°15–121°18E, 41°23–42°17N). Continental monsoon climate of the northern temperate zone is the typical climate of this area. The mean annual temperature is 8.3 °C and the mean annual precipitation ranges from 450 to 550 mm. The elevation ranges from 200 to 1,074 m and 145–150 days are frost-free. Brown forest soil is the main soil type in this area. Pinus tabuliformis, Populus alba, Salix matsudana, Ulmus macrocarpa, Quercus mongolica, and Tilia mandshurica are the main tree species for this region of China.

The study area is located in Pinus tabuliformis plantation in the Linghe experimental region (location A) from Liaoning ecological and experimental farm, Songzhangzi experimental region (location B) from Qitian forest farm, and Shahai forest farm (location C). Location B and C are within Jianping county with the distance of approximate 37 km, and Location A is from Beipiao county with the distance to Jianping of 140 km. The forest age and density for location A, B, and C are 45, 43, and 46 years and 1,035, 1,058, and 1,002 trees per hectare, respectively. In the spring of 2014, a clear-cut strip experiment was established on around 50 hectares in a mature Pinus tabuliformis plantation. All the three experimental locations show identical topographic characteristics with the slope smaller than 5° and southeast-facing aspect. The length of strips varied from 200 to 300 m. The orientation of the strips for the three locations were in the south–north direction. The widths of the clear-cut strips and uncut strips varied from 13 to 25 m, and 5 to 20 m, respectively. To avoid the effect of width on the results between the three locations, our investigation was conducted at the clear-cut strips with the widths of 15, 20, and 25 m and the uncut strips with the widths of 10 and 18 m for all the three locations (Fig. 1). The average mature pine height in locations A, B and C was 15.9, 15.4, and 15.5 m, respectively, and all understory trees were naturally regenerated. Harvesting was conducted as carefully as possible to avoid damage to saplings and the ground surface. Timber and slash were removed immediately after harvesting was completed.

Data collection

Field investigations were conducted in the strip clear-cut experimental zone of the Pinus tabuliformis plantation in August 2017 for location A, in December 2017 for location B and C. All measurements were conducted in clear-cut strips with widths of 15, 20, and 25 m, uncut strips with widths of 10 and 18 m, as well as untreated sites (approximate 100 m × 100 m for location A, 95 m × 95 m for location B, 110 m × 100 m for location C) served as a control. In each study location, 12 to 15 sample plots were established within each treatment in the clear-cut strip, uncut strip, and control. Plots measured 4 m² (2 m × 2 m), and to avoid edge effects between the clear-cut and uncut strips, all sample plots were established in the center of each strip at distances of 10 m. For each sample plot, the number of seedlings and saplings were recorded in the field. The sapling age was estimated by the number of whorls for the live branches, and the scar remaining on a stem after the branch falls along the sapling bole. Seedling was defined as a height below 30 cm and sapling was defined as a height above 30 cm and below 1.3 m. A total of 1 to 4 saplings with healthy crowns and no apparent damage to the branches was randomly selected from each sample plot. The sapling selection criterion was a distance larger than 1 m between two saplings. For each selected sapling the total sapling height and height of the leader (herein referred to as current year increment) was recorded using a measuring tape to the nearest cm. The lengths of the first whorl branches were also measured and the number of branches for the first whorl was also recorded. From each strip, one large (with the average sapling height for the 10 largest saplings), one medium (with average sapling height), and one small (with the sapling height for the 10 smallest saplings) sampled sapling in the center of the strip were randomly selected to measure the photosynthetic rate using a portable photosynthesis system (LI-6400XT; LI-COR, Inc., Lincoln, NE, USA). Two cloud-free days in August were chosen to conduct the photosynthesis measurements. The photosynthetic rates of all saplings were measured between 9:30–10:30, 11:00–12:00, 13:00–14:00, and 15:00–16:00. Measurements were conducted twice in each time period. Five continuous measurements were carried out during each measurement and their mean values were used for the analysis. To account for shrubs and herbs in each treatment, three 4 m² (2 m × 2 m) sample plots were established to tally individuals at the species level. In addition, a 1 m² (1 m × 1 m) plot was established to estimate the percent cover of the ground area occupied by shrubs and herbs. In all the plots we evaluated the water content of decomposition and semi-decomposition. The decomposition layer was defined as having an intact profile of leaves, while the semi-decomposition layer had no intact profile if leaves and organic material was broken down and brown. In each layer around 0.5 kg fresh weight samples were collected and oven dried at the lab at a temperature of 85 °C for 24 h to reach a constant weight and the dry weight was measured.

Data analysis

The current year increment, branch length and number of branches of the first whorl for all sampled saplings were used to reflect the effect of clear-cut strip on the saplings growth status. The net photosynthesis rates, stomatal conductance, and transpiration rates of the leaves of each sapling were used to determine the photosynthetic characteristics of the saplings. The biological diversity and water content of the shrubs and herbs were used to characterize these understory plant communities. The biological diversity of the shrubs and herbs of each sample plot was calculated using the Shannon–Wiener index and Eq. (1).

(1) $$H_i=-\sum _{j=1}^{N{S}_i}(P_{ij}\ln (P_{ij}))$$

where H_i is Shannon–Wiener index, i = 1 is for the shrubs, and i = 2 is for herbs. NS_i is the number of the species for shrubs and herbs. P_ij = N_ij/N_i, N_ij is the number of the individuals of the jth species for shrubs and herbs in each sample plot, and N_i is the total number of individuals of all species for shrubs and herbs in each sample plot.

For all the three locations, one-way ANOVAs were performed to determine the influence of strip clear-cutting (15, 20, 25 m widths of clear-cut strip, 10, 18 m widths of uncut strip) and control on the growth status of saplings and the regeneration of shrubs and herbs. The effect of strip clear-cutting and time periods (9:30–10:30, 11:00–12:00, 13:00–14:00, and 15:00–16:00) on the net photosynthesis rates, stomatal conductance, and transpiration rates were analyzed with two-way ANOVAs. All analyses were performed using R software version 4.0.5 (R Core Team, 2021). HSD-Tukey was used to conduct the multiple comparisons.

Growth status of saplings

There was almost no seedling in the plots and the average count of saplings from each sapling were 11.2, 10.9, and 11.0 for location A, B and C, respectively. The density of saplings from the clear-cut strip was the largest, and the control was the smallest for location A, B and C (Table 1). The current year increment, branch length and number of branches of the first whorl of the sample saplings from clear-cut strips, uncut strips, and control were compared (Table 2). Normality and homogeneity of variances of the data were tested to meet the requirement of ANOVAs and HSD-Tukey. On the whole, current year increment shows the following trend: clear-cut strip > control > uncut strip for location A and B, and clear-cut strip > uncut strip > control for location C. Statistically significant differences were found for the current year increment of sapling height among the different widths of the strips and the control for all the three locations (F = 5.21, P < 0.05 for location A; F = 4.21, P < 0.05 for location B; F = 2.88, P < 0.05 for location C). The current year increment of saplings was largest in the 25 m wide clear-cut strips in all the three locations and was significantly larger than uncut strips and control (Table 2). No significant difference was found for the current year increment between the clear-cut strip with the width of 15, 20 m, uncut strip and the control except for location A. For branch length of the first whorl, significant differences among the strip types and control were confirmed for all the three locations (F = 6.20, P < 0.001 for location A; F = 11.17, P < 0.001 for location B; F = 11.57, P < 0.001 for location C), and the orders follow: clear-cut strip > control > uncut strip for all the three locations. In HSD-Tukey analysis, the branch length of clear-cut strip with width of 15 and 25 m were all significantly larger than uncut strips and control for location A, and the branch length of clear-cut strip with the width of 25 m was significant larger than the width of 15 and 20 m, uncut strip and control for location B and C. No statistical difference was also found between the uncut strip and control for all the three locations. Statistically significant differences were confirmed as for the number of branches for the first whorl among strip types for location A, B, and C (F = 12.02, P < 0.05 for location A; F = 17.46, P < 0.01 for location B; F = 16.85, P < 0.01 for location C). The number of the branches of the first whorl from clear-cut strips were significantly larger than the uncut strip and control, and follows the trend: clear-cut strip > uncut strip > control for all the three locations. No difference was found between the uncut strip and control for all the three locations.

Photosynthetic rate of saplings

For the large saplings from location A, strip type was a significant effect on the net photosynthesis rates (F = 90.90, P < 0.05) but the time period was not (F = 1.75, P = 0.16). In contrast, both of strip type (F = 16.37, P < 0.05; F = 13.53, P < 0.05) and time period (F = 20.31, P < 0.05; F = 3.81, P < 0.05) have significant effect on the stomatal conductance and transpiration rates. For the large saplings from location B, both of strip type and time period have significant effect on net photosynthesis rates (F = 73.14, P < 0.01 for strip type, F = 7.32, P < 0.01 for time period), stomatal conductance (F = 16.23, P < 0.01 for strip type, F = 20.88, P < 0.01 for time period) and transpiration rates (F = 16.80, P < 0.01 for strip type, F = 6.01, P < 0.01 for time period). For the large saplings from location C, both of strip type and time period have significant effect on net photosynthesis rates (F = 73.57, P < 0.01 for strip type, F = 7.48, P < 0.01 for time period), stomatal conductance (F = 12.34, P < 0.01 for strip type, F = 18.68, P < 0.01 for time period) and transpiration rates (F = 17.17, P < 0.01 for strip type, F = 5.48, P < 0.01 for time period).

As for the large sapling across all the strip types from location A, the largest values of net photosynthesis rates, stomatal conductance, and transpiration rates were 4.265 umol·m⁻²·s⁻¹ for the 25 m width, 0.0743 mol·m⁻²·s⁻¹ for the 15 m width, and 1.738 mmol·m⁻²·s⁻¹ for the 20 m widths of the clear-cut strips, respectively (Figs. 2A1–2A3). The control has the lowest stomatal conductance and transpiration rates but not the lowest net photosynthesis rates. As for the large saplings from location B and C, the net photosynthesis rates, stomatal conductance, and transpiration rates were also the largest for 25, 15, and 20 m width, respectively, and the control has the lowest stomatal conductance and transpiration rates (Figs. 2B1–2B3 and 2C1–2C3). Across four time periods (9:30–10:30, 11:00–12:00, 13:00–14:00, 14:30–15:30), net photosynthesis rates (F = 10.59, P < 0.05), stomatal conductance (F = 6.21, P < 0.05), and transpiration rates (F = 1.80, P < 0.05) for the large saplings from location A were significantly larger than uncut strips and control. The net photosynthesis rate was the largest in the clear-cut strips at 11:00–12:00 and that in the uncut strip with a width of 10 m and control was the largest at 13:00–14:00 (Figs. 3A1–3A3). As for the location B and C, the net photosynthesis rates of the clear-cut strip across four time periods were significantly larger than the uncut strip and control (Figs. 3B1 and 3C1). The stomatal conductance was the largest at the period of 14:30–15:30 from clear-cut strip with the width of 15 m for both of location B and C (Figs. 3B2 and 3C2), and the same to the transpiration rates (Figs. 3B3 and 3C3).

For the medium-sized sapling from location A, strip type showed statistically significant effect on the photosynthesis rates (F = 25.71, P < 0.05), stomatal conductance (F = 24.21, P < 0.05), and transpiration rates (F = 25.13, P < 0.05). In contrast, time period was statistically significant only for the stomatal conductance (F = 16.92, P < 0.05), and transpiration rates (F = 10.08, P < 0.05). Notably, the largest photosynthesis rates was from 25 m width of the clear-cut strip with 3.307 umol·m⁻²·s⁻¹, and the smallest values is from control with 0.499 umol·m⁻²·s⁻¹ (Fig. 2A1). The largest stomatal conductance and transpiration rates were from the 25 and 15 m width, respectively, for the clear-cut strip, and the control was the lowest. Notably, the stomatal conductance and transpiration rates of medium-sized saplings from the clear-cut strips is the largest, followed by the uncut strips and control (Figs. 2A2–2A3). Overall, net photosynthesis rates (F = 4.63, P < 0.05), stomatal conductance (F = 5.05, P < 0.05), and transpiration rates (F = 3.02, P < 0.05) for the medium-sized saplings from the clear-cut strips were the largest across the four time periods, and the control was the smallest for net photosynthesis rates and transpiration rates, but not stomatal conductance (Figs. 4A1–4A3). Overall, also for location B and C, the net photosynthesis rates, transpiration rates, and stomatal conductance for the clear-cut strip were larger than the uncut strip and control (Figs. 4B1–4B3, and 4C1–4C3).

As for the small sapling from location A, both of strip type and time period were confirmed to show statistically significant differences for net photosynthesis rates (F = 43.00, P < 0.05; F = 14.77, P < 0.05), stomatal conductance (F = 16.88, P < 0.05; F = 25.58, P < 0.05), and transpiration rates (F = 14.45, P < 0.05; F = 12.16, P < 0.05). Notably, the net photosynthesis rates for the clear-cut strip was still the largest. Different from the large and middle-sized sapling, the net photosynthesis rates for the uncut strip with 18 m was the smallest (Fig. 2A1). On the whole, the clear-cut strip has the largest stomatal conductance and transpiration rates and the control has the smallest values (Figs. 2A2–2A3). Across four time periods, net photosynthesis rates (F = 13.31, P < 0.05), stomatal conductance (F = 8.24, P < 0.05), and transpiration rates (F = 4.85, P < 0.05) for the small saplings were significantly larger than uncut strips and control. The net photosynthesis rates of small-sized saplings decreased from 9:30–10:30 to 13:00–14:00 and then increased in the clear-cut strips. In the uncut strips and control, the lowest values were determined at 11:00–12:00 (Fig. 5A1). The stomatal conductance increased, except for the clear-cut strip with a width of 25 m (Fig. 5A2). The maximum transpiration rate was reached at 11:00–12:00 in clear-cut strips and at 13:00–14:00 in uncut strips. The net photosynthesis rate, stomatal conductance, and transpiration rate of large, medium, and small saplings from the different types of strips exhibit the same order: clear-cut strips > uncut strips > controls, which is similar to the regularity to the location B and C (Figs. 5B1–5B3, and 5C1–5C3).

Diversity of herb and shrub and water content of decomposition layer

Across all the strip types for the three locations, a total of 21 understory herbs and 12 shrubs were observed. Height and coverage for shrubs and herbs were showed in Table 3. The biological diversity (Shannon–Wiener index) of shrubs and herbs from clear-cut strips, uncut strips, and the control was calculated for three locations. No statistically significant effect of the strip type on the biodiversity of herbs and shrubs was observed (F = 2.16, P > 0.05 for herbs, F = 0.49, P > 0.05 for shrubs for location A; F = 2.08, P > 0.05 for herbs, F = 1.29, P > 0.05 for shrubs for location B; F = 2.44, P > 0.05 for herbs, F = 1.33, P > 0.05 for shrubs for location C). As for location A, the biological diversity of herbs in treatments with widths of 15, 20, and 25 m is 1.723, 1.180, and 0.928, respectively, and that in uncut strips with widths of 10 and 18 m and the control is 1.518 and 0.843 and 1.372, respectively. The biological diversity of shrubs in clear-cut strips with widths of 15, 20, and 25 m is 0.868, 0.898, and 0.693, respectively, and that in uncut strips with widths of 10 and 18 m and control is 0.980 and 0.915 and 0.786, respectively. The largest biological diversity of herbs with 1.655 and shrubs with 1.811 were from clear-cut strips with widths of 15 from location A and B, and 1.685 and 1.711 were also from clear-cut strips with widths of 15 from location B, respectively. Across the clear-cut strip, the control has the largest water content for both of decomposition and semi-decomposition layer, and the clear-cut strip was the smallest for decomposition and semi-decomposition layer (Table 4). However, the strip type was not a statistically significant treatment for the water content of decomposition (F = 1.15, P = 0.39) and semi-decomposition layer (F = 1.08, P = 0.42) for location A. On the whole, there was no significant difference of the water content for decomposition layer from three locations (Table 4), and the water content for the semi-decomposition layer from the clear-cut strip was significantly larger than the control and uncut strip for all the three locations (Table 4).