Spatial pattern of forest aboveground biomass and its environmental influencing factors in Qinling-Daba Mountains, central China

Abstract

Accurate estimation of forest aboveground biomass (AGB) is crucial for understanding and managing forest ecosystems in the context of global environmental change, and also provides a scientific basis for national and regional ecological planning and carbon emission reduction policies. In order to investigate the regional pattern of forest AGB and its influencing factors in central China, a total of 469 sample plots were measured along four sample transects and on six mountains in field survey. The results showed that: 1) Two longitudinal distribution patterns of forest AGB were found, and one was that the AGB in the Qinling Mountains and the Daba Moutains gradually decreased from east to west, and the other was that the AGB in the areas between the two mountains gradually increased from east to west. The latitudinal distribution pattern of the forest AGB showed an "increasing–decreasing-increasing–decreasing" trend from south to north. The altitudinal distribution pattern showed a "first increasing-then decreasing" pattern with increasing altitude. 2) The influence of each factor on the spatial pattern of forest AGB was manifested as temperature > precipitation > HAI > topography, indicating that the spatial pattern of forest AGB in central China was the result of the interaction of climate, human activities and natural factors.

Introduction

Forests, which cover more than 30% of the Earth's surface, store 45% of terrestrial carbon and contribute 50% of terrestrial net primary production^1–3, mitigating the impacts of climate change in the process^4,5. Forests sequester carbon through photosynthesis and store it primarily as aboveground living biomass of trees (AGB), which accounts for between 70 and 90% of total forest biomass^6–9, and about 50% of AGB is carbon^10,11. Understanding and quantifying forest AGB is essential for assessing and managing global carbon sinks, and for developing effective forest management and conservation strategies.

In recent years, with the increase of global warming and forest degradation, increasing attention has been paid to estimating forest AGB with sufficient resolution at larger spatial scales^12–14. Overall, there are three types of means for AGB estimation. One is field measurement, which estimates forest AGB by the parameters such as diameter at breast height (DBH) and height of trees in a specific forest area, combined with the biomass Eqation^15–19. Although, this method is highly accurate, it is time-consuming, labor-intensive, and difficult to apply to forests over large areas. The other method is to use satellite remote sensing data (e.g., LIDAR, RADAR) in combination with ground measurements to estimate AGB over a large area^20–22. In the last decades, many AGB maps based on remote sensing technology have been produced at continental and national scales, or for some specific areas with resolutions ranging from 30 m to 1km^23–30. This method is highly efficient, but is limited by the resolution and accuracy of remote sensing technology. Another method is to estimate the AGB by ecological modeling combined with climate data, soil data, vegetation types, etc.^31–33. This method can simulate the change of AGB under future climate change scenarios.

Over the past decade, research utilizing field measurements to study aboveground biomass (AGB) has made significant contributions to our understanding of AGB, particularly in the context of environmental factors such as drought. For example, Kröel-Dulay et al.³⁴ reviewed the discrepancies between experimental and observational studies in estimating AGB responses to drought, suggesting that experimental setups may significantly underestimate the impacts of drought due to factors such as the small scale of experiments and unaccounted for environmental conditions, ​and applying new methods for more accurate biomass estimation. Some studies have focused on improving the precision of field estimations of AGB by constructing tree-level horizontal biomass distributions (HBDs) for trees both within and extending into the plot perimeter. For instance, Kleinn et al.³⁵ proposed a more refined approach by combining stem biomass models with branch and leaf biomass models to provide a more realistic distribution of biomass across a tree's crown projection area. Research has also been conducted to estimate AGB in planted forests, taking into account stand age and environmental variables³⁶. These studies demonstrate the dynamic and evolving nature of AGB and highlight the ongoing efforts to improve the accuracy and applicability of AGB estimation methods, which is critical for understanding and managing forest ecosystems in the context of global environmental change.

Although the field measurement method has been widely used to estimate AGB thanks to its directness and accuracy, it is time-consuming, labor-intensive, and high-cost, and has usually been applied at local scales or small areas and limited to one or a few species, seldom on regional or national scales. Therefore, these studies are somewhat inadequate in reflecting the spatial differentiation pattern of forest AGB. In order to investigate the regional pattern and status of forest AGB, and to provide support for regional forest management and protection policies, this study analyzed the spatial distribution pattern of forest AGB in the Qinling-Daba Mountains of the central China through 469 sample plots obtained from field surveys. The objectives of this study were as follows: (1) To dicover the current status of forest AGB in the study area under the forest conservation policy. Since the late 1970s, China has initiated ambitious programs to conserve and expand forests with the goal of mitigating land degradation, air pollution and climate change, covering approximately 44.8% of the national forests and 23.2% of the national grasslands, including the Qinling-Daba Mountains in central China^37,38. (2) To understand the spatial pattern of forest AGB and its environmental influencing factors at the regional scale.

Materials and methods

Study area

The Qinling-Daba Mountains (102°–114° E, 30°–36° N) is an important geographical-ecological transition zone in central China (Fig. 1), developing diverse and unique forest ecosystems, whose forest ecosystems and carbon sequestration potentials are important for the regional ecological balance and the specific design of the national dual-carbon policy. It stretches from Diebu in Gansu to Shennongjia in Hubei, with an east–west length of about 1,000 km and a north–south width of about 240 to 450 km, covering an area of about 300,000 km² and crossing six provinces: Gansu, Shaanxi, Sichuan, Chongqing, Hubei, and Henan. The Daba Mountains are situated in the south with subtropical climate and the Qinling Mountains are located in the north with temperate climate, the Han River Valley is located between of the two mountains. The terrain presents a pattern of high in the west and low in the east, high in the north and south and low in the middle: the elevation of the eastern part and the Han River valley is usually lower than 1000 m, the elevation of the western part is even higher than 3000 m, most of the other places are between 1000 and 2000 m. The climate is warm and humid, and gradually changes from a subtropical climate in the south to a temperate climate in the north³⁹. The mean annual temperature in this area is between 12 °C and 16 °C, and the mean annual precipitation is about 400–1500 mm^40,41. Forests cover 55% of the Qinling-Daba Mountains, of which 21%, 13%, and 66% are broadleaved, coniferous, and mixed forests, respectively⁴². In addition to field surveys along four sampling transects throughout the Qinling-Daba Mountains, field surveys were also conducted in three subtropical mountains (Mt. Shennongjia, Mt. Micang and Mt. Xionghuang) and three temperate mountains (Mt. Baotianman, Mt. Taibai and Mt. Jifeng), all of which are nature reserves.

Fig. 1.

Location of the study area, sampling transects, and distribution of sample plots (created in ArcGIS10.8, the base map was obtained from the Ministry of Natural Resources, China and the forest data was downloaded from http://data.starcloud.pcl.ac.cn/zh/resource/15).

Field measurement and sample plots

Four sampling transects (three north–south sampling transects and one east–west sampling transect) and six mountains (Mt. Baotianman, Mt. Shennongjia, Mt. Micang Mt. Xionghuang, Mt.Jifeng and Mt. Taibai) covering subtropical and temperate climate zones, and important nature reserves, vegetation distribution and change areas in central China were designed to collect samples in the field survey (Fig. 1). From 2017 to 2022, a total of 318 sample plots were measured along the four sampling transects and the five typical mountains. The other 151 samples (covering Mt.Taibai and Foping) were obtained from the Shaanxi Mt. Taibai National Nature Reserve Comprehensive Expedition Project. Square plots of a 20 m × 20 m were spaced at 5 km intervals along the north–south(S–N) transects and 10 km intervals along the east–west (E–W) transect, 100 m intervals along the flanks of the mountains. Tree height, diameter at breast height (DBH), and crown spread of individual trees with a DBH greater than 5 cm in the sample plots were measured and recorded. There were a total of 267 broadleaved forest plots, 104 coniferous forest plots, and 98 mixed coniferous and broadleaved forest plots (Fig. 1).

Other materials

Temperature, precipitation, topography and human activity intensity (HAI) data were used in this study to analyze the influencing factors of forest AGB. The mean annual temperature and mean annual precipitation data were downloaded from the Resource and Environment Science Data Center of the Chinese Academy of Sciences (http://www.resdc.cn), which were generated based on daily observations from more than 2,400 meteorological stations across the country with a spatial resolution of 500 m for the period of 1980–2015. The Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTER GDEM) topographic data was downloaded from https://earthdata.nasa.gov/ with a spatial resolution of 1 arc second (~ 30 m in the equatorial region). The HAI data with a 100 m resolution was used to reflect the impact of human activities on forest AGB, which was generated using principal component analysis (PCA) by selecting population density, degree of land transformation, degree of remoteness from highway/railway, degree of remoteness from settlements, highway density, topography, and river density as indicators⁴³.

Methods

Biomass equations in China are mainly constructed based on single or independent samples or special sites, with limited samples, and the accuracy of the established biomass equations may be reduced when they are used to estimate the biomass outside the study sites or different species and different diameter levels¹⁶. In addition, most of the equations are specific to a particular site or forest type and lack consistent methods and standards, which are not suitable for estimating forest biomass at the regional level. Since the sample sites in this study were located in six provinces (Shaanxi, Henan, Hubei, Sichuan, Chongqing and Gansu) and the samples included different forest types with different tree species, as well as the biomass equations for all species of the samples could not be obtained at present, the biomass equations for mixed species (groups) in the provinces constructed by Zhou et al.¹⁶ (Table 1) were adopted to calculate the forest AGB with DBH (the sum of stem, branch and leaf) as the independent variable, and the AGB per square meter was obtained by diving the area of the sample site (400 m²) and then was standardized as t/hm². These biomass equations (Table 1) were constructed by assigning values to the collected published biomass equations based on 7800 typical forest sample plots on field measurement across China, and were optimized by the least squares support vector machine (LSSVM)¹⁶. The estimation error was less than 10%, which was verified by the measurement data of the standard timber.

Table 1.

Mixed species (group) biomass equation by provinces.

Province	Forest type	Biomass equation	R2	Extent of DBH(cm)
Shaanxi	Broadleaved forest	${\text{B}}_{\text{stem}}=0.1449\times {\text{DBH}}^{2\text{.0845}}$	0.96	5 ~ 70
		${\text{B}}_{\text{branch}}=0.134\times {\text{DBH}}^{2\text{.4995}}$	0.74
		${\text{B}}_{\text{leaf}}=0.0106\times {\text{DBH}}^{2\text{.2040}}$	0.87
	Coniferous forest	${\text{B}}_{\text{stem}}=0.0556\times {\text{DBH}}^{2\text{.4201}}$	0.98	5 ~ 60
		${\text{B}}_{\text{branch}}=0.0446\times {\text{DBH}}^{1\text{.9545}}$	0.95
		${\text{B}}_{\text{leaf}}=0.286\times {\text{DBH}}^{2\text{.221}}$	0.97
	Mixed forest	${\text{B}}_{\text{stem}}=0.627\times {\text{DBH}}^{2\text{.3707}}$	0.97	5 ~ 70
		${\text{B}}_{\text{branch}}=0.0270\times {\text{DBH}}^{2\text{.268}}$	0.91
		${\text{B}}_{\text{leaf}}=0.0221\times {\text{DBH}}^{2\text{.1633}}$	0.87
Henan	Broadleaved forest	${\text{B}}_{\text{stem}}=0.1449\times {\text{DBH}}^{2\text{.0845}}$	0.96	5 ~ 70
		${\text{B}}_{\text{branch}}=0.0134\times {\text{DBH}}^{2\text{.4995}}$	0.74
		${\text{B}}_{\text{leaf}}=0.0106\times {\text{DBH}}^{2\text{.2040}}$	0.87
	Coniferous forest	${\text{B}}_{\text{stem}}=0.0283\times {\text{DBH}}^{2\text{.6268}}$	0.95	5 ~ 80
		${\text{B}}_{\text{branch}}=0.0227\times {\text{DBH}}^{2\text{.1989}}$	0.84
		${\text{B}}_{\text{leaf}}=0.0741\times {\text{DBH}}^{1\text{.5627}}$	0.77
	Mixed forest	${\text{B}}_{\text{stem}}=0.0991\times {\text{DBH}}^{2\text{.2757}}$	0.92	5 ~ 95
		${\text{B}}_{\text{branch}}=0.0214\times {\text{DBH}}^{2\text{.2363}}$	0.90
		${\text{B}}_{\text{leaf}}=0.0291\times {\text{DBH}}^{1\text{.8656}}$	0.87
Hubei	Broadleaved forest	${\text{B}}_{\text{stem}}=0.1120\times {\text{DBH}}^{2\text{.3230}}$	0.98	5 ~ 81
		${\text{B}}_{\text{branch}}=0.0270\times {\text{DBH}}^{2\text{.4080}}$	0.94
		${\text{B}}_{\text{leaf}}=0.0200\times {\text{DBH}}^{2\text{.0470}}$	0.71
	Coniferous forest	${\text{B}}_{\text{stem}}=0.1970\times {\text{DBH}}^{1.929}$	0.97	5 ~ 66
		${\text{B}}_{\text{branch}}=0.280\times {\text{DBH}}^{2\text{.5210}}$	0.94
		${\text{B}}_{\text{leaf}}=0.0850\times {\text{DBH}}^{1\text{.8440}}$	0.73
	Mixed forest	${\text{B}}_{\text{stem}}=0.1220\times {\text{DBH}}^{2\text{.2630}}$	0.96	5 ~ 81
		${\text{B}}_{\text{branch}}=0.0270\times {\text{DBH}}^{2\text{.4280}}$	0.94
		${\text{B}}_{\text{leaf}}=0.0240\times {\text{DBH}}^{2\text{.0290}}$	0.67
Sichuan	Broadleaved forest	${\text{B}}_{\text{stem}}=0.3780\times {\text{DBH}}^{1\text{.9170}}$	0.92	5 ~ 56
		${\text{B}}_{\text{branch}}=0.0520\times {\text{DBH}}^{2\text{.3680}}$	0.90
		${\text{B}}_{\text{leaf}}=0.0530\times {\text{DBH}}^{1\text{.7390}}$	0.96
	Coniferous forest	${\text{B}}_{\text{stem}}=0.5324\times {\text{DBH}}^{1\text{.7381}}$	0.90	5 ~ 76
		${\text{B}}_{\text{branch}}=0.0181\times {\text{DBH}}^{2\text{.3560}}$	0.97
		${\text{B}}_{\text{leaf}}=0.0473\times {\text{DBH}}^{1\text{.8867}}$	0.90
	Mixed forest	${\text{B}}_{\text{stem}}=0.5157\times {\text{DBH}}^{1\text{.7450}}$	0.90	5 ~ 76
		${\text{B}}_{\text{branch}}=0.0764\times {\text{DBH}}^{2\text{.0595}}$	0.76
		${\text{B}}_{\text{leaf}}=0.0426\times {\text{DBH}}^{1\text{.8967}}$	0.91
Gansu	Broadleaved forest	${\text{B}}_{\text{stem}}=0.0527\times {\text{DBH}}^{2\text{.4912}}$	0.99	5 ~ 80
		${\text{B}}_{\text{branch}}=0.0036\times {\text{DBH}}^{3\text{.0278}}$	0.89
		${\text{B}}_{\text{leaf}}=0.0060\times {\text{DBH}}^{2\text{.1451}}$	0.67
	Coniferous forest	${\text{B}}_{\text{stem}}=0.500\times {\text{DBH}}^{2\text{.4043}}$	0.99	5 ~ 125
		${\text{B}}_{\text{branch}}=0.0154\times {\text{DBH}}^{2\text{.5052}}$	0.97
		${\text{B}}_{\text{leaf}}=0.0275\times {\text{DBH}}^{2\text{.0100}}$	0.87
	Mixed forest	${\text{B}}_{\text{stem}}=0.528\times {\text{DBH}}^{2\text{.4011}}$	0.97	5 ~ 125
		${\text{B}}_{\text{branch}}=0.0107\times {\text{DBH}}^{2\text{.6348}}$	0.92
		${\text{B}}_{\text{leaf}}=0.0162\times {\text{DBH}}^{2\text{.1444}}$	0.78

Statistical analysis and the box-and-whisker plots (analyzed in RStudio 4.3.2) were used to characterize the spatial pattern of forest AGB along the longitudinal, latitudinal and altitudinal directions based on the above AGB estimation.

Correlation analysis and Geodetector analysis were used to analyze the effects of human activities, temperature, precipitation and topography on forest AGB. The Geodetector has been widely used to detect the driving factors in many studies such as land use⁴⁴, public health⁴⁵, meteorology and environment⁴⁶, and vegetation cover change^47,48. Q-statistic in Geodetector can be used to measure spatial stratified heterogeneity, detect explanatory factors and analyze the interactive relationship between variables⁴⁹. The range of the Q-statistic is [0, 1], and a larger Q-statistic indicates that the independent variable has the stronger explanatory power on the dependent variable⁴⁹. The P-statistic corresponding to the Q-statistic of the independent variable can be used to represent the significance of the independent variable on the dependent variable, for example, a P-statistic of less than 0.05 or 0.01 means that the effect of the independent variable on the dependent variable is significant or highly significant. Since the Geodetector requires the input data to be category data, the mean annual temperature and mean annual precipitation were classified into 10 categories using the natural breakpoint method, and the DEM was divided into 10 categories according to < 500, 500–800, 800–1000, 1000–1500, 1500–2000, 2000–2500, 2500–3000, 3000–3500, and > 3500. HAI was divided into six classes as < 0.3, 0.3–0.4, 0.4–0.6, 0.6–0.7, 0.7–0.8, 0.8–1.0.

Results

Spatial pattern of forest AGB in the Qinling-Daba mountains

Overall, the mean AGB of forest was 133.34 t/hm², which was higher than that of the subtropical region of China (129.0 t/hm²) and that of the national forest (65.3 t/hm²)⁵⁰. The forest AGB in the eastern part of the study area was lower than that in the central and western parts of the study area (Table 2).

Table 2.

Statistical results of forest aboveground biomass based on survey samples.

Transects or sites	Number of sample plots	Extent of AGB (t/hm2)	Average AGB (t/hm2)	Median AGB (t/hm2)	Standard deviation of AGB (t/hm2)
Eastern S–N transect	65	8.01–301.91	82.03	54.99	62.96
Central S–N transect	48	18.05–483.24	142.59	133.93	74.23
Western S–N transect	44	16.86–275.26	112.50	108.68	65.83
E-W transect	68	9.60–334.52	102.24	84.78	67.16
Foping-Changqing	49	1.73–366.80	126.26	88.43	98.94
Mt. Taibai	102	29.03–491.31	142.16	108.95	105.84
Mt. Micang	19	103.26–346.95	229.81	222.51	77.60
Mt. Baotianman	23	59.86–279.43	151.08	135.34	62.58
Mt. Shengnongjia	38	54.98–479.21	202.98	186.21	103.38
Mt. Xionghuang	13	42.95–325.29	134.31	107.84	80.21

The longitudinal pattern of forest AGB along the E-W transect (along Longnan-Hanzhong-Shiyan in the central of the study area) gradually decreased from west to east, while there was a slight increase between 105°E-107°E, reaching a high value in Hanzhong Crested Ibis Reserve and then decreasing sharply (Fig. 2a).

Fig. 2.

Forest aboveground biomass along four sampling transects in the Qinling-Daba Mountains (a) Forest aboveground biomass along the E-W transect; (b) Forest aboveground biomass along the eastern S–N transect; (c) Forest aboveground biomass along the central S–N transect; (d) Forest aboveground biomass along the western S–N transect.

The latitudinal pattern of forest AGB showed an "increasing–decreasing-increasing–decreasing" trend from south to north (Figs. 2b–d). Along the eastern S–N transect, the forest AGB increased slowly from the south to the north, reaching the highest value at Mt. Shennongjia (south flank, > 200 t/hm²), then decreased rapidly, with a gentle transition from 32.5°N to 33.0°N, and then increased slightly at Mt. Furniu (Fig. 2b), then decreased gradually from Mt. Furniu to the north. Along the central S–N transect, forest AGB also increased from the south to the north, reached a high value on the south flank of the Daba Mountains, then decreased and reached a low value in the Han River valley, and then gradually increased on the south flank of the Qinling Mountains (Fig. 2c), reaching the highest AGB (491.31 t/hm²) in the Quercus serrata forest, then gradually decreased from the Qinling Mountains to the north. Along the western S–N transect, forest AGB slowly increased from the south to the north, reached a high value (> 200 t/hm²) near the Jianshan Nature Reserve, slowly decreased from 32.7°N to 33.5°N, then increased to a secondary peak at about 34°N, and then gradually decreased on the north flank of the Qinling Mountains (Fig. 2d).

The altitudinal pattern of forest AGB showed a "first increasing and then decreasing" trend with increasing altitude (Fig. 3): the highest AGB on the southern flank of Mt. Shennongjia was found at about 1500–2000 m (Fig. 3a), and that on the northern flank was found at about 1600–2000 m (Fig. 3b); the highest AGB on the southern flank of Mt. Baotianman was found at about 1100–1600 m (Fig. 3c), and that on the northern flank was found at about 1400–1600 m (Fig. 3d); the highest AGB on the southern flank of Mt. Yicang was found at about 1700–2200 m (Fig. 3e), and that on the northern flank of Mt. Yicang was found at about 1500–2000 m (Fig. 3f); the highest AGB on the southern flank of Mt. Taibai was found at about 2200–3100 m (Fig. 3g), and that on the northern flank of Mt. Taibai was found at about 1400–2600 m (Fig. 3h).

Fig. 3.

Changes of forest aboveground biomass with altitude (a) Southern flank of Mt. Shennongjia; (b) Northern flank of Mt. Shennongjia; (c) Southern flank of Mt. Baotianman; (d) Northern flank of Mt. Baotianman; (e) Southern flank of Mt. Micang; (f) Northern flank of Mt. Micang; (g) Southern flank of Mt. Taibai; (h) Northern flank of Mt. Taibai.

AGB of different forest types in the Qinling-Daba Mountains

On the eastern S–N transect, the AGB extent of broadleaved forest (mainly Quercus variabilis forest, Form.Robiniapseudoacacia forest) ranged from 23.06 to 301.91 t/hm²; that of coniferous forest (mainly Platycladus orientalis (L.) Franco forest and Pinus tabuliformis forest) ranged from 7.20 to 210.98 t/hm² ; that of mixed forest (mainly Quercus variabilis & Pinus tabuliformis forest, Quercus variabilis & Pinus massoniana forest, and Quercus aliena var. acutiserrata & Pinus armandii forest) ranged from 26.45 to 270.11 t/hm² (Fig. 4a). According to the mean and median of AGB, AGB of coniferous forest > AGB of mixed forest > AGB of broadleaved forest (Fig. 4a).

Fig. 4.

Boxplots of forest aboveground biomass along the transects (a) The eastern S–N transect; (b) The central S–N transect; (c) The western S–N transect; (d) The E-W transect.

On the central S–N transect, the AGB extent of broadleaved forest ranged from 18.05 to 248.64 t/hm², mainly including Quercus variabilis forest, Toxicodendron vernicifluum forest and Acer cappadocicum forest; that of coniferous forest ranged from 38.79 to 184.01 t/hm², mainly including Pinus massoniana forest and Cunninghamia lanceolata forest; that of mixed forest ranged from 99.99 to 483.24 t/hm², mainly including Quercus variabilis & Pinus massoniana forest, and Quercus aliena var. acutiserrata & Pinus armandii forest. The mean of AGB was mixed forest > broadleaved forest > coniferous forest, while the median of AGB was mixed forest > coniferous forest > broadleaved forest (Fig. 4b).

On the western S–N transect, the AGB extent of broadleaved forest ranged from 19.93 to 275.26 t/hm², mainly including Form.Robiniapseudoacacia forest; that of coniferous forest ranged from 29.62 to 185.51 t/hm², mainly consisting of Pinus tabuliformis forest and Platycladus orientalis (L.) Franco forest; that of mixed forest ranged from 38.06 to 204.37 t/hm², mainly consisting of Pinus tabuliformis or Platycladus orientalis (L.) Franco & other delicious broadleaved forest. According to the mean and median of AGB, AGB of coniferous forest > AGB of mixed forest > AGB of broadleaved forest (Fig. 4c).

On the E-W transect, the AGB extent of broadleaved forest (mainly including Quercus acutissima forest and Robinia pseudoacacia L. forest) ranged from 9.60 to 334.52 t/hm²; that of coniferous forest (mainly including Pinus tabuliformis forest and Pinus massoniana forest) ranged from 22.97 to 288.37 t/hm²; that of mixed forest (mainly Pinus massoniana & Quercus acutissima or Quercus variabilis forest) ranged from 29.80 to 264.76 t/hm². According to the mean and median of AGB, AGB of coniferous forest > AGB of mixed forest > AGB of broadleaved forest (Fig. 4d).

The above results showed that the AGB of coniferous forest or mixed forest was higher than that of broadleaved forest in the Qinling-Daba Mountains.

Forest AGB of typical mountains and nature reserves

In the southern subtropical region, the AGB of broadleaved forest in Mt.Shennongjia ranged from 54.98 to 479.21 t/hm², that of coniferous forest ranged from 129.55 to 408.38 t/hm², and that of mixed forest ranged from 75.55 to 197.37 t/hm², the mean and median of AGB were coniferous forest > broadleaved forest > mixed forests (Fig. 5a). The AGB of broadleaved forest in Mt.Micang (dominating by broadleaved forest) ranged from 103.26 to 336.95 t/hm², with the maximum AGB occurring in the Betula albosinensis forest at 1506 m (Fig. 5b). The AGB of broadleaved forest in Mt.Xionghuang ranged from 42.95 to107.84 t/hm², that of coniferous forests was 214.57 to 305.29 t/hm², that of mixed forest was 72.2 to 205.68 t/hm², the mean and median of AGB were coniferous forest > mixed forest > broadleaved forest (Fig. 5c). Forest AGB in the southern subtropical mountains showed a gradual downward trend from east to west (Fig. 6).

Fig. 5.

Box-and-whishker plots of forest aboveground biomass of different forest types on different mountains (a) Mt. Shennongjia; (b) Mt. Micang; (c) Mt. Xionghuang; (d) Mt. Baotianman; (e) Mt. Taibai; (f) Mt. Jifeng.

Fig. 6.

Box-and-whishker plots of forest aboveground biomass on different mountains (a) Mt. Shennongjia; (b) Mt. Micang; (c) Mt. Xionghuang; (d) Mt. Baotianman; (e) Mt. Taibai; (f) Mt. Jifeng.

In the northern temperate region, the AGB of broadleaved forest in Mt. Baotianman ranged from 59.86 to 279.43 t/hm², that of mixed forest was 77.42 to 231.82 t/hm², the mean and median of AGB were mixed forest > broadleaved forest (Fig. 5d). The AGB of broadleaved forest in Mt. Taibai ranged from 28.94 to 491.31 t/hm², that of coniferous forest was 31.91–368.70 t/hm², that of mixed forest was 40.74 to 376.07 t/hm², the mean and median of AGB were mixed forest > coniferous forest > broadleaved forest (Fig. 5e). The AGB of broadleaved forest in Mt. Jifeng ranged from 21.61 to108.58 t/hm², that of coniferous forest was 67.90 to170.60 t/hm², the mean and median of AGB were coniferous forest > broadleaved forest(Fig. 5f). Forest AGB in the northern temperate mountains also showed a gradual downward trend from the east to the west (Fig. 6).

The above results showed that: 1) the AGB of coniferous forest or mixed forest was higher than that of broadleaved forest in typical mountains and nature reserves (Fig. 5); 2) the forest AGB in subtropical mountains was higher than that in temperate mountains (Fig. 6); and 3) the forest AGB not only in the southern subtropical mountains, but also in the northern temperate mountains, showed a gradual downward trend from the east to the west (Fig. 5 and Fig. 6).

There is an inconsistency in the forest AGB trend from the east to the west between typical mountains (gradual decrease) and the E-W transect (gradual increase). The reason for the inconsistency is that: all of these six mountains are nature reserves with relatively little interference from human activities, and the AGB is mainly determined by habitat conditions, so it decreases accordingly with the gradual change of the water and heat conditions from the east to the west. While the E-W transect passes through the Han River valley with greater interference from human activities, and the intensity of human activities in the east is much stronger than that in the west, therefore, most of the forests are secondary forests, and their AGB are lower in the east.

That is to say, there are two longitudinal patterns of forest AGB in the Qinling-Daba Mountains: one was that the AGB in the Qinling Mountains and in the Daba Mountains gradually decreased from the east to the west (Fig. 6), and the other was that the AGB in the areas between the Qinling mountains and Daba Mountains (Han River valley, along the Longnan-Hanzhong-Shiyan line) gradually increased from the east to the west (Fig. 2a).

Influencing factors of forest AGB in the Qinling-Daba Mountains

The result of correlation analysis indicated that temperature, precipitation, topography and HAI have profound effect on forest AGB: temperature and precipitation were significantly correlated with AGB at 0.05 significant level, and topography and HAI were significantly correlated with AGB at 0.01 significant level (Table 3). The Geodetector results showed that the explanatory power of the four factors was as follows: temperature > precipitation > HAI > topography (Table 3). Climatic factors (temperature and precipitation) play an important role in plant growth, development and spatial distribution and are therefore the most important factors affecting forest AGB, and topography affects forest AGB by regulating local redistribution of water and heat and by influencing soil nutrient composition. It is worth noting that the distribution pattern of forest AGB in the study area is also strongly influenced by human activities. For example, in the eastern part of the study area and the Han River valley, the lower elevation, higher population and stronger agriculture and industry resulted in lower forest AGB. Broadleaved forests are usually located in the low elevation areas, which are easily affected by human activities such as agriculture, industry, or urbanization, and thus the AGB of broadleaved forest is the lowest among the three major forest types. The AGB of coniferous forest or mixed forest, on the other hand, is somewhat higher in high-altitude mountainous areas or nature reserves because there is less disturbance from human activities.

Table 3.

Correlation and Geodetector results of environmental factors influencing the forest aboveground biomass.

	Temperature	Precipitation	Topography	HAI
Q-statistics(explanatory power)	0.156	0.1106	0.079	0.101
P-statistics(significance)	0.000	0.000	0.000	0.000
Pearson Correlation	− 0.098*	− 0.116*	0.135**	− 0.213**
Significance (two-tailed)	0.034	0.012	0.003	0.000

**Correlations are significant at the 0.01 level (two-tailed).

*Significant at the 0.05 level (two-tailed).

Discussion

Factors influencing forest AGB and its spatial pattern

Many studies have found that climate, species diversity, community-level functional traits and stand structures have important effects on forest AGB^8,51–53, but the effects of human activities and the physical environment on AGB are rarely considered. In the field survey, it was found that besides the species diversity and community characteristics of the forest itself and the climate, human activities and topography in the Qinling-Daba Mountains also play important roles in the spatial pattern of forest AGB. In this study, the correlations between HAI / topography and AGB were even higher than those of climate factors, confirming the influence of human activities and topography on forest AGB. Among the four selected influencing factors, because climate (temperature and precipitation) is closely associated with forest AGB^8,54–56 and plays a critical role in tree growth and development, the explanatory power of climatic factors (mean annual temperature and mean annual precipitation) is higher than that of HAI and topography factors. Human activities in the Qinling-Daba Mountains have a great impact on vegetation^41,43, and anthropogenic disturbance indirectly affects the forest AGB by influencing species density, community-level functional traits, and stand structure. While topography affects forest AGB by regulating local water and heat redistribution, soil nutrients and moisture, and constraining human activities. That is why the explanatory power of the four factors on AGB spatial pattern was temperature > precipitation > HAI > topography (Table 3). It is important to note that none of the four factors mentioned above have a high explanatory power for forest AGB (Table 3), as they are all indirect factors influencing forest AGB. Future work will further consider the impacts of the species diversity, community characteristics and stand structure of the forest itself on the AGB.

Impact of human activities on AGB of different forest types

The distribution of forests in the Qinling-Daba Mountains is characterized by an obvious altitudinal zonation (mountain altitudinal belts)^57,58. Broadleaved, mixed coniferous and broadleaved, and coniferous forests are sequentially distributed from low to high altitudes. Due to constraining by topography, human activities in the Qinling-Dabashan Mountains are mainly concentrated in the middle and low elevation areas^41,43, where are mainly covered by broadleaved forests or croplands, so the AGB of broadleaved forest is more strongly affected by human activities than that of coniferous forests. In a word, the impacts of human activities on AGB of different forest types are different. While the influencing factors for different forest types were not discussed in this study and can be investigated in-depth in future work. Additionally, human activities have both negative (e.g., lgging) and positive (e.g., forest restoration) effects, which also need to be carefully analyzed in future studies.

Conclusion

As shown in the above analysis and results, forest AGB was low in the eastern and central Qinling-Daba Mountains with low elevation and high in mountainous areas with high elevation and in nature reserves. The spatial pattern of forest AGB was as follows: 1) there were two longitudinal spatial patterns, and one was that the AGB in the Qinling Mountains and the Daba Mountains gradually decreased from east to west, and the other was that the AGB in the areas between the two mountains (the central part of the study area, the Han River valley, and along the Longnan-Hanzhong-Shiyan line) gradually increased from east to west; 2) the latitudinal distribution pattern of the forest AGB showed an "increasing–decreasing-increasing–decreasing" trend from south to north; 3) the altitudinal distribution pattern showed a "first rising-then falling" pattern with increasing altitude. This spatial pattern of forest AGB was attributed to the interaction of climate, human activities and topography of the study area. In addition to climate factors such as the mean annual temperature and mean annual precipitation that play important roles in plant growth, development and distribution, the influence of human activities and topography on the forest AGB in the central China should be paid more attention.

The results of the field survey and the analysis in this study also showed that the forest AGB in the southern subtropical mountains of the study area was higher than that in the northern temperate mountains, and the AGB of coniferous forest and mixed forest was higher than that of broadleaved forest in the study area. The current status of forest AGB in the Qinling-Daba Mountains is good (the average AGB is 133.34 t/hm²), and great achievements have been made in forest management and ecological restoration. The Qinling-Daba Mountains in central China is not only an ecological functional area for biodiversity conservation, but also a water-source conservation area for the "South to North Water Transfer Project", and one of the poorest regions in China⁵⁹. It is hard work to maintain the balance between poverty alleviation and ecological restoration. Since the late 1970s, China has initiated ambitious programs to conserve and expand forests with the goal of mitigating land degradation, air pollution and climate change^37,38. More than 30 nature reserves, about 37 national forest parks, 11 national geological parks, and many other scenic spots have been established in the study area from the 1960s to the present⁶⁰. Meanwhile, the "Grain for Green" project has also been made great progress in this area⁴¹. Moreover, the precision poverty alleviation projects implemented by China in this region have also released the human-land conflict^61,62. All of these ecological projects and poverty alleviation projects contribute to the good condition of the forest.

The results of this study can support the research on forest ecosystems and their carbon storage capacity, climate change in central China, and also provide scientific basis for national and regional ecological planning and carbon emission reduction policies.

Author contributions

Yonghui Yao: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Data curation, Writing- Original draft, Writing- Reviewing and Editing, and Visualization.

Funding

This work was funded by the Strategic Priority Research Program of the Chinese Academy of Sciences [Grant No.XDB0740200]; the National Natural Science Foundation of China (Grant number: 41871350); and the Key Project of Innovation LREIS (Grant number: KPI008).

Data availability

The sample plots and HAI data used in this study are available from the authors upon request; ASTER GDEM can be downloaded from https://earthdata.nasa.gov/; temperature and precipitation data can be accessed from the Data Center of Resources and Environment Science, Chinese Academy of Sciences (http://www.resdc.cn).

Competing interests

The author declares no competing interests.