全球森林固碳潜力巨大(英文)

J. Resour. Ecol. 2012 3 (3) 193-201 DOI:10.5814/https://www.wendangku.net/doc/1b13596416.html,

Sept., 2012

Journal of Resources and Ecology V

ol.3 No.3Received: 2012-04-05 Accepted: 2012-05-14

Foundation : the National Basic Research Program of China (2010CB833504), the CAS Strategic Priority Research Program (XDA05050600), and the National Natural Science Foundation of China (30590381).* Corresponding author: YU Guirui. Email: yugr@https://www.wendangku.net/doc/1b13596416.html,.

1 Introduction

Forests are important terrestrial ecosystems. On one hand, the biomass carbon storage of global forests is 289–356 Pg C (IPCC 2000; FAO 2010; Pan et al . 2011), accounting for 77% biomass carbon storage of global terrestrial ecosystems (IPCC 2000). On the other hand, the carbon exchange between global forests and the atmosphere is huge. For example, the average GPP of global forests between 1998 and 2005 is about 59 Pg C (Beer et al . 2010), and the global intact forests sequestrated 2.4±0.4 Pg C y -1 from 1990 to 2007 (Pan et al . 2011). There are two factors primarily limiting the forest carbon sequestration. First, the forest can not absorb carbon permanently. With forest growth, the carbon stock would achieve a saturated state, called carbon carrying capacity (CCC) (Keith et al . 2009). That means that the forest carbon sequestration has an upper limit (Odum 1969), named carbon sequestration potential (CSP) (Keith et al . 2009). Second, not all the forests can be conserved due to deforestation and forest degradation. Therefore, focusing attention onto the forests with high CCC and CSP would be a

trade-off between protecting forest carbon sink and meeting human demand for forest products.

For the regional average biomass of existing forests, the tropical rainforests (IPCC 2006), Northwest of USA (Hudiburg et al . 2009) and Southeast of Australia (Keith et al . 2009) have high biomass carbon density. There are two factors determining the high carbon density. One is forest age or recovery years from disturbance, which determines forest biomass directly (Pregitzer and Euskirchen 2009). The other is climate (temperature, precipitation, etc.), which have important impacts on the ecosystem succession, pattern and production (Odum 1969; Stegen et al . 2011).

Climatic gradient is believed to determine the spatial pattern of forest biomass. Correspondingly, the average forest biomass decreases from tropical to subtropical, temperate and boreal forests based on IPCC reports (IPCC 2006). Some scholars proposed that the average biomass of old-growth forests, much higher than that of current forests as advised by IPCC (2006), could represent the potential biomass carbon storage (carbon carrying capacity) of the biomes (Keith et al . 2009). According to Keith (Keith et al . 2009), the old-

Huge Carbon Sequestration Potential in Global Forests

LIU Yingchun 1,2, YU Guirui 1*, WANG Qiufeng 1 and ZHANG Yangjian 1

1 Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China;

2 University of Chinese Academy of Sciences, Beijing 100049, China

Abstract: Forests play an important role in mitigating climate change by absorbing carbon from atmosphere. The global forests sequestrated 2.4±0.4 Pg C y -1 from 1990 to 2007, while the quantitative assessment on the carbon sequestration potential (CSP) of global forests has much uncertainty. We collected and compiled a database of site above-ground biomass (AGB) of global mature forests, and obtained AGB carbon carrying capacity (CCC) of global forests by interpolating global mature forest site data. The results show that: (i) at a global scale, the AGB of mature forests decline mainly from tropical forests to boreal forests, and the maximum AGB occurs in middle latitude regions; (ii) temperature and precipitation are main factors influencing the AGB of mature forests; and (iii) the above-ground biomass CCC of global forests is about 586.2±49.3 Pg C, and with CSP of 313.4 Pg C. Therefore, achieving CCC of the existing forests by reducing human disturbance is an option for mitigating greenhouse gas emission.Key words: climatic gradient; global forests; mature forest; above-ground biomass (AGB); carbon carrying capacity (CCC); carbon sequestration potential (CSP)

Journal of Resources and Ecology V ol.3 No.3, 2012

194

growth forests in temperate moist forests have the highest carbon storage, higher than both of tropical and boreal forests. That means the CCC of temperate moist forests is higher than those of tropical and boreal forests. The forest CSP is the difference between the current carbon storage and CCC under current climate regime and disturbances (Odum 1969; Keith et al . 2009; Liu et al . 2011). It can be inferred that temperate moist forests also have higher CSP than tropical forests. Therefore, the relationship between the CCC, CSP and current biomass (or climatic factors) of global forests may be not a linear positive correlation in different ecological zones. The assessments of CCC and CSP of global forests are mostly based on simulating carbon balance of productivity and respiration determined by climatic factors (Cramer et al . 2001). However, it is difficult to evaluate the uncertainty in simulation results (Cramer et al . 2001; Keenan et al . 2012) and assess the carbon sequestration potential of global forests due to the lack of old-growth forest site data.

We collected and compiled global mature forest site inventory data, because the biomass and growth stage of mature forests were close to those of old-growth forests under similar climate. Based on these data, we calculated the average above-ground biomass carbon density of mature forests in each ecological zone, analyzed the spatial pattern of carbon carrying capacity of global forests, and finally evaluated the carbon sequestration potential of global existing forests.

2. Materials and methods

2.1 Data collection and compilation

The data used in this study include: global mature forests biomass site data, climate data, global ecological zoning data from FAO (https://www.wendangku.net/doc/1b13596416.html,/), and global forests distribution data from the Global Land Cover 2000 database (http://bioval.jrc.ec.europa.eu/products/glc2000/glc2000.php) with a spatial resolution of 1 km.2.1.1 Mature forest biomass

Mature forests have a similar meaning with old-growth

forests, especially for the above-ground biomass and growth stage of forests. A mature forest is when the growth of timber has reached a stage of being extremely slow or being almost saturated, and the timber volume begins to decrease or quality begins to degrade due to various reasons (Meng 2007). The methods are different for determining whether a forest is old-growth forests (Kira and Shidei 1967; Odum 1969; Luyssaert et al. 2007; Pregitzer and Euskirchen 2009; Goulden et al . 2011). To make the data comparable between different sites and cover more ecological zones, we treated both of “the old-growth forest sites” in the literature and the forests ≥ 80 years (Odum 1969) as mature forests, taking into account of both the amount and pattern of the forest sites. The stand age of mature forests were obtained from the literature or from tree core rings in our field work (Liu et al . 2011).

The global mature forest biomass data, with a total of 728 sites, are primarily from literatures and site surveys, among which 118 from Keith et al . (2009), 79 from Lewis et al . (2009), 112 from Luyssaert et al . (2007), 297 from Luo (1996), and 119 from other literatures (CERN) and 3 from our field investigations (Liu et al . 2011). All the sites biomass data were collected from forest inventory. The forest sites were invented in permanent plots, or temporary plots located at random. The size of each sample plot is ≥0.06 h a in boreal and temperate forests (Liu et al . 2011), and ≥0.1 ha in subtropical and tropical forests (Feng et al . 1999; Lewis et al . 2009). All trees ≥4 cm in diameter at breast height (DBH) were measured in boreal and temperate forests (Liu et al . 2011), and ≥10 cm in tropic forests (Lewis et al . 2009). The diameter and height measurements were converted to biomass using the published allometric equations (Feng et al . 1999; Lewis et al . 2009; Liu et al . 2011).

The factor of biomass to carbon stock was assumed 0.5 g C g -1 (Lewis et al . 2009). All the site locations are shown in Fig. 1. The dataset covers 15 ecological zones, including all the zones listed in the IPCC except tropical shrubs, subtropical dry forests and subtropical grasslands. For the sites that lack of latitude or longitude, we obtained them in

Fig. 1 Distribution of mature forest sites and added sites.

Note: Mature forest (green point) data is collected from literatures and field survey; of added sites (blue point), assuming the above-ground biomass is 0 Mg C hm -2. The added sites are distributed in Antarctic, Sahara, Arabia, Patagonia, Kalahari, Great Sandy, Kara-kum, Taklimakan Desert, Gurbantunggut Desert, Tenger Desert, and Gobi Desert

Added site

Old-growth forest

LIU Yingchun, et al .: Huge Carbon Sequestration Potential in Global Forests

195

Google Earth according to the site names. 2.1.2 Spatial dataset of climate

The mean annual temperature and precipitation of China from 1980 to 2000, with a spatial resolution of 1km (Yu et al . 2004; He et al . 2004; Liu et al . 2004), were collected from the Chinese Ecosystem Research Network (CERN). The climate data in other regions of the world is the average monthly temperature and precipitation from 1961 to 1990, were obtained from IPCC (https://www.wendangku.net/doc/1b13596416.html,/, New et al . 2002). Two spatial resolutions of the global climate were used, 10′ and 0.5°, respectively. The mean annual temperature data used in this paper is the average of the 12 monthly temperature, and the mean annual precipitation is the sum of the 12 monthly precipitation. The mean annual temperature and mean annual precipitation of mature forest sites are mainly collected from the literature. We only extracted for the sites that lack of temperature and precipitation from the global (with a spatial resolution of 10′) or China’s climatic data. The global climatic data, with a spatial resolution of 0.5°, is used in Partial Thin Plate

Smoothing Spline interpolation (See 2.2.3).

2.2 Forest above-ground biomass carbon carrying

capacity According to the classical theory of ecology (Odum 1969): the carbon storage increases rapidly when the forests are in developmental stage or recovery stage from disturbance, thus, the forests act as carbon sink. When the forests are older than 80 or 100 years, called old-growth forests, the carbon stocks grow slowly, and the carbon exchange between forests and the atmosphere gradually approaches an equilibrium state, thus, the forests act as a relatively weak carbon sink or mainly in a state of carbon neutrality (Jarvis et al . 1989; Zhou et al . 2002). Therefore, the carbon stocks of old-growth forests can be regarded as a reference of the carbon carrying capacity of the forests under similar climate.

The spatial climate data was interpolated from site observations for the limited amount of observation sites. Similarly we can obtain regional carbon carry capacity from site data by interpolation. Or, if old-growth forest biomass and climatic variables, such as temperature and precipitation, can be combined with empirical regression relationships, the regional carbon carry capacity could be simulated based on spatial climate data.

We applied Above-ground Biomass-Climate Regression Kriging, Inverse Distance Weighted interpolation (Bartier and Keller 1996), and Thin Plate Smoothing Spline interpolation (Hutchinson 2001) to simulate carbon carrying capacity of forest above-ground biomass, with a spatial resolution of 1 km. We added 82 points with 0 Mg C ha -1 of above-ground biomass in the Arctic, Antarctic and deserts to increase the number of control points and enhance the precision of interpolation (Fig. 1).

2.2.1 Above-ground Biomass-Climate Regression Kriging We used equation (1), primarily based on the “law of the minimum” (Lieth 1973), to determine the relationship of mature forest above-ground biomass to mean annual temperature and mean annual precipitation (Fig. 2).

B t =exp(0.000048T 3-0.003959T 2+0.094659T+4.535219), R 2=0.97, P < 0.01

B p =–0.000024P 2+0.155735P+5.15818, R 2=0.82, P <0.01 B m =min(B t , B p ) (1)

where B t is the above-ground biomass of mature forests primarily limited by mean annual temperature (Mg C ha -1); T is the mean annual temperature (℃); B p is the above-ground biomass of mature forests primarily limited by mean annual precipitation, Mg C ha -1; P is the mean annual precipitation, mm; and B m is the above-ground biomass of mature forests controlled by both mean annual temperature and mean annual precipitation, Mg C ha -1.

Apart from the equation (1), we also tried binary linear equation (B m =61.7+2.3294T +0.00427P , R 2=0.31, P <0.01) in describing the relationships of above-ground biomass

Fig. 2 The pattern of above-ground biomass of global mature forests with mean annual temperature (a) and mean annual precipitation (b).

The points are the average above-ground biomass of mature forests, calculated with every 3?C mean annual temperature or 300mm mean annual precipitation

L n (A b o v e -g r o u n d b i o m a s s )(M g C h a -1)

Mean annual temperature (?C)

-2001020-101

3452(a)

30

6

Mean annual precipitation (mm)

200030004000100050006000

A b o v e -g r o u n d b i o m a s s (M g C h a -1)

010015020050250300

(b)