Forest carbon storage in Guizhou Province based on field measurement dataset

Chunzi Guo·Yangyang Wu·Jian Ni·Yinming Guo

Abstract Accurate estimation of forest carbon storage is crucial in understanding global and regional carbon cycles and projecting future ecological and economic scenarios.Guizhou is the largest karst landform province in China;61.9%of its land area is characterized as karst.However,monitoring its field biomass and carbon storage is difficult.This study synthesized and analyzed a comprehensive database of direct field observations of forest vegetation and soil carbon storage in Guizhou Province by using data from existing literature.The total vegetation carbon storage in Guizhou Province was 488.170 TgC,the average vegetation carbon density(VCD)was 27.866 MgC hm-2,the total amount of soil organic carbon(SOC)(20 cm)was 1017.364 TgC,and the average SOC density was 58.074 MgC hm-2.Among all vegetation types,needleleaf forest had the highest vegetation carbon stocks,and scrub presented the highest SOC storage.The vegetation and SOC storage values of the karst landform were 282.352 and 614.825 TgC,respectively,which were higher than those of the non-karst landform.VCD was concentrated at 10-40 MgC hm-2,and SOC density was concentrated at 40-60,60-80,and 80-100 MgC hm-2.This comprehensive regional data synthesis and analysis based on directfield measurement of vegetation and soil will improve our understanding of the forest carbon cycle in karst landforms under a changing climate.

Keywords Forest carbon storage·Field measurement dataset·Karst landform

1 Introduction

Forests contribute to climate change mitigation by storing carbon in tree biomass.Forests play a decisive role in the material cycle of the biosphere and atmosphere,and this situation reveals the dominant position of the forest ecosystem in the global carbon cycle(Dixon et al.1994).Knowledge of the distribution of carbon storage and changes over time is critical for understanding the mechanisms that control the global terrestrial carbon cycle.Accurate estimation of carbon storage is crucial in understanding the processes of global and regional carbon cycles and precisely projecting ecological and economic scenarios for the future(Pan et al.2011).

Global-scale carbon research has become a popular topic worldwide.Many studies have examined the carbon storage and carbon uptake of terrestrial ecosystems in different regions and explored the effects of natural and anthropogenic factors on carbon storage and carbon cycle(Schimel et al.2000;Fang et al.2001,2014;Ni 2001,2013;Pacala et al.2001;Janssens et al.2003;Pan et al.2011).Carbon storage can be measured through direct and indirect methods.Mean biomass density,volume-derived,and remote sensing methods are commonly used approaches for estimating forest biomass carbon storage.Various global models based on biotic and environmental factors have also been developed and improved in recent years.

At present,estimation of forest biomass at large scales is based primarily on forest inventory data or remote sensing technology.The carbon budget of forest ecosystems includes four main components,namely,forest biomass,surface litter,wood residues,and soil organic carbon(SOC).Estimating biomass and carbon density in smallscale ecosystems on the basis of measured data from plots and standard trees offers an accuracy advantage.However,this method is independently performed on a small scale due to the practical difficulties of field measurements.Establishment of large-scale or regional community field carbon storage databases can lay a good foundation for accurate estimation of regional-scale carbon storage.

Guizhou Province in southwestern China has the largest and most continuous karst landform in China.Topography on limestone and dolomite makes its forest type and forest biomass different from others in the same subtropical region(Zhang et al.2012;Ni et al.2015;Liu et al.2016).The influence of geological background on forest biomass should always be considered.Regional syntheses of biomass data were conducted by Luo et al.(2013).However,only a few vegetation administrative divisions in Guizhou were included in the study.In response,extended searches were conducted by other researchers to identify forest biomass,carbon density,and SOC observations in Guizhou Province(An et al.1991;Tu and Yang 1995;Liu et al.2009;Wu et al.2012;Li and Ding 2013).We used these data to estimate forest carbon storage(including vegetation and soil)in Guizhou Province and the carbon storage of various vegetation types and geomorphological features.We also analyzed the spatial distribution of vegetation and SOC density.Through this study,we aim to obtain a thorough understanding of carbon storage by forests on karst landforms and hope to expand our current understanding of the forest carbon cycle in karst areas under a changing climate.

2 Methods

2.1 Study area

Guizhou Province has a land area of approximately 176,167 km2and is located in the northern Yunnan-Guizhou Plateau of southwestern China at approximately 24°37′-29°13′N and 103°36′-109°35′E(Fig.1a).92.5%of its land area is comprised mountains and hills(Fig.1b).Its altitude is between 147.8 and 2900.6 m,with a mean value of 1100 m above sea level.The East Asian and Indian monsoon climate systems,together with the tall mountains in the area,result in diverse climates across the province.The mean annual temperature in most parts of Guizhou Province is between 14 and 16°C(Fig.1c),and the total annualprecipitation is between 1100 and 1400 mm(Fig.1d).The annualsunshine percentage is low(25-30%),so possible sunshine hours in this area range from 1200 to 1500 h(Fig.1e).The province experiences approximately 270 frost-free days and has an active accumulated temperature(≥ 10 °C)of 4000-5500 °C.The province has a broad and continuous distribution of karst landforms which account for 61.9%of the provincial land area(Forestry Department of Guizhou Province 2016)(Fig.1f).Soil erosion and rocky desertification are severe in this area.In 2000,17.1%of the total land area and 25.7%of the exposed carbonate rock area were classified as having undergone rocky desertification(Jiang et al.2014).Areas of rocky desertification expanded 3.76 folds from 1970 to 2005(Cao et al.2009).

The province belongs to a subtropical red soil-yellow soil area(Fig.1g).Among all soil types,yellow earths have the largest distribution area throughout the main part of Guizhou plateau,followed by limestone soils.

2.2 Vegetation type

The province has abundant vegetation because of its warm and moist climates,undulating terrain,and complex soil types.On the basis of the vegetation atlas of China,the vegetation in this province can be divided into eight types,namely,broadleaf forest,needleleaf forest,mixed needleleaf and broadleaf forest,scrub,meadow,grass-forb community,cultivated vegetation,and no vegetation(Fig.1h).Needleleaf forest is the most widely distributed vegetation type in Guizhou Province and has the highest economic value.Evergreen broadleaf forest is the zonal vegetation.Since the 1950s,most of the original forests in the province

have been cut down or degraded by human disturbances.Many protection measures and re-vegetation programs have been implemented to counteract this trend.Secondary vegetation succession has occurred widely after the implementation of these measures(Liu et al.2016).Scrub that grew after the destruction of forest vegetation is the most commonly distributed vegetation type(Forestry Department of Guizhou Province 2016).

Fig.1 Location and environmental factors of Guizhou Province.a Location in China;b DEM(void-filled seamless SRTM data V1,2004,International Center for Tropical Agriculture,available from the CGIAR-CSI SRTM 90 m Database);c mean annual temperature;d mean annual precipitation;e mean annual sunshine percentage;f carbonate area(http://www.sges.auckland.ac.nz/sges_research/karst.shtm);g soil map(Shi et al.2004);and h vegetation distribution(Chinese Academy of Sciences 2007).Baseline climatology data were derived from records of mean monthly temperature,precipitation,and percentage of possible sunshine hours obtained by 1814 meteorological stations across China(740 stations have observation data that cover the period of 1971 to 2000,whereas the remaining stations have observation data that cover the period of 1981 to 1990;China Meteorological Administration,unpublished data)then interpolated to a 0.01 grid by using a three-dimensional thin-plate spline[ANUSPLIN version 4.36(Hancock and Hutchinson 2006)]

The vegetation geographical distribution with a spatial resolution of 1 km in Guizhou Province and related information were obtained from the 1:100 million China Vegetation Map(Chinese Academy of Sciences 2007),which includes vegetation code,vegetation type groups,vegetation types,community types(vegetation formations or subformations),and areas.The vegetation in Guizhou Province can be divided into 7 vegetation type groups,15 vegetation types,and 58 community types.The most widely distributed vegetation type group is scrub.Subtropical and tropical broadleaf evergreen and deciduous scrub have the largest distributions(Table 2).

2.3 Data collection

Many researchers have studied the biomass,carbon storage,and distribution characteristics of different forest types in different regions of Guizhou Province through field measurementssincetheearly 1990s.Thesestudies involved not only forest vegetation carbon but also SOC.We collected and sorted field data on measured forest biomass and SOC in Guizhou Province obtained over the past 30 years,then screened the data on vegetation biomass or carbon and SOC storage according to the community types in the vegetation map for further integration analysis(Supplementary Materials).We recorded the longitude,latitude,altitude,and community type on the basis of the original information of each published paper and sorted thoroughly the biomass/carbon storage data for each sample plot.In the studies we considered,the aboveground biomass of the vegetation tree layer was generally estimated using allometric biomass models or the standard sample tree method and the underground biomass was estimated using allometric biomass models,the standard sample tree method,or the ratio of aboveground biomass to belowground biomass.The biomass of individuals in liana,shrub,and herb layers and the litter biomass were obtained with the clear-cut method.For SOC storage,we used the organic carbon density of the arable topsoil(i.e.,soil layer with a depth of 20 cm)as a standard for assignment and calculation based on the relevant literature and the fact that the soil in southwestern China is shallow and loose.

2.4 Carbon density assignment

We assigned the vegetation carbon density(VCD)and SOC density of 58 community types in Guizhou Province by using the collected biomass,carbon storage,and SOC density data of different community types in combination with vegetation distribution information.First,we assigned the carbon density of each community type in Guizhou Province according to literature records.When a community had multiple biomass research results,we used the average value of these results as the carbon density value of this community type.Second,for a community type in which only aboveground biomass was studied,its underground biomass was estimated by using the ratio of aboveground to belowground biomass(Li and Ren 2004).Third,for a community type in Guizhou with no relevant results available,the results of research on community types that belong to the same vegetation type in Guizhou were used as alternative values.Fourth,the result of the same community type in adjacent provinces was adopted as an alternative value when no research result was available for other community types of the same vegetation type in Guizhou Province.Fifth,for a community type where only vegetation biomass data were available(no carbon density data),the VCD of this community type was calculated as

where VCD is expressed in Mg C hm-2,B is the biomass,and Cc is the carbon-biomass conversion factor.The Cc value used in this study were derived from previous studies(Table 1).

We fully considered the possible errors caused by the distribution of the community types,landforms,elevation,and other factors in the carbon density assignment to ensure that the assignment process was as precise as possible.

2.5 Carbon storage calculation

According to the carbon density of vegetation and SOC(CD),and community area(A)of each community type,the vegetation carbon storage and SOC storage(CS)of different community types in Guizhou Province can be estimated as

where CD is expressed in Mg C hm-2and A is expressed in hm2.

Using this equation,we calculated the total carbon storage of each vegetation type within the province and for the province as a whole.

Vegetation and soil carbon storage for the karst and nonkarst topography of Guizhou Province were calculatedusing vegetation and topographic maps that were stacked and rasterized.

Table 1 Carbon-biomass conversion factor(Cc)of different community types in Guizhou Province

All statistical calculation and drawing were performed with SPSS version 22 and ARCGIS version 10.2.

3 Results

3.1 Carbon density and carbon storage of different community types

The total vegetation carbon storage in Guizhou Province(Table 2)was 488.170 Tg C,and the average VCD was 27.866 Mg C hm-2.The total amount of SOC(0 cm to 20 cm soil)was 1017.364 TgC,and the average SOC density was 58.074 MgC hm-2.The VCD of different community types in Guizhou Province ranged from 3.162 to 261.609 MgC hm-2,and the SOC density ranged from 30.4to126.841 MgC hm-2.TheArundinellasetosa community,Quercus fabri,Quercus serrata var.brevipetiolata scrub,Quercus variabilis and Quercus acutissima scrub exhibited the lowest VCD,which was less than 5 MgC hm-2.The carbon densities of Castanopsis eyrei and Castanopsis carlesii forest,Lysidice rhodostegia and Amesiodendron tienlinensis forest,Quercus rehderiana and Quercus senescens forest,Tsuga chinensis var.tchekiangensis,Cyclobalanopsis multinervis and Styrax subnivers forest were all greater than 100 MgC hm-2.The cultivated vegetation had the lowest carbon storage,and the total carbon storage of Pinus massoniana forest was the highest.The SOC density in cultivated plant communities,Phyllostachys pubescens forest and Sinarundinaria nitida scrub,were lower than 40 MgC hm-2.Although the VCDs of Lysidice rhodostegia and Amesiodendron tienlinensis forest were large,their corresponding SOC density was low.The SOC densities of Cupressus funebris forest,Cyclobalanopsis oxyodon and Cyclobalanopsis gracilis forest,and Castanopsis fargesii and Castanopsis fordii forest ranked

in the top three.The SOC storage values of Sageretia theezans,Rosa cymosa,Pyracantha fortuneana and Bauhinia championii scrub,and Castanea seguinii and Quercus fabri scrub were high mainly because of their wide distribution areas.

Table 2 Carbon density and carbon storage of various community types in Guizhou Province

Table 2 continued

Table 2 continued

Table 3 Carbon storage of different vegetation types

3.2 Carbon storage of different vegetation types and geomorphology

For the seven vegetation type groups in Guizhou Province,the needleleaf forest had the highest vegetation carbon storage,192.332 TgC,and the carbon density was 64.814 MgC hm-2.The mixed needleleaf and broadleaf forest had the lowest vegetation carbon storage,which was only 0.488 TgC.The carbon storage of scrub was second only to that of the needleleaf forest and accounted for 18.86%of the total carbon storage of vegetation.The carbon storage of cultivated vegetation was slightly lower than that of scrub.The carbon storage of broadleaf forest vegetation was 65.886 TgC,which was approximately 13.50%of the total vegetation carbon storage.The SOC storage of the mixed needleleaf and broadleaf forest was also the lowest among all vegetation type groups(0.085 TgC),whereas the SOC storage values of scrub and meadow were the top two and accounted for 37.00%and 22.43%of the total SOC,respectively.Needleleaf forest and cultivated vegetation showed high carbon storage of vegetation and contained high SOC storage(184.343 and 136.933 TgC).

Fig.2 Carbon storage of different landform types in Guizhou Province.a Vegetation and b SOC.A-Subtropical needleleaf forest;BSubtropical mountains mixed needleleaf,broadleaf evergreen and deciduous forest;C-Subtropical broadleaf deciduous forest;D-Subtropical mixed broadleaf evergreen and deciduous forest;E-Subtropical broadleaf evergreen forest;F-Subtropical broadleaf evergreen sclerophyllous forest;G-Tropical monsoon rain forest;H-Subtropical,tropical bamboo forest and scrub;I-Subtropical and tropical broadleaf evergreen and deciduous scrub;J-Subtropical and tropical evergreen xeromorphic succulent thorny scrub;K-Subalpine broadleaf evergreen sclerophyllous scrub;L-Subtropical and tropical grass-forb community;M-Temperate grass and forb meadow;N-Two crops containing upland and irrigation annually,evergreen and deciduous orchards,economic forest;O-Two crops or three crops containing upland and irrigation rotate crops annually(with double-cropping rice),evergreen orchards and subtropical economic forest

Among the vegetation in Guizhou Province,only one vegetation type,subtropical needleleaf forest,existed in the needleleaf forest.Therefore,in all vegetation types,subtropical needleleaf forest presented the highest vegetation carbon storage,and its SOC storage was also high(Table 3).Subtropical and tropical broadleaf evergreen and deciduous scrub accounted for 98.76%of scrub vegetation carbon storage and 97.54%of scrub SOC storage.In other words,this scrub is the most important and most widely distributed scrub vegetation in Guizhou Province.

The karst landform vegetation and SOC storage were 282.352 and 614.825 TgC,respectively,which were higher than those of the non-karst landform(210.865 and 412.674 TgC).Subtropical mountains mixed needleleaf,broadleaf evergreen and deciduous forest,and subtropical,tropical bamboo forest and scrub were all distributed in non-karst landforms.The vegetation and SOC storage of subtropical needleleaf forest,subtropical and tropical broadleaf evergreen and deciduous scrub,and economic forest in karst landforms were higher than those in non-karst landforms.In particular,the vegetation and SOC storage of subtropical and tropical broadleaf evergreen and deciduous scrub accounted for an absolute advantage in karst landforms(Fig.2).This result further shows that the role of scrub in forest carbon accumulation should not be underestimated in Guizhou Province.

3.3 Spatial distribution of carbon density

The carbon density classes of forest vegetation and soil indicated that VCD was mainly concentrated between 10 and 40 MgC hm-2,and SOC density was mainly distributed from 40 to 60,60 to 80,and 80 to 100 MgC hm-2(Fig.3).The locations with high vegetation carbon densities were mostly concentrated in the southeast of Guizhou Province and in small areas in the northwest,which are mostly areaswith concentrated non-karstevergreen broadleaf forests.High SOC densities were distributed in the southeast,small parts of the northwest,and most of the west.Regardless of VCD or SOC density,low vegetation communities were widespread in the karst area.

4 Discussion

4.1 Current status of forest carbon storage in Guizhou Province

Unlike previous studies(Zhou et al.2000;Fang et al.2001;Li 2002;Li and Lei 2010;Wu et al.2012),our work showed that the VCD in Guizhou Province is much lower than that in southwest China,the national and global average value(86.000 MgC hm-2)(Dixon et al.1994).The integrated analysis of field measurements showed that the average VCD in Guizhou Province was 27.866 MgC hm-2.Wu et al.(2012)estimated the forest biomass in the southwest of China to be 162.15 t hm-2by summarizing previous research.If the biomass was converted into carbon density by 0.48,the value would be much higher than the results of this study.Li and Lei(2010)estimated the carbon density of arbor forests in China to be 42.82 MgC hm-2and the carbon density of arbor forests in Guizhou Province to be 34.75 MgC hm-2.The average carbon density of forest vegetation estimated by Zhou et al.(2000)was 57.07 MgC hm-2.Li(2002)estimated the actual average carbon density of forest vegetation in China to be 41.938 MgC hm-2,which is close to the result of forest vegetation inventory data estimated by Fang et al.(2001)(44.91 MgC hm-2).Although the average VCD in Guizhou Province was low,the values of several vegetation types were higher than the national average.For example,the VCDs of subtropical broadleaf evergreen forest and tropical monsoon rain forest(132.830 and 156.619 MgC hm-2)were higher than the national average(26.290 and 70.197 MgC hm-2)and approached the potential average carbon density values(138.448 and 191.826 MgC hm-2)(Li 2002).Different research methods might have varying effects on the research results.Nevertheless,these results indicate that although the forest vegetation resource in Guizhou Province is insufficient,the space for growth is still large.Therefore,implementation of forest protection measures remains necessary and useful.

The average SOC density in Guizhou Province was 58.074 MgC hm-2,which was lower than the average organic carbon density in the 20 cm soil layer of the Chinese forest ecosystem (60.43 MgC hm-2)(Li 2002).Compared with the SOC density in the surface layer of different vegetation types on the national scale based on an ARCGIS model(Xie et al.2004),the SOC densities of subtropical needleleaf forest,subtropical broadleaf evergreen forest,and subtropical and tropical broadleaf evergreen and deciduous scrub were higher,and the SOC densities of subtropical mixed broadleaf evergreen and deciduous forest and subtropical and tropical evergreen xeromorphic succulent thorny scrub were similar.

Fig.3 Spatial distribution of carbon density in vegetation and soil in Guizhou Province(MgC hm-2).a Vegetation and b SOC

The factors that determine carbon accumulation in a community are hydrothermal and soil nutrient conditions.In the short term,the direct impact factors are light,temperature,atmospheric CO2concentration,leaf nitrogen content,leaf area index,and growing season length(Chapin et al.2012);however,from the perspective of longterm community succession,the factors that ultimately determine community carbon inputs are climate,topography,soil parent material,biota,and development time(He 2012).The carbon storage of China's terrestrial ecosystems is closely related to the types of vegetation present,and its spatial pattern is primarily affected by the combined effects of temperature and precipitation(Liu 2009).In addition to climatic factors,the growth of plants and the accumulation of biomass are also affected by soil characteristics.Soil texture directly determines the water-holding capacity and permeable capacity of soil,and soil thickness determines the amount of soil fertility.Vegetation type is also a major control factor for the distribution of SOC(Huang 2000;Jobbágy and Jackson 2000).As a typical karst development area,Guizhou Province has unique characteristics,such as geomorphology,climate,intrinsic feature and distribution of soil,and zonal biological conditions determined by climate.The high temperature and humid climate conditions in Guizhou are conducive to vigorous plant growth and cause abundant plant residues to return to the soil each year.The conditions of damp heat increase the decomposition rate of surface organic matter and reduces accumulation.This is one reason that the growth and accumulation of carbon in forest vegetation and soil in Guizhou Province differ from other regions.However,how these factors quantitatively affect forest carbon accumulation in karst geomorphic areas requires further research.

4.2 Method based on field measurement data integration

When measuring the carbon accumulation and carbon storage of ecosystems on a regional or global scale,the accuracy of some data is often sacrificed because of the need to cover the largest possible spatial scale.The actualfield measurements performed on standard trees are able to obtain as much plant components biomass as possible including understory and litter.When several biomass components could not be directly obtained or when plant biomass was promoted to the sample site biomass,they were estimated based on the interrelationship among the components.However,unlike methods involving forest inventory and remote sensing data,the presented method is as detailed as possible.Therefore,estimating forest carbon storage at the regional scale by using a field measurement dataset based on different research samples is feasible and accurate.

In the southwestern karst region where Guizhou is the center,the high heterogeneity of topography and the depletion of soils make it difficult to conduct field measurements of biomass and carbon storage.Most existing research focused on the southeast and central regions.Some data on biomass measured in the field,such as litter decomposition and animal feed intake,are missing.Measurements from different studies to determine vegetation and soil carbon storage only be carried out in a small area and differences also exist in the respective measurement methods.Existing work could not cover all types of vegetation due to the heavy workload.In this study,although the method of superimposing vegetation and geomorphologic maps to divide landforms was used to replace the method of judging the landforms based only on soil types,the assignment of carbon density to the same community type in different landforms was still the same because of the lack of collected data.

The results of this study not only describe the current situation of forest carbon reserves in Guizhou Province but also provide input data for the establishment and verification of ecosystem models in karst areas.A complete dataset of the forest ecosystem,including vegetation and soil,must be established.In follow-up work,an analysis of forest carbon storage based on dataset integration should be performed separately for the same community type that belongs to both karst and non-karst topographies to improve the accuracy of estimation.This would be of great benefit for further analysis of the differences in forest carbon accumulation between karst landforms and nonkarst landforms.

In order to create a field measurement dataset,a unified standard should be established for the method of field measurement of biomass,and determination of vegetation and soil should be conducted at the same time.Establishment of a perfect cooperation mechanism and a dataset at the national scale is imminent.For example,the China Forest Ecosystem Biomass Database was constructed by Luo et al.(2013),and ‘‘the Biomass and Allometry''dataset(Falster et al.2015)was established for the individual morphological indicators and characteristics of woody plants and the growth environment.A large dataset on ecosystem carbon sequestration was also established by the Chinese Academy of Sciences Strategic Priority Research Program,and the dataset covers more than 16,000 plots in China,including forests,scrubs,grasslands,and farmland ecosystems(Fang et al.2015).The establishment of these datasets has far-reaching implications for the accurate estimation of large-scale forest carbon storage,analysis of the main factors that affect carbon accumulation,and response of forest carbon storage to climate change.

5 Conclusions

We established a new regional dataset of forest vegetation and SOC densities for Guizhou Province.This dataset was used to estimate forest carbon storage and the carbon storage of various vegetation types and geomorphological features.This study is an attempt to estimate forest carbon storage at the regional scale and provide a detailed description of the current situation of forest resources in Guizhou Province.However,in future work,additional data should be collected to establish a complete and sustainable dataset.

AcknowledgementsThis study wasfinancially supported by National Natural Science Foundation of China(Nos.41471049 and 31870462).