Research Article Received: 13 November 2014
Revised: 13 April 2015
Accepted article published: 17 April 2015
Published online in Wiley Online Library: 12 May 2015
(wileyonlinelibrary.com) DOI 10.1002/jsfa.7219
Change in soil organic carbon between 1981 and 2011 in croplands of Heilongjiang Province, northeast China Lu-Jun Li,a* Martin Burger,b Shu-Li Du,a Wen-Xiu Zou,a Meng-Yang You,a Xiang-Xiang Hao,a Xin-Chun Lu,a Lin Zhenga and Xiao-Zeng Hana* Abstract BACKGROUND: Soil organic carbon (SOC) is fundamental for mitigating climate change as well as improving soil fertility. Databases of SOC obtained from soil surveys in 1981 and 2011 were used to assess SOC change (0–20 cm) in croplands of Heilongjiang Province in northeast China. Three counties (Lindian, Hailun and Baoqing) were selected as typical croplands representing major soil types and land use types in the region. RESULTS: The changes in SOC density (SOCD) between 1981 and 2001 were −6.6, −14.7 and 5.7 Mg C ha−1 in Lindian, Hailun and Baoqing Counties respectively. The total SOC storage (SOCS) changes were estimated to be −11.3, −19.1 and 16.5% of those in 1981 in the respective counties. The results showed 22–550% increases in SOCS in rice (Oryza sativa L.) paddies in the three counties, but 28–33% decreases in dry cropland in Lindian and Hailun Counties. In addition, an increase of 11.4 Mg C ha−1 in SOCD was observed in state-owned farms (P < 0.05), whereas no significant change was observed in family-owned farms. CONCLUSION: Soil C:N ratio and initial SOCD related to soil groups were important determinants of SOCD changes. Land use and residue returning greatly affected SOC changes in the study region. To increase the topsoil SOCD, the results suggest the conversion of dry croplands to rice paddies and returning of crop residue to soils. © 2015 Society of Chemical Industry Keywords: black soil; carbon storage; crop residue returning; state farm; land use; soil organic carbon density
INTRODUCTION
J Sci Food Agric 2016; 96: 1275–1283
different changes in SOC from different starting SOC levels under equivalent treatments. Besides management practices and soil properties, environmental factors such as precipitation, temperature and elevation showed indirect or direct impacts on SOC stock.11 Numerous studies have investigated the SOC dynamics in croplands in China.9,12 Among these studies, there is considerable concern for the soils with high SOM content in northeast China. Inappropriate management practices (such as intensive land use) would lead to potentially large emissions of carbon dioxide (CO2 ) and threaten the regional environment. However, large discrepancies remain in the estimation of SOC storage (SOCS) in northeast China. Numerous studies have reported that croplands have been losing SOC in northeast China during the past several years,12,13 mainly because of soil erosion and low inputs to soils.12 Some authors argued that the large SOC losses in croplands in the
∗
Correspondence to: Lu-Jun Li or Xiao-Zeng Han, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, 138 Haping Road, Harbin 150081, China. E-mail: [email protected] (Li); [email protected] (Han)
a Key Laboratory of Mollisols Agroecology, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Harbin 150081, China b Department of Land, Air and Water Resources, University of California, Davis, Davis, CA, 95616, USA
www.soci.org
© 2015 Society of Chemical Industry
1275
Soil organic matter (SOM) is fundamental for mitigating climate change as well as maintaining and improving soil fertility.1 Thus an accurate estimation of soil organic carbon (SOC) stock is of importance for the detection of the potential for carbon sequestration and has been of increasing concern recently.2 The estimation of SOC stock and its change is frequently constrained by the variability of SOC between soil types and the spatial scale of soil distribution,3,4 land use change and management5,6 as well as the methodologies associated with soil sampling, SOC measurements and estimation procedures, or statistical approaches.7 Among the determinants of SOC changes, agricultural management practice is a critical factor influencing SOC changes in croplands. The application of crop straw is commonly considered as an effective management practice to sequester carbon in soils; however, different effects of straw return were observed on SOC changes.8 In addition, land use types in croplands can also greatly affect the SOC dynamics. By conducting a holistic statistical analysis of SOC data from 1081 observations in croplands across China, Pan et al.9 revealed a much higher increase in SOC for rice paddies than for dry croplands since the 1980s. Moreover, soil basic characteristics could alter the responses of SOC changes to agricultural management. Based on 20 long-term field experiments, Körschens et al.10 found that changes in SOC resulting from agricultural management depended on the starting SOC level, with
www.soci.org
L-J Li et al.
Figure 1. Geographical distribution of study area in Heilongjiang Province, northeast China.
region were predominantly observed in natural soils converted to croplands.14,15 In contrast to these findings, an increase in SOCS was also observed in this region based on a case study.16 Using published data at monitoring sites from croplands across China, Pan et al.9 found that half of the observations showed a decline in SOC for northeast China. In any case, the overall change in SOCS for a given county or province of northeast China remains uncertain, partly owing to the small number of sampling sites within those studies focusing on a national scale. Northeast China accounts for approximately 30% of the maize (Zea mays L.) production in China.17 Heilongjiang Province, located in northeast China, has the largest area of croplands in China and thus plays a unique and central role in China’s agriculture. A model estimation predicted that the cultivation layer of black soils (Phaeozems), the predominantly distributed soil type in northeast China, will disappear within 40–100 years at the actual loss rate of the depth in hilly cropland.13 In this context, an assessment of SOCS changes and exploration of their influencing factors could help manage the black soils as well as predict SOC change in future. The assessment of SOC changes is urgently needed to propose countermeasures to mitigate the potential loss of SOC in this region. Such information could also be particularly important for cropland management in other areas with similar climate and soils to northeast China. Using SOC databases compiled from two surveys in 1981 and 2011, we aimed to estimate the temporal and spatial variations of SOC in croplands and to clarify the influencing factors on SOC change in Heilongjiang Province, northeast China. Based on previous studies, we hypothesized that soil types, land use types and residue management would affect the SOC stock and its change in the study region.
METHODS
1276
Description of study area Northeast China has a typical temperate continental monsoon climate characterized by hot summers and cold winters. Three
wileyonlinelibrary.com/jsfa
counties (Lindian, Hailun and Baoqing) were selected as typical cropland areas in Heilongjiang representing the major soil types (Haplic Phaeozem, Haplic Chernozem, Luvic Phaeozem and Albic Luvisol (FAO/Unesco)) and including different land uses (dry cropland and rice (Oryza sativa L.) paddy) in the province (Fig. 1). The four major soil types occupy over 80% of the total area of croplands in the province, with clay content of approximately 36–39% (w/w) and SOC concentration of 5.9–58.7 g C kg−1 . The mean annual temperature ranged from 2.4 ∘ C for Hailun County to 4.4 ∘ C for Baoqing County, and the mean annual precipitation ranged from about 440 to 550 mm. The general description of the three study counties is shown in Table 1. In the three counties, the main crops in dry croplands are maize and soybean (Glycine max L.). The type of straw returned to soils corresponded to the current crop types. The mean acreage percentages of straw return to soils were 4, 1 and 37% in Lindian, Hailun and Baoqing Counties respectively, and the heights of stubble retained were 7, 6 and 13 cm respectively (Fig. 2). The topography of Hailun County is characterized by gentle undulation with elevation declining from the northeastern to the southwestern area. Along the elevation gradient, Hailun County can be separated into four areas (A, 200 m; B, 225 m; C, 325 m; D, 375 m). With our investigation, we found that cultivation was more intensive in the southwestern area compared with the northeastern area in the long run. State-owned farms in Heilongjiang account for 45% of farmlands in all China’s state farms.18 There are several typical state-owned farms in Baoqing County, including State Farms 597, 851, 852 and 8511. The field management of state-owned farms differs completely from that of family-owned farms. With field investigation, we noticed that the markedly different practice in state-owned farms was that more crop residues were incorporated during the past several years owing to mechanical harvest compared with family-owned farms (Fig. 2). The highly mechanized agricultural system of state-owned farms in Baoqing County resulted in a higher percentage of straw return (69%) compared with family-owned farms (2%), and larger height of stubble retained (15
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 1275–1283
Change in SOC between 1981 and 2011 in croplands of Heilongjiang, China
www.soci.org
Table 1. General description of three counties in Heilongjiang Province, northeast China Item Sampling geographical range MATa (∘ C) MAPb(mm) Elevation (m) Soil mechanical composition (% w/w) Sand (2–0.02 mm) Silt (0.002–0.02 mm) Clay (2 mm). The occurrence of coarse particles in soils was rare in our study area, and thus F was considered to be negligible. The Pedological Knowledge Based (PKB) method22 was used to convert SOCD data sets from discrete points to spatial databases. For each county, we calculated the SOCD change as the difference between the final and initial values in the reported monitoring period (1981–2011) based on an assumption that no land use change did occur. To decrease errors in the processes of site selection and sampling, we calculated SOCD change based on a ‘face-to-face’ method, i.e. the final and initial SOCD values were
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
1277
Data sources Soils were sampled in two different years. The first sampling was performed during the Second National Soil Survey in China in 1981, and a total of 292 samples in croplands were taken within the three counties considering spatial homogeneity. The current soil samples were collected in 2011 under a project financed by the Chinese Academy of Sciences. A total of 260 sampling sites were selected as close to the sites recorded in 1981 as possible. Subsamples taken from five random locations within about 200 m2 of croplands were collected on the topsoil disturbed by tillage (0–20 cm) to form a composite sample for each site. During the sampling within each county, soil types, residue management and land use types (dry cropland and rice paddy) were simultaneously recorded in the field. In addition, because there is a clear elevation gradient only in Hailun and several state farms only in Baoqing, the elevation and soil management (family
vs state farm) were recorded within Hailun and Baoqing Counties respectively. The incorporation of crop residues to soils during the past three decades was surveyed simultaneously by rural household survey combined with field statistics, including the mean height of stubble retained and the acreage percentage of crop straw return to soils. In addition, a portable Global Positioning System (GPS) receiver was used to locate the position of sampling sites. After the visible roots and organic debris were removed, soil samples were sieved (
Revised: 13 April 2015
Accepted article published: 17 April 2015
Published online in Wiley Online Library: 12 May 2015
(wileyonlinelibrary.com) DOI 10.1002/jsfa.7219
Change in soil organic carbon between 1981 and 2011 in croplands of Heilongjiang Province, northeast China Lu-Jun Li,a* Martin Burger,b Shu-Li Du,a Wen-Xiu Zou,a Meng-Yang You,a Xiang-Xiang Hao,a Xin-Chun Lu,a Lin Zhenga and Xiao-Zeng Hana* Abstract BACKGROUND: Soil organic carbon (SOC) is fundamental for mitigating climate change as well as improving soil fertility. Databases of SOC obtained from soil surveys in 1981 and 2011 were used to assess SOC change (0–20 cm) in croplands of Heilongjiang Province in northeast China. Three counties (Lindian, Hailun and Baoqing) were selected as typical croplands representing major soil types and land use types in the region. RESULTS: The changes in SOC density (SOCD) between 1981 and 2001 were −6.6, −14.7 and 5.7 Mg C ha−1 in Lindian, Hailun and Baoqing Counties respectively. The total SOC storage (SOCS) changes were estimated to be −11.3, −19.1 and 16.5% of those in 1981 in the respective counties. The results showed 22–550% increases in SOCS in rice (Oryza sativa L.) paddies in the three counties, but 28–33% decreases in dry cropland in Lindian and Hailun Counties. In addition, an increase of 11.4 Mg C ha−1 in SOCD was observed in state-owned farms (P < 0.05), whereas no significant change was observed in family-owned farms. CONCLUSION: Soil C:N ratio and initial SOCD related to soil groups were important determinants of SOCD changes. Land use and residue returning greatly affected SOC changes in the study region. To increase the topsoil SOCD, the results suggest the conversion of dry croplands to rice paddies and returning of crop residue to soils. © 2015 Society of Chemical Industry Keywords: black soil; carbon storage; crop residue returning; state farm; land use; soil organic carbon density
INTRODUCTION
J Sci Food Agric 2016; 96: 1275–1283
different changes in SOC from different starting SOC levels under equivalent treatments. Besides management practices and soil properties, environmental factors such as precipitation, temperature and elevation showed indirect or direct impacts on SOC stock.11 Numerous studies have investigated the SOC dynamics in croplands in China.9,12 Among these studies, there is considerable concern for the soils with high SOM content in northeast China. Inappropriate management practices (such as intensive land use) would lead to potentially large emissions of carbon dioxide (CO2 ) and threaten the regional environment. However, large discrepancies remain in the estimation of SOC storage (SOCS) in northeast China. Numerous studies have reported that croplands have been losing SOC in northeast China during the past several years,12,13 mainly because of soil erosion and low inputs to soils.12 Some authors argued that the large SOC losses in croplands in the
∗
Correspondence to: Lu-Jun Li or Xiao-Zeng Han, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, 138 Haping Road, Harbin 150081, China. E-mail: [email protected] (Li); [email protected] (Han)
a Key Laboratory of Mollisols Agroecology, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Harbin 150081, China b Department of Land, Air and Water Resources, University of California, Davis, Davis, CA, 95616, USA
www.soci.org
© 2015 Society of Chemical Industry
1275
Soil organic matter (SOM) is fundamental for mitigating climate change as well as maintaining and improving soil fertility.1 Thus an accurate estimation of soil organic carbon (SOC) stock is of importance for the detection of the potential for carbon sequestration and has been of increasing concern recently.2 The estimation of SOC stock and its change is frequently constrained by the variability of SOC between soil types and the spatial scale of soil distribution,3,4 land use change and management5,6 as well as the methodologies associated with soil sampling, SOC measurements and estimation procedures, or statistical approaches.7 Among the determinants of SOC changes, agricultural management practice is a critical factor influencing SOC changes in croplands. The application of crop straw is commonly considered as an effective management practice to sequester carbon in soils; however, different effects of straw return were observed on SOC changes.8 In addition, land use types in croplands can also greatly affect the SOC dynamics. By conducting a holistic statistical analysis of SOC data from 1081 observations in croplands across China, Pan et al.9 revealed a much higher increase in SOC for rice paddies than for dry croplands since the 1980s. Moreover, soil basic characteristics could alter the responses of SOC changes to agricultural management. Based on 20 long-term field experiments, Körschens et al.10 found that changes in SOC resulting from agricultural management depended on the starting SOC level, with
www.soci.org
L-J Li et al.
Figure 1. Geographical distribution of study area in Heilongjiang Province, northeast China.
region were predominantly observed in natural soils converted to croplands.14,15 In contrast to these findings, an increase in SOCS was also observed in this region based on a case study.16 Using published data at monitoring sites from croplands across China, Pan et al.9 found that half of the observations showed a decline in SOC for northeast China. In any case, the overall change in SOCS for a given county or province of northeast China remains uncertain, partly owing to the small number of sampling sites within those studies focusing on a national scale. Northeast China accounts for approximately 30% of the maize (Zea mays L.) production in China.17 Heilongjiang Province, located in northeast China, has the largest area of croplands in China and thus plays a unique and central role in China’s agriculture. A model estimation predicted that the cultivation layer of black soils (Phaeozems), the predominantly distributed soil type in northeast China, will disappear within 40–100 years at the actual loss rate of the depth in hilly cropland.13 In this context, an assessment of SOCS changes and exploration of their influencing factors could help manage the black soils as well as predict SOC change in future. The assessment of SOC changes is urgently needed to propose countermeasures to mitigate the potential loss of SOC in this region. Such information could also be particularly important for cropland management in other areas with similar climate and soils to northeast China. Using SOC databases compiled from two surveys in 1981 and 2011, we aimed to estimate the temporal and spatial variations of SOC in croplands and to clarify the influencing factors on SOC change in Heilongjiang Province, northeast China. Based on previous studies, we hypothesized that soil types, land use types and residue management would affect the SOC stock and its change in the study region.
METHODS
1276
Description of study area Northeast China has a typical temperate continental monsoon climate characterized by hot summers and cold winters. Three
wileyonlinelibrary.com/jsfa
counties (Lindian, Hailun and Baoqing) were selected as typical cropland areas in Heilongjiang representing the major soil types (Haplic Phaeozem, Haplic Chernozem, Luvic Phaeozem and Albic Luvisol (FAO/Unesco)) and including different land uses (dry cropland and rice (Oryza sativa L.) paddy) in the province (Fig. 1). The four major soil types occupy over 80% of the total area of croplands in the province, with clay content of approximately 36–39% (w/w) and SOC concentration of 5.9–58.7 g C kg−1 . The mean annual temperature ranged from 2.4 ∘ C for Hailun County to 4.4 ∘ C for Baoqing County, and the mean annual precipitation ranged from about 440 to 550 mm. The general description of the three study counties is shown in Table 1. In the three counties, the main crops in dry croplands are maize and soybean (Glycine max L.). The type of straw returned to soils corresponded to the current crop types. The mean acreage percentages of straw return to soils were 4, 1 and 37% in Lindian, Hailun and Baoqing Counties respectively, and the heights of stubble retained were 7, 6 and 13 cm respectively (Fig. 2). The topography of Hailun County is characterized by gentle undulation with elevation declining from the northeastern to the southwestern area. Along the elevation gradient, Hailun County can be separated into four areas (A, 200 m; B, 225 m; C, 325 m; D, 375 m). With our investigation, we found that cultivation was more intensive in the southwestern area compared with the northeastern area in the long run. State-owned farms in Heilongjiang account for 45% of farmlands in all China’s state farms.18 There are several typical state-owned farms in Baoqing County, including State Farms 597, 851, 852 and 8511. The field management of state-owned farms differs completely from that of family-owned farms. With field investigation, we noticed that the markedly different practice in state-owned farms was that more crop residues were incorporated during the past several years owing to mechanical harvest compared with family-owned farms (Fig. 2). The highly mechanized agricultural system of state-owned farms in Baoqing County resulted in a higher percentage of straw return (69%) compared with family-owned farms (2%), and larger height of stubble retained (15
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 1275–1283
Change in SOC between 1981 and 2011 in croplands of Heilongjiang, China
www.soci.org
Table 1. General description of three counties in Heilongjiang Province, northeast China Item Sampling geographical range MATa (∘ C) MAPb(mm) Elevation (m) Soil mechanical composition (% w/w) Sand (2–0.02 mm) Silt (0.002–0.02 mm) Clay (2 mm). The occurrence of coarse particles in soils was rare in our study area, and thus F was considered to be negligible. The Pedological Knowledge Based (PKB) method22 was used to convert SOCD data sets from discrete points to spatial databases. For each county, we calculated the SOCD change as the difference between the final and initial values in the reported monitoring period (1981–2011) based on an assumption that no land use change did occur. To decrease errors in the processes of site selection and sampling, we calculated SOCD change based on a ‘face-to-face’ method, i.e. the final and initial SOCD values were
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
1277
Data sources Soils were sampled in two different years. The first sampling was performed during the Second National Soil Survey in China in 1981, and a total of 292 samples in croplands were taken within the three counties considering spatial homogeneity. The current soil samples were collected in 2011 under a project financed by the Chinese Academy of Sciences. A total of 260 sampling sites were selected as close to the sites recorded in 1981 as possible. Subsamples taken from five random locations within about 200 m2 of croplands were collected on the topsoil disturbed by tillage (0–20 cm) to form a composite sample for each site. During the sampling within each county, soil types, residue management and land use types (dry cropland and rice paddy) were simultaneously recorded in the field. In addition, because there is a clear elevation gradient only in Hailun and several state farms only in Baoqing, the elevation and soil management (family
vs state farm) were recorded within Hailun and Baoqing Counties respectively. The incorporation of crop residues to soils during the past three decades was surveyed simultaneously by rural household survey combined with field statistics, including the mean height of stubble retained and the acreage percentage of crop straw return to soils. In addition, a portable Global Positioning System (GPS) receiver was used to locate the position of sampling sites. After the visible roots and organic debris were removed, soil samples were sieved (
