Development of a high-spatial-resolution annual emission inventory of greenhouse gases from open straw burning in Northeast China from 2001 to 2020

Open straw burning has been widely recognized as a significant source of greenhouse gases (GHGs), posing critical risks to atmospheric integrity and potentially exacerbating global warming. In this study, we proposed a novel method that integrates crop cycle information into extraction and classification of fire spots from open straw burning in Northeast China from 2001 to 2020. By synergizing the extracted fire spots with the modified fire radiative power (FRP) algorithm, we developed high-spatial-resolution emission inventories of GHGs, including carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O). Results showed that the northern Sanjiang Plain, eastern Songnen Plain, and eastern Liao River plain were areas with high intensity of open straw burning. The number of fire spots was evaluated during 2013–2017, accounting for 58.2 % of the total fire spots observed during 2001–2020. The prevalent season for open straw burning shifted from autumn (pre-2016) to spring (post-2016), accompanied by a more dispersed pattern in burning dates. The 2-decade cumulative emissions of CO₂, CH₄, and N₂O were quantified at 198 Tg, 557 Gg, and 15.7 Gg, respectively, amounting to 218 Tg of CO₂-eq (equivalent). Significant correlations were identified between GHG emissions and both straw yield and straw utilization (p<0.01). The enforcement of straw-burning bans since 2018 has played a pivotal role in curbing open straw burning and has reduced fire spots by 51.7 % on an annual basis compared to 2013–2017. The novel method proposed in this study considerably enhanced the accuracy in characterizing spatiotemporal distributions of fire spots from open straw burning and quantifying associated pollutant emissions.

1 Introduction

Open straw burning, a customary practice in agricultural areas, serves multiple purposes, including rapid straw disposal, weed control, nutrient release, and pest management (Korontzi et al., 2006; Wen et al., 2020). This practice results in short-term yet intense emissions of greenhouse gases (GHGs), such as carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O). The accumulation of these gases in the atmosphere adversely impacts climate and atmospheric chemistry (Weldemichael and Assefa, 2016; Tang et al., 2020; Hong et al., 2023). To date, open straw burning remains prevalent in grain-producing areas globally, despite the many drawbacks of such a practice (Gadde et al., 2009; Huang et al., 2013; Zhu et al., 2015; Ahmed et al., 2019; Mehmood et al., 2020; Fu et al., 2022; Huang et al., 2023; Xu et al., 2023). Thus, accurate and high-spatial-resolution emission inventories for GHGs from this source sector are needed from regional to global scales to assess potential climate and air quality impacts and formulate carbon mitigation policies.

The “bottom-up” approach, which is based on the amount of straw burned and corresponding emission factors, has been widely employed to establish emission inventories for various pollutants emitted from open straw burning (van der Werf et al., 2017; Wang et al., 2018; Liu et al., 2021; Zheng et al., 2023). Emission factors for diverse pollutants released from different types of straw burning have been extensively investigated in laboratory studies (Li et al., 2007; Liu et al., 2011; Stockwell et al., 2014; Pan et al., 2017; Peng et al., 2016; Sun et al., 2016). However, estimation of the amount of straw burned is subject to large uncertainties since it involves many parameters, such as grain yield, ratio of straw and grain, open burning proportion, burning efficiency, and dry matter fraction (Guan et al., 2017; Zhou et al., 2017). Consequently, existing regional-scale emission inventories based on the “bottom-up” approach generally have large uncertainties and low spatiotemporal resolutions (Tian et al., 2011; Jin et al., 2017).

The advent of satellite technologies, such as the Moderate Resolution Imaging Spectroradiometer (MODIS, remote sensing instrument), Visible Infrared Imaging Radiometer Suite (VIIRS, remote sensing instrument), and Himawari-8 (geostationary satellite), has markedly revolutionized the monitoring of open straw burning, enabling real-time and high-spatiotemporal-resolution fire spot products to be accessible to the general public (Schroeder et al., 2014; Giglio et al., 2016; Xu et al., 2017; Wu et al., 2018; Zhuang et al., 2018; Lv et al., 2024). Many studies have effectively utilized satellite fire spot products for constructing emission inventories, based on either the burned area (BA) or fire spot counts (FCs) (Ke et al., 2019; Cui et al., 2021). Several studies have also developed a hybrid inventory strategy using the “bottom-up” approach to allocate GHG emissions spatially and temporally based on BA or FCs (Huang et al., 2012; Jin et al., 2018; Zhang et al., 2019; Kumar et al., 2021). These approaches have significantly improved the spatiotemporal resolutions of the emission inventories for open straw burning (Wu et al., 2023).

MODIS and VIIRS, both operating in polar orbits, provide only two observations per day. MODIS has provided 1 km resolution fire data since 2000, which are suitable for long-term trend analyses (Chen et al., 2022), while VIIRS has provided fire data at a 375 m resolution since 2012, which is more suitable for detecting small fires (Chen et al., 2022). Himawari-8 (geostationary orbit) has provided 10 min temporal resolution and 2 km spatial resolution fire data since 2015, ideal for real-time monitoring across the Asia–Pacific region (Zhang et al., 2020). However, the aforementioned datasets remain inadequate for accurately capturing small-area, short-duration open straw burning, particularly in scattered farmlands (Wiedinmyer et al., 2014). It should also be noted that meteorological disturbances, such as cloud cover and rainfall, can reduce the accuracy of these products (Schroeder et al., 2014; Ying et al., 2019). Furthermore, straw burning during non-satellite transit periods, on cloudy days, at night, and under heavy haze may not be captured in these datasets (Liu et al., 2020). For example, Liu et al. (2019) found that same-day omission error of MODIS burned area product could be as high as 95 % for agricultural fire detection during the post-monsoon season in northwestern India.

With continuous enrichment of satellite data, a strong relationship was observed between fire radiative power (FRP) and emission amounts from open straw burning (Wu et al., 2023). Consequently, the FRP algorithm has been widely accepted for estimating emissions (Wooster et al., 2005; Freeborn et al., 2008; Vermote et al., 2009; Yang and Zhao, 2019). The FRP algorithm has been optimized by integrating multi-source satellite fire spot data, field survey data, and ground observation data and combined with advanced modeling techniques to improve the accuracy of emission inventory for open straw burning. For example, Liu et al. (2020) revised FRP by combining household survey results with satellite observations in northern India to capture small fires, fill cloud/haze gaps in satellite observations, and adjust partial-field burns and diurnal cycle of fire activity disturbances. Yang et al. (2020) improved the FRP algorithm by calibrating the contributions of open straw burning to ground observation data in Northeast China based on model simulation results using the coupled Weather Research and Forecasting model and Community Multiscale Air Quality (WRF-CMAQ) model.

At present, the identification of straw types in open straw burning typically relies on crop data, such as the International Geosphere-Biosphere Programme (IGBP)-Modified MODIS Land Use and MapSPAM datasets (Ke et al., 2019; Yang et al., 2020). These low-spatiotemporal-resolution crop data contribute to errors in both the extraction of fire spots and the identification of straw types (Ke et al., 2019; Liu et al., 2022). Additional errors come from planting structure adjustment and frequent variations in crop phenology. For instance, fire spots that occurred during crop growth might be incorrectly classified as open straw burning, while those that occurred prior to crop growth could be inaccurately attributed to burning of straw from subsequent harvests (Zhou et al., 2022). Therefore, high-spatiotemporal-resolution data on crop types and phenology are critical, and such data should be integrated into the extraction and classification of fire spots from open straw burning to accurately estimate emissions of various pollutants from this source sector.

To control emissions from open straw burning, the Air Pollution Prevention and Control Action Plan (APPCAP) went into effect in 2013 in China (Huang et al., 2021). In addition, China committed to achieving a carbon peak by 2030 and carbon neutrality by 2060, which draws unprecedented challenges in reducing carbon emissions from open straw burning (Wu et al., 2023). As a significant grain-producing region in China, Northeast China produced 1.35×10⁸ t of major grains (corn, rice, beans, and wheat) in 2020, accounting for 21.4 % of total production in China (National Bureau of Statistics of China, 2021). During 2013–2018, open straw burning in Northeast China exhibited an increasing trend while decreasing in all other regions of China (Huang et al., 2021). The constant increase reflects the expansion of the agricultural sector and economic development in Northeast China yet relatively unconstrained open burning activities (Huang et al., 2021). Liu et al. (2022) estimated CO₂ emissions from open straw burning in Northeast China to be as high as 344 Tg from 2012 to 2020.

In this study, high-spatial-resolution fire spot products were used to develop annual emission inventories of GHGs, including CO₂, CH₄, and N₂O, from open straw burning in Northeast China for the period of 2001–2020. To improve the accuracy of the developed emission inventory, a novel concept that integrates the crop cycle information into fire spot extraction and classification was adopted. Furthermore, this study conducted a thorough analysis to assess the driving factors influencing GHG emissions during the 2 decades. This study comprehensively examined GHG emissions from open straw burning in Northeast China and offered valuable insights for policy makers to mitigate carbon emissions and air pollution in agricultural areas.

2 Methodology

2.1 Extraction and classification of fire spots

The MODIS fire product (MCD14ML, Collection 6.1) was selected from 1 January 2001 to 31 December 2020 for the whole region of Northeast China (Giglio et al., 2016, https://sftp://fuoco.geog.umd.edu, last access: 4 July 2024). The dataset, with a spatial resolution of about 1 km², includes essential variables, such as latitude, longitude, acquisition date and time (in UTC), satellite (Aqua or Terra), FRP, and fire type (presumably vegetation fire, active volcano, other static land source, and offshore), among others (https://modis-fire.umd.edu/files/MODIS_C6_C6.1_Fire_User_Guide_1.0.pdf, last access: 4 July 2024). Non-vegetation fire activities (active volcano, other static land source, and offshore) were then filtered out from the selected dataset for subsequent analysis.

To clarify, the MCD14ML underestimated fire spots in 2001 and 2002 because only the Terra satellite was operational before 3 July 2002. Therefore, data for the years 2003 to 2020 were used for developing annual emission inventories, with relevant results for 2001 and 2002 as reference only. Also, failure of the Aqua satellite on 16 August 2020 led to the loss of fire spot data for about 2 weeks (https://modis-fire.umd.edu/files/MODIS_C61_BA_User_Guide_1.1.pdf, last access: 4 July 2024). However, as August is a crop-growing period in Northeast China, this failure would not lead to an underestimation of fire spots from open straw burning.

The ChinaCropArea1 km and ChinaCropPhen1 km datasets were used to extract and classify fire spots from open straw burning (Luo et al., 2020a, b). These datasets present annual data on the type and phenology (day of year (DOY) of emergence and maturity) of grain crops (corn, rice, and wheat). Considering that Northeast China is a major bean-producing area, we also compiled bean distribution datasets (Li et al., 2021; Xuan et al., 2023). However, bean distribution in Jilin and Liaoning provinces was not recorded during 2001–2012 in this dataset. The dataset was extended to the whole region of Northeast China (Heilongjiang, Jilin, and Liaoning provinces) after 2013. Thus, some gaps still exist in these datasets compared to the comprehensive information required for this study, as detailed in Table S1 in the Supplement.

Figure 1 describes the meticulous process of accurately extracting and classifying fire spots from open straw burning in areas experiencing one harvest season every year. The process involves several key steps:

•  

  In Step 1, the current year's ChinaCropPhen1 km and ChinaCropArea1 km data, along with the previous year's ChinaCropArea1 km data, were extracted to fire spots (MCD14ML) by ArcGIS 10.2 software to obtain the Fire spots–Crop–Phen dataset.

•  

  In Step 2, considering the crop cycle, the extraction of fire spots was divided into two stages. The first stage is before crop growth (spring) and requires the fire spot to satisfy two conditions: (a) there was a crop planted in the previous year, and (b) the burning date is before emergence. The second stage is after crop growth (autumn) and also involves two conditions: (a) there was a crop planted in the current year, and (b) the burning date is after maturity.

•  

  In Step 3, for fire spots in spring, the type of straw burned is identified based on the previous year's crop type. For autumn fire spots, the straw type is determined according to the crop type of the current year.

Furthermore, fire spots from open straw burning were extracted using the traditional method that does not integrate crop cycle information. Only the current year's ChinaCropArea1 km data were extracted to fire spots (MCD14ML). Then, fire spots occurring on agricultural land with growing crops were identified as open straw burning.

Figure 1 Extraction and classification method for fire spots from open straw burning.

2.2 Development of high-spatial-resolution annual emission inventories for GHGs and exploration of driving factors

Annual emission inventories for GHGs were developed for the region of Northeast China at a grid resolution of 5 km × 5 km for the years 2001 to 2020. The domain grids were created using Fishnet of ArcGIS 10.2 software.

The modified FRP algorithm (Yang et al., 2020) is used to estimate the emissions of GHGs from open straw burning in this study:

$$\begin{array}{}\text{(1)}& \begin{array}{rl}E& =\mathit{\alpha }\times \underset{t_{\mathrm{1}}}{\overset{t_{\mathrm{2}}}{\int }}{\mathrm{FRP}}^{\ast }\mathrm{d}t\times \mathit{\beta }\times F=\mathit{\alpha }\times \mathrm{FRP}\\ & \times f_{\mathrm{FRP}}\times (t_{\mathrm{2}}-t_{\mathrm{1}})\times \mathit{\beta }\times F,\end{array}\end{array}$$

where E (in g) is the emissions of GHGs; α is a correction factor used to adjust for FRP detection errors between MODIS and VIIRS, which is given a value of 2.5 following Vadrevu and Lasko (2018), indicating that the FRP VIIRS sum is 2.5 times the FRP MODIS sum. t₁ and t₂ are the beginning and ending time of fire spots, respectively. The average burning time (3 h) of a fire spot in Northeast China was obtained by delivering questionnaires to local farmers (Yang et al., 2020). FRP^∗ (in MW) is the adjusted satellite-detected FRP. FRP (in MW) is the instantaneous FRP observed by satellite. f_FRP is a correction factor that is used to adjust the underestimated emissions by fire spots, and Yang et al. (2020) determined an optimal value of 5 for f_FRP by calibrating the contributions of open straw burning to ground observation data in Northeast China using WRF-CMAQ. β (in kg MJ⁻¹) is the biomass combustion rate, and the average value of 0.411 kg MJ⁻¹ from previous studies is used here (Wooster et al., 2005; Freeborn et al., 2008). F (in g kg⁻¹) is the emission factor for individual straw type (Table 1) (Li et al., 2007; Liu et al., 2011; Peng et al., 2016).

Table 1 Emission factors of open straw burning for different crop types.

Driving factors such as the output of major grains and rural residential coal consumption for temporal variations of annual GHG emissions were explored through Pearson correlation analysis using SPSS 20.0. Information on the above data is also detailed in Table S1.

3 Results and discussion

3.1 Spatial and temporal distributions of fire spots

Cultivated lands in Northeast China are primarily distributed in Sanjiang Plain (northeastern Heilongjiang Province), Songnen Plain (western Heilongjiang Province and midwestern Jilin Province), and Liao River plain (central Liaoning Province) (Fig. 2a). Fire spots were widely spread, covering most cultivated lands, including both dry and paddy fields across Northeast China (Fig. 2a and b). A total of 156 044 fire spots from open straw burning were recorded during 2001–2020. Note that the traditional method overestimated the total number of fire sports by 7190 over the 20-year period, with the largest in 2017 (an overestimation of 4060) (Fig. 2c). This highlights the importance of integrating crop cycle information into fire spot extraction for open straw burning to enhance data accuracy and reliability. Considering the 20-year period (2001–2020), high occurrence frequencies of open straw burning (also referred to as intensity of fire spots below) appeared in the northern Sanjiang Plain, eastern Songnen Plain, and eastern Liao River plain, as well as scattered areas close to Inner Mongolia (Fig. 2a and b).

Interannual variations of fire spot distributions are shown in Fig. S1. In the Sanjiang Plain, low occurrence frequencies of fire spots were observed in a few cultivated lands during 2003–2006 (Fig. S1c to f) and in most cultivated lands in the northern part of the plain during 2007–2013 (Fig. S1g to m). Note that in 2014 and later years, fire spots were extended to the entire Sanjiang Plain, and the northern part of the plain became an area with a high intensity of fire spots (Fig. S1n to q), although a few cultivated lands in this plain recorded a low intensity of fire spots after 2018 (Fig. S1r to t). In the Songnen Plain, most cultivated lands recorded fire spots from 2014 to 2017, with the highest intensity in the northern and eastern parts of the plain (Fig. S1n to q). The occurrence frequencies of fire spots have decreased across the plain since 2018, particularly in the northern part of the plain (Fig. S1r to t). In the Liao River plain, although fire spots were observed in most cultivated lands in the eastern part of the plain during 2014–2017, high occurrence frequency was only recorded in 2014 (Fig. S1n to q).

Apparently, open straw-burning events have decreased in all of the three plains since 2018 (Fig. S1r to t), which was likely due to the intensified effort from the Chinese government in banning open straw burning (Hong et al., 2023). The reduction in the number of fire spots was more significant in the Sanjiang Plain and northern Songnen Plain than in the Liao River plain (Fig. S1), indicating more compliance with straw-burning bans from state farms in the former two regions.

Figure 2 (a) Spatial distributions of cultivated land in 2020 in Northeast China (https://www.resdc.cn, last access: 17 November 2023), (b) spatial distributions of the total number of fire spots during 2001–2020 in Northeast China, and (c) seasonal distributions of the annual fire spots and annual overestimated fire spots with the traditional method from 2001 to 2020. The overestimated fire spots are calculated as the number of fire spots identified by the traditional method minus those extracted by the novel method.

Fire spots from open straw burning were concentrated in spring and autumn, with few burning events in the other two seasons in Northeast China. Open straw burning events in this region during 2003–2020 can be roughly divided into three distinctive periods (Fig. 2c). During Period I (2003–2012), the annual average number of fire spots in this region was 3732. There were more fire spots in autumn than in spring in most of these years. During Period II (2013–2017), there was a substantial surge in fire spots, with an annual average of 18 177 spots, accounting for 58.2 % of the 20-year total. Notably, the number of fire spots peaked at 25 759 in 2014. Spring fire spots consistently increased annually, reaching the highest in 2017 at 12 094 spots. The variations in autumn fire spots fluctuated, with a peak of 18 951 spots in 2014. During 2013–2015, autumn fire spots were higher than spring; however, this trend reversed in 2016 and 2017, with spring fire spots becoming more dominant. During Period III (2018–2020), the number of fire spots experienced a significant decrease, averaging 8788 spots annually, which was a 51.7 % decrease from Period II. Spring emerged as the primary season of fire spots, accounting for approximately 93.8 % of the annual total. Zhao et al. (2021) have reported a similar phenomenon, in which the primary season of open straw burning in Northeast China gradually shifts to spring (April to June). The apparent seasonal variation of open straw burning primarily stems from strict government bans imposed after the autumn harvest (Yang et al., 2020). In addition, farmers' increasing awareness regarding how open straw burning contributes to the thawing of spring soil may also be a factor (Saxton et al., 1993; Song et al., 2024).

However, the “sudden drop” in fire spots should also be partially attributed to strategies employed by farmers to avoid detection by satellite and government regulations, such as burning straw on smaller scales and in more dispersed areas, or during non-transit times of the satellites (Liu et al., 2019, 2020). Chen et al. (2022) also found that farmers in East China frequently burned straw in 2019 during non-transit times of the MODIS and VIIRS instruments, as indicated by Himawari satellite data. To further verify the reliability of the “sudden drop” in fire spots in Northeast China, we analyzed the trend of particulate matter concentrations (PM₁₀ and PM_2.5) during the periods of open straw burning from 2014 to 2020 in Northeast China (Fig. S2). Atmospheric particulate matter concentrations during autumn open straw burning in Northeast China decreased with a “sudden drop” in fire spots (Fig. S2c). However, a similar trend was not observed in spring (Fig. S2b), possibly due to limitations in fire spot detection by current satellite techniques and avoidance strategies. Kumar et al. (2021) suggested that a hybrid inventory that accurately allocates emissions estimated using the “bottom-up” approach based on satellite data may be more advantageous in this scenario.

The straw-burning dates in Northeast China also changed during the three periods, besides varying with crop type. During Period I (2003–2012), the autumn burning dates of corn and rice straw were concentrated from early October to mid-November (DOY range of 270 to 320). Spring burning dates of corn and rice straw were concentrated between mid-March and late April (DOY range of 70 to 120) in 2003, while the were dispersed from early March to mid-May (DOY range of 60 to 140) in 2012 (Fig. 3a and b). During Period II (2013–2017), the dispersion of spring burning dates for corn and rice straw became more pronounced, extending from early February to mid-May (DOY range of 30 to 140) (Fig. 3a and b). During Period III (2018–2020), the dispersion of spring burning dates for corn and rice straw persisted (Fig. 3a and b). During Period I (2003–2012), the spring and autumn burning dates of bean straw in Heilongjiang Province were concentrated from mid-March to late April (DOY range of 70 to 120) and from early October to mid-November (DOY range of 270 to 320), respectively (Fig. 3c). During 2013–2020, the spring burning dates of bean straw in Northeast China were concentrated between early February and late April (DOY range of 30 to 120), while the autumn burning dates remained consistent with those during Period I in Heilongjiang Province (Fig. 3c). Unlike other crops, the burning dates for wheat straw did not conform to the aforementioned pattern of variation, likely due to a limited number of fire spots (Fig. 3d). The changing dispersion of burning dates for each crop type indicates shifts in agricultural practices that may be influenced by regional straw-burning ban policies, environmental conditions, and farming practices (Yang et al., 2020).

Figure 3 The daily frequency distribution of fire spots from various straw burning: (a), (b), (c), and (d) represent corn, rice, bean, and wheat straw, respectively. Note that the x axis is year, the y axis is DOY, and the range of color bars (indicating fire spots) is different for each crop, with values ranging from 1 to 1029 for corn, 1 to 615 for rice, 1 to 345 for beans, and 1 to 35 for wheat.

3.2 High-spatial-resolution annual emission inventory of GHGs

The cumulative emissions of CO₂, CH₄, and N₂O from open straw burning in Northeast China from 2001 to 2020 amounted to 198 Tg, 557 Gg, and 15.7 Gg, respectively (or 218 Tg CO₂-eq (equivalent) in total). The spatial distributions of GHG emissions correspond well with those of fire spots, particularly in high-emission areas (Figs. 2 and 4). However, the amounts of GHG emissions in the northern Songnen Plain unexpectedly exceeded those in the eastern Songnen Plain and eastern Liao River plain, suggesting that even low-intensity fire spots can generate considerable emissions of GHGs due to higher FRP detected via remote sensing. Therefore, the FRP algorithm proves to be more effective than burned-area-based algorithms in identifying emission intensity resulting from open straw burning while reducing the uncertainty associated with high-spatiotemporal-resolution emission inventories (Wu et al., 2023).

The annual emissions of CO₂, CH₄, N₂O, and CO₂-eq from 2001 to 2020 are presented in Figs. S3, S4, S5, and S6, respectively. The spatiotemporal patterns of GHG emissions correspond well to the observed trends in fire spots during Period I (2003–2012). However, during Period II (2013–2017) and Period III (2018–2020), the emissions of GHGs in the eastern Songnen Plain and eastern Liao River plain did not exhibit a proportional increase with the rise in fire spots. This discrepancy can be attributed to the dispersed burning dates among individual farmers in these regions, resulting in high-intensity fire spots with relatively low emissions. In contrast, several state farms located in the northern Sanjiang Plain and northern Songnen Plain demonstrated a higher level of synchronization in open straw-burning activities, resulting in parallel trends between fire spots and emissions (Cui et al., 2021).

Figure 4 The cumulative GHG emissions from open straw burning in Northeast China from 2001 to 2020 for CO₂ (a), CH₄ (b), N₂O (c), and CO₂-eq (d) emissions, respectively. Note that the range of color bars (indicating emissions) is different for each GHG, with values ranging from 0 to 225 Gg for CO₂, 0 to 597 Mg for CH₄, 0 to 17.4 Mg for N₂O, and 0 to 245 Gg for CO₂-eq.

During Period I (2003–2012), average annual CO₂-eq emission was at 4.20 Tg, and the cumulative CO₂-eq emission amounted to 42.0 Tg. During Period II (2013–2017), average annual CO₂-eq emission increased substantially to 26.1 Tg, and the cumulative emission during this period amounted to 130 Tg, which accounted for 59.9 % of the total emissions over the 2 decades. During Period III (2018–2020), average annual CO₂-eq emissions decreased significantly to 14.3 Tg, and the cumulative emission during this period amounted to 42.8 Tg (Fig. 5a). The trend of CO₂-eq emission from 2003 to 2020 generally corresponds with the occurrence of fire spots, except for 2015 when higher emissions were obtained despite having fewer fire spots than the case in 2014 (Fig. 5a). Such a trend is consistent with those of carbonaceous gas and aerosol (CGA) emissions estimated by Liu et al. (2022). This discrepancy between fire spots and pollutant emissions in 2015 highlights the limitations of estimating pollutant emissions based solely on burned areas (Ke et al., 2019; Wu et al. 2023). The combustion of corn and rice straw was identified as the primary contributor to CO₂-eq emissions, accounting for 51.1 % and 30.8 %, respectively, of the total emissions (Fig. 5b). Specifically, corn straw burning released 99.6, 9.06, and 2.42 Tg, while rice straw burning released 61.8, 3.78, and 1.27 Tg of CO₂, CO₂-eq for CH₄, and CO₂-eq for N₂O, respectively.

Figure 5 (a) Regional total annual CO₂-eq emissions and (b) percentage contributions from open burning of individual crop straw type.

3.3 Validation and limitations

Our estimated total CO₂ emissions from 2012 to 2020 with MODIS (161 Tg) or with VIIRS (165 Tg) were much lower than those (∼ 523 Tg) estimated by Liu et al. (2022); the latter was based on a modified FRP algorithm and fire spot products by VIIRS, which has limitations in its traditional straw extraction methods in accurately identifying fire spots during certain times of the year. Our estimated CO₂ emission from 2002 to 2020 in Northeast China (196 Tg) was slightly lower than that (195 Tg) estimated by Global Fire Emissions Database Version 4.1 (GFED4.1s) by van der Werf et al. (2017) and slightly higher than that (181 Tg) estimated by the Fire INventory from NCAR version 2.5 (FINNv2.5) by Wiedinmyer et al. (2023), which addresses the underestimation of open biomass burning in China by the older version FINNv1.5 (Stavrakou et al., 2016; Yang et al., 2020) (Fig. 6). However, our estimated total CO₂ emission from 2012 to 2020 was significantly higher than that (35.6 Tg) estimated by VIIRS-based Fire Emission Inventory version 0 (VFEIv0) by Ferrada et al. (2022), which relies on the traditional FRP algorithm (Fig. 6). Furthermore, Northeast China surpassed East China (27.1 Tg) as the highest emitter of open straw burning in China since 2014, with CO₂ emissions reaching 30.4 Tg (Zhang et al., 2020).

Figure 6 Annual total emissions of CO₂ from open straw burning (agricultural waste burning) in Northeast China from this study with MODIS (red, 2001–2020) and VIIRS (blue, 2012–2020), the Fire INventory from NCAR version 2.5 (FINNv2.5) with MODIS-only (green, 2002–2020), FINNv2.5 with MODIS and VIIRS (purple, 2012–2020), Global Fire Emissions Database Version 4.1 (GFED4.1s) (orange, 2001–2020), and VIIRS-based Fire Emission Inventory version 0 (VFEIv0) (grey, 2012–2020).

Although this study effectively improved the accuracy of emission inventory for open straw burning through the novel method that integrates crop cycle information into extraction and classification of fire spots and the modified FRP algorithm, certain limitations still exist. The uncertainty in this study stems mainly from the inherent limitations of satellite fire detection systems. The MODIS fire spot product, although widely used, is limited by its temporal resolution and tends to miss transient or small-scale fires. In addition, straw burning during non-satellite transit periods, on cloudy days, at night, and under heavy haze further exacerbates the underestimation of fire incidence, leading to potential gaps in emission inventories.

Additionally, the novel method that integrates crop cycle information into extraction and classification of fire spots presents a promising advancement. However, its applicability is constrained to regions where comprehensive and detailed crop data are available. In countries or regions lacking such agricultural information, this method may face challenges, thereby limiting its broader applicability. These factors underscore the need for continued refinement of satellite detection technologies and the expansion of agricultural data collection efforts to reduce uncertainties and enhance the robustness of emission inventories on regional to global scales.

3.4 Driving factors of open straw burning

Open straw burning is more prominently influenced by anthropogenic activities compared to other types of open biomass burning, such as forest, shrubland, and grassland fires (Syphard et al., 2017; Wu et al., 2020). Open straw burning is influenced by changes in straw yield and utilization rate, straw-burning ban policy, and farmers' awareness of straw-burning consequences (Chen et al., 2016; Li et al., 2017; Tao et al., 2018; Fang et al., 2019; Xu and You, 2023).

Northeast China has experienced a remarkable expansion in its sown area for major grain crops over the past 2 decades. By 2020, the sown area reached 231 937 km², 61.4 % more than that in 2001 (National Bureau of Statistics of China, 2002–2021). In the meantime, annual straw yield reached 143 Tg in 2020, 142 % higher than that (59.2 Tg) in 2001 (Fig. 7) (numbers are calculated based on the major grain yields in Northeast China presented in the National Bureau of Statistics of China (2002–2021) and the ratio of straw and grain (Wang et al., 2012)). Note that the annual straw yields have stabilized around 140 Tg since 2017, and this trend is expected to persist for many years to come (Fig. 7). From 2003 to 2020, a strong positive correlation was observed between the straw yields and the emissions of CO₂-eq from open straw burning across Northeast China, as well as in Heilongjiang and Jilin provinces (p<0.01, Table S2). When looking at individual periods, significant correlations were only observed during Period I (2003–2012) for the whole of Northeast China (p<0.01) and Heilongjiang (p<0.01) (Table S2). This highlights that increased straw yields exacerbated the challenges of straw disposal in Northeast China and have been a major contributor to the increase in the emissions of GHGs from open straw burning in the aforementioned region and period.

Besides open burning, crop straw is also used for cooking and heating in rural households in Northeast China (Ke et al., 2023; Liu et al., 2023). Crop straw can also be converted to bioenergy, used as animal feed, and returned to the fields (Alengebawy et al., 2022; Fang et al., 2022). However, exact quantification of straw utilization in different sectors in Northeast China is still lacking (Shi et al., 2023). Knowing that coal combustion and straw burning are major energy sources for rural households in Northeast China, we tried to explore potential changes in straw utilization on open straw burning through coal consumption changes (Fang et al., 2019). The abrupt increase in rural residential coal consumption in 2013 in Northeast China coincided with a spike in CO₂-eq emissions from open straw burning (Fig. 7a). Furthermore, a significant positive correlation between rural residential coal consumption and CO₂-eq emissions in Northeast China was revealed, especially in Heilongjiang and Jilin provinces (Table S3). We thus speculate that the increase in rural commercial energy consumption may have reduced the demand for straw as an energy source for agricultural households, thus facilitating the increased open straw burning. This needs to be confirmed in future studies once various straw utilization pathways are quantified.

Figure 7 Annual CO₂-eq emissions, yield of straw, rural residential coal consumption, and straw-burning bans in (a) Northeast China, (b) Heilongjiang, (c) Jilin, and (d) Liaoning from 2001 to 2020. Note that the range of the y axis is different for each region. The blue y axis indicates CO₂-eq emissions, with values ranging from 0 to 50 Tg for Northeast China, 0 to 35 Tg for Heilongjiang, 0 to 10 Tg for Jilin, and 0 to 6 Tg for Liaoning. The green y axis indicates yield of straw, with values ranging from 0 to 160 Tg for Northeast China, 0 to 90 Tg for Heilongjiang, 0 to 60 Tg for Jilin, and 0 to 40 Tg for Liaoning. The red y axis indicates rural residential coal consumption, with values ranging from 0 to 7 Tg for Northeast China, 0 to 2 Tg for Heilongjiang, 0 to 3 Tg for Jilin, and 0 to 4 Tg for Liaoning.

We also evaluated the efficacy of straw-burning ban policy in Heilongjiang, Jilin, and Liaoning (Table S4). Despite the implementation of the policy in 2013 in this region, a significant reduction in CO₂-eq emissions from open straw burning was only observed after 2018 (Fig. 7). Compared to the other regions of China, the effective control of open straw burning was delayed by several years in Northeast China (Huang et al., 2021). An important phenomenon was observed regarding the geographical and temporal expansion of the ban policy, for example, initially focused on key areas and specific seasons (autumn and winter) and progressively extended to the entire region and throughout the whole year (see Heilongjiang Province as an example; Table S4). Therefore, enhanced enforcement of the ban policy likely reduced CO₂-eq emissions during Period III and shifted the burning season to spring.

In conclusion, the enforcement of region-specific straw-burning bans tailored to spatiotemporal variations is crucial to control GHG emissions, given the anticipated sustained high straw yields in the future. Additionally, promoting diverse methods for utilizing straw is highlighted as an effective strategy for mitigating carbon emissions resulting from open straw burning in Northeast China. A combined effort of policy enforcement and alternative straw usage would play a pivotal role in addressing the environmental challenges posed by agricultural practices in the region.

4 Conclusions

This study provides a comprehensive analysis of the spatiotemporal variations of open straw burning across Northeast China from 2001 to 2020 and develops regional-scale high-spatial-resolution annual emission inventories of GHGs. Open straw burning in Northeast China emitted a total of 218 Tg of CO₂-eq during 2001–2020, of which 19.3 % was from Period I (2003–2012), 59.9 % from Period II (2013–2017), and 19.7 % from Period III (2018–2020). Analysis results demonstrate the necessity of integrating the crop cycle information into the extraction and classification of fire spots from open straw burning to enhance the accuracy of emission inventories of various pollutants. This study also highlights the inconsistencies between the number of fire spots and pollutant emissions caused by remote sensing detection techniques. In Northeast China, regions such as the northern Sanjiang Plain, eastern and northern Songnen Plain, and eastern Liao River plain are identified as high-emission areas of GHGs from open straw burning, which emitted 38.1, 45.5, 31.9, and 10.8 Tg of CO₂-eq, respectively, during 2001–2020. Additionally, it is observed that the season for open straw burning has shifted from autumn to spring, with dispersed burning dates. This spatiotemporal analysis provides crucial insights into policy effectiveness as well as geographical variations regarding compliance with regulations banning open straw burning. Consequently, government policies prohibiting open straw burning should be adjusted according to the observed spatiotemporal variations in different regions. Simultaneously promoting diversified applications of straw, such as bioenergy conversion, animal feeding, and soil amendment, is recommended – a strategy that is aligned with China's dual-carbon objectives aiming at achieving carbon peak and carbon neutrality.