Contributions of open crop straw burning emissions to PM2.5 concentrations in China

PM2.5 inventories have been developed in major Chinese cities to quantify the contributions from various sources based on annual emissions. This approach, however, could substantially underestimate the contribution from open straw burning during the harvest or other active burning periods. This study examines this issue by estimating monthly and annual straw-burning PM2.5 emissions in China and comparing with them with the corresponding emissions from other anthropogenic sources. Annually burned straw PM2.5 emissions during 1997 ∼ 2013 for 31 China provinces were calculated based on crop and related burning information for 12 months based on satellite detection of agricultural burning. Annual emissions from other anthropogenic sources were collected from the literature and allocated to monthly values using air pollution index measurements. The results indicate that the annual PM2.5 emissions from open straw burning in China were 1.036 m tons. The monthly PM2.5 emission ratios of straw burning to other anthropogenic sources during June, the harvest period for many regions, were several times larger than the annual ratios at national, regional, and province levels, suggesting that, in contrast to annual emissions that were used in the PM2.5 inventories in Chinese cities to assess the contributions from other sources, monthly emissions should be used to assess the contributions from straw burning during the harvest or other active burning periods. The larger contributions from straw burning shown in this study also suggest that substantial reduction of open field straw burning would dramatically improve air quality in many Chinese regions during the harvest or other active burning periods.


Introduction
China is among the major agricultural nations in the world. Crop straw resources in China, ranking the first in the world, account for 17.29% of the global production (Bi et al 2010). Corn, wheat, and rice straws are the major crop straw resources in China. Farmers need to remove a large amount of crop straws in a short period of time after the harvest in order to plant crops for the next growing season. In comparison with other approaches such as cutting, burning is the most effective and is less expensive to remove straws.
However, straw burning has many adverse environmental and ecological impacts (Shi et al 2014).
Straw burning wastes valuable biological resources and, more importantly, often causes serious air pollution. Straw burning releases a large amount of pollutants such as PM 2.5 , SO 2 , CO, NH 3 , VOC, and NO X (Koppmann et al 2005, Li et al 2008. Further, straw burning is one of the primary contributors to haze and smog formation during the harvest periods in China (Zha et al 2013).
Air pollution is a serious environmental problem in China. Regional haze and smog events have dramatically increased in recent years, due to the rapid development of the economy since the late 1970s. This situation has worsened in recent decades due to dramatic urban expansion and an in increase in automobile use in China . For example, there were 124 recorded haze and smog days in Beijing in 2012, 25 of which occurred in January (13 more than the monthly average over previous years), and 18 in June (three times the monthly average over the previous 10 years) (http://beijing.tianqi.com/ news/19519.html).
Reversing these declining air quality trends and eventually solving the air pollution problem is one of the fundamental goals of China today. Many research projects have been initiated to assess the intensity of air pollution and to identify the major contributors. China started monitoring PM 10 , SO 2 , and NO 2 as early as 2000 in more than 40 major cities to formulate the air pollution index (API). In 2012, China added PM 2.5 , the most important contributor to smog and haze and one of the major air pollution indicators used by other countries, to the monitoring list together with O 3 and CO to formulate a new air pollution index (i.e., the air quality index (AQI)).
A recent major effort has been the development of PM 2.5 inventories and the identification of major emission sources based on the PM 2.5 data together with various analysis tools. This task was accomplished for nine cities in the three most industrialized cities in Beijing-Tianjin-Hebei, the Yangtze River Delta, and the Pearl River Delta in 2014 and 26 more cities will be added to the list in 2015. The inventories are essential for air pollutant control, prediction, and prevention.
The PM 2.5 particles come from both local and transported sources. In Beijing, for example, they accounted for 64%∼72% and 28%∼36%, respectively. The local sources were mainly automobiles (31.1%), coal consumption (22.4%), industry (18.1%), construction dust (14.3%), and other sections (14.1%). The percentage of transported sources were not specified, but smoke particles from cropstraw burning were expected to be one of the major components for Beijing , Wang et al 2014.
Straw-burning emissions in China have been estimated for 2000∼2003   Wang and Zhang 2008, Bi et al 2009, Zuo et al 2015. According to Guan et al. (2014) and , annual biomass burning contributed about 26% of total particulate matter (PM) emissions, half of which were from open-field straw burning. These results provide useful information to assess how important straw-burning emissions are as a transported source to the PM 2.5 inventories in China cities.
However, straw burning is essentially a seasonal activity and it peaks with crop harvests. The harvest periods are mostly in June-July and around October in southern China and late May-early June, and late September-early October in northern China . This strong seasonal dependence suggests that the PM 2.5 emission ratios of straw burning to other anthropogenic sources during the harvest periods are much larger than the annual ratios. Thus, the current approach to developing PM 2.5 inventories in China cities using annual values underestimates the contributions of straw-burning emissions during the harvest periods. For the PM 2.5 inventory in Beijing, the contributions from the transported sources were more than 50% during severe air pollution periods, 15%∼20% more than the annual values. This is a common feature of biomass burning.
The aim of this study was to examine the above issue by estimating and analyzing the monthly PM 2.5 emissions from open straw burning during 1997∼2013 in China and comparing the emission ratios of straw burning to other anthropogenic sources. The results are expected to provide useful information for improving our understanding of the roles of straw-burning emissions in air pollution formation and for better assessing the contributions of transported sources in the PM 2.5 inventories in Chinese cities.

Methods
The methods used in this study are briefly described as follows. An illustration of the calculation procedure (figure S1) with a detailed description of emission estimation is provided in the supplemental material.

Emissions
Annual emissions of open straw burning were obtained using (Lu et al 2011) here M i is emission of ith straw-burning pollutant, m crop species, P annual crop yield, N yield-to-straw ratio, R r portion of open straw burning in region r, F combustion efficiency, and EF i emission factor of the pollutant.
To estimate emissions from open straw burning, the annual crop yields of 31 China provinces during 1997∼2013 (Chinese Statistical Yearbook), the yield-to-straw rates from Bi et al (2009) and Zuo et al (2015), the straw-burning proportions from Wang and Zhang (2008), and the average combustion efficiency from Zhang et al (2008) were used. The emission factors of the straw burning were provided in many studies (e.g. , Zhang et al 2015, Zhu et al 2005. This study used the results from Ni et al (2005) and Li et al (2007) with the PM 2.5 emission factors of 8.5, 9.5, 12.0, and 9.5 g kg −1 for rice, wheat, corn, and other crops, respectively. An average value of 9.65 g kg −1 was also used without distinguishing crop types. In addition, PM 10 emissions were calculated using an emission factor of 11.7 g kg −1 based on Li et al (2007)  The data combine MODIS information on fire activity and vegetation productivity to estimate gridded monthly burned area and fire emissions. The RS detection has a spatial resolution of 0.25 degrees and is available from 1997 and onwards. Note that GFED4.0s includes an algorithm to detect small fires that is helpful to improve the accuracy of the estimation of agricultural straw burning. However, straw burning in China is scattered in residential areas with small burn sizes. In addition, the MODIS detections during the day are available around 10 am and 1 pm local time. The flaming stage of straw burning may be last shorter than the interval between the two detection times. Thus, it is likely that the GFED data could have missed many small burns in China. Our comparison indicated that the detected emissions were much smaller than those obtained using crop yield data (figure S2). Nonetheless, little impact on the detection of straw burning season is expected , Li et al 2009). Thus, we estimated straw-burning emissions based on crop data while using the GFED data for monthly allocation.

Air pollution index
Monthly means of API in 42 key cities were obtained based on daily values (http://www.mep.gov.cn) and averaged over 2001∼2012. They were used to allocate the monthly variations of PM emissions from other anthropogenic sources. The API values are divided into five grades of 0∼50, 51∼100, 101∼200, 201∼300, and 301∼500 with increasing severity of air pollution. The annual PM 2.5 emission from open straw burning in China was 1.036 m tons, calculated using one single emission factor for all kinds of crops, which was very close to the value of 1.044 m tons calculated using three different emission factors for cereal crops and one emission factor for other crops. The emission value is comparable to the estimate by , but about half of the estimate by Wang and Zhang (2008), who used much larger yield-to-straw rate, proportion of burning, burning efficiency, and emission factors. Consistent with the amounts of burned straw shown in figure 1(a), larger PM emissions from straw burning occurred in East, Central, and North China (figures 1(b)) and 2(a)). The PM 2.5 emissions from Jiangsu and Anhui in East China reached 107.04 and 99.64 k tons, respectively (1 k ton=1 Gg=10 9 g). In comparison, the annual PM 10 emission was 1.256 m tons nationally, and 129.77 and 120.80 k tons, respectively, for Jiangsu and Anhui. This spatial distribution is consistent with other studies (e.g., Lu et al 2011).
The PM 2.5 emissions from other anthropogenic sources occurred mainly in North and East China, as well as in Sichuan of Southwest China and Guangdong of South China ( figure 1(b)). These regions have high population density and more advanced industry and economy than other regions. In contrast, PM 2.5 emissions in the less industrialized areas of Xizang, Qinghai, Ningxia of Southwest and Northwest China, and the tourist-oriented Hainan of South China were very low. In addition, emissions were low in the four municipalities of Beijing, Tianjin, Shanghai, and Chongqing. The largest PM 2.5 emission of about 1200 k tons occurred in Shandong of North China (figure 2(b)), followed by about 800∼1000 k tons in Hebei, Jiangsu, Henan, and Sichuan. Different from PM 2.5 emissions from straw burning, all regions except Northwest China had one or more provinces with considerably large emissions from anthropogenic sources. Note that the proportion rate for Hunan is 32.9% (district 6), which is nearly two times larger than the rate of 10.7% for Hubei (district 1) (table S2). Thus, despite having very similar straw yields, the amount of burned straw is about two times larger in Hunan than in Hubei. Also note that the emissions in Beijing, Shanghai, and Tianjin are relatively small. There are two reasons: (1) The emissions for each area is the total amount rather than the amount per unit. All three cities are extensively urbanized with little farming land. In other regions, the maximum emissions occurred in spring and/or fall months, with much smaller values of about 25 k tons. The large emissions in June and during the fall months were a result of the summer and fall harvest seasons, respectively. Some areas did not burn straw immediately after the fall harvest, but waited until the next spring planting season. This caused large emissions in some spring months.

Monthly PM 2.5 emissions
The monthly API looked different than the monthly straw-burning emissions. Only one pattern was found for all regions (figure S3), with the worst air pollution (the highest API) found in winter, the best in summer, and a second peak in spring. Heating using coal is one of the major air pollution sources in China, mostly in the winter. In addition, the atmosphere is more stable and rainfall is the lowest in winter but the opposite is true in summer. Dust storms are a major environmental hazard, mainly in northern China during spring. These natural and human activities explain the seasonal API pattern. The main seasonal cycle is consistent with Zhang et al (2009) who estimated the monthly variations based on the monthly distributions of energy consumption of residents according to temperature and other energy and industrial emissions according to power generation, cement production, and gross domestic product.
Consistent with the seasonal API variations, larger emissions from other anthropogenic sources in China occurred in the winter months of January and December with PM 2.5 of about 1.303 and 1.325 m tons, respectively. In the spring months of March and April, PM 2.5 emissions were about 1.207 and 1.181 m tons, respectively. At the regional level, the largest monthly variations occurred in North China, followed by East China, and relatively small variations were seen in Northwest, South, and Northeast China ( figure 3(b)).

PM 2.5 emission ratios
The annual PM 2.5 emission ratio of open crop straw burning averaged over 1997∼2013 to other anthropogenic sources for the year of 2006 was 7.8% at the national level (table 1). The monthly PM 2.5 emission ratios were 26% in June, 11% in May, 9.7% in October, and 8% in March and April. The June ratio was about 3.4 times the annual ratio. At the regional level, the annual ratios were about 10% for Central, East, and South China, and as low as about 5% for Southwest and North China. The June ratios were about 56% in East China and 30% in North and Central China. The October ratios were about 29% in Northwest China and 22% in Northeast China ( figure 4 and table 1). The monthly ratios were about 3∼5 times the annual ratios.
At the province level, the annual emission ratios were the largest at nearly 35% in Xizang of Southwest China, over 20% in Hunan of Central China, over 15% in Guangxi of South China, Anhui of East China, and Xinjiang of Northwest China, and below 10% in 21 out of the 31 provinces ( figure S4; also see figure S5 (a) for the geographic distribution). The June ratios (figures S4 and S5(b); table 1) were about 107% in Anhui and 81% in Jiangsu of East China, 63% in Henan of Central China, 47% in Shaanxi of Northwest China, and 45% in Shandong of North China, which were about 6∼7 times the corresponding annual ratios. The other months with relatively large ratios were about 33% in Guangxi of South China and 41%  (table S3).
Straw-burning emissions had some interannual variability (figure S6). To examine the possible impact of the variability on the emission ratios, the PM 2.5 emission ratios using open crop straw burning for the year 2006 only instead of the average over 1997∼2013 were calculated. The results are the same as those described above, with slightly larger magnitude ( figure S7 and table S4).

Discussion
The findings from this study, the PM 2.5 emission ratios of straw burning compared to other anthropogenic sources, may be valuable for the development of the PM 2.5 inventories in China cities. First, it provides new evidence for the important contribution of strawburning emissions to air pollution in China, suggesting that the PM 2.5 inventories should include straw burning as an emission source. Previous studies have provided this evidence based on burning information of single or several years, while this study analyzed the long-term burning information of 17 years and thus is more meaningful from a statistical point of view.
Secondly, this study further indicates that the contributions of straw-burning emissions during harvest or other active burning periods are much larger than the contributions from previous assessments, which were based on annual emissions of straw burning and other anthropogenic sources. This suggests that the  PM 2.5 inventories should use monthly rather than annual straw-burning emissions while assessing its contribution as a transported source. The importance of more detailed wildland fire information than the annual data from the US EPA national emission inventories for improving air quality simulations is indicated (e.g., Battye and Battye 2002). This study provides evidence for this by showing the large difference between monthly and annual contributions of straw-burning PM 2.5 emissions to other anthropogenic emissions in China. Thirdly, this study identifies the specific time of a year (June, October, March, etc) for a region or province when the monthly contribution from strawburning emissions was the largest. Thus, the PM 2.5 inventories for a specific city could focus on the corresponding time of year to determine the strawburning contribution. Dynamic models such as CMAQ (Byun and Schere 2006) and HYSPLIT (Draxler and Hess 1998, Zhu et al 2010) are needed to simulate transport of smoke particles from the rural burning sites to urban area where the impact of smoke on the air quality conditions are assessed , Su et al 2012. Simulations of such models requires a large amount of computational resources, while much less is needed for simulations if a specific harvest or other active burning period is used instead of an entire year.
The June peak in the major straw-burning regions of China is consistent with the finding from a global MODIS detection over 2001∼2003 for this region (Korontzi et al 2006). The peak times, however, are different in other regions, as shown in that study. The peaks occur during April to May and August for the global average. North America has a similar two-peak pattern during March to June and September to November. One peak pattern is dominant in other continents during February and March in South and Southeast Asia, April and May in Central and Northeast Asia and Central America, July and August in Europe, and November and December in North Africa. In the southern hemisphere, the peak occurs during March and April in Australia and New Zealand, July to September in South Africa, and August to October in South America. The amount of PM 2.5 emissions from agricultural burning in the continental United States detected using MODIS from 2003∼2007 (McCarty 2011) is much smaller than in China, probably because of the extensive use of powerful mechanical chopping. The largest emissions occur mainly in several regions of the Southeast, the Great Plains, and the Pacific Northwest.
There are alternatives to burning for removing crop straw residues in China. Straw can be chopped and directly decomposed in soil or composted outside fields as nutrients and organic materials. However, farmers in China usually use simple tools that can't chop straw fine enough to be well decomposed and absorbed by soil before the next crop season. Using power generation has been explored but has largely failed due to the low efficiency and high costs of straw cutting and transportation. Crop straws were once used as domestic fuel for some farmers. However, this is no longer a practice for the majority of famers due to extensive usage of electric power and natural gas in rural areas in recent years. The Chinese government has banned crop straw burning in most regions in an effort to reduce air pollution but has compensated farmers for the economic burdens of using alternatives to burning. However, this policy is unlikely to be a long-term solution.

Conclusions
It can be concluded from the findings of this study that the contributions of straw-burning PM 2.5 emissions during harvest or other active burning periods to air pollutions in China are substantially larger than previous assessments based on annual emissions of straw burning and other anthropogenic sources in the major burning provinces. China recently identified main urban air pollutant sources from transportation, energy, industrial, and construction dust based on annual emissions. For sources such as crop straw burning, monthly emissions should be used to estimate their contributions to air pollutants. Otherwise, the contributions will be significantly underestimated for certain periods of a year.
The larger contribution of straw burning to air pollution in China for harvest or other active burning periods suggests that a substantial reduction or complete abandonment of open-field straw burning would dramatically improve air quality in many China regions during these periods.