Assessments of the impacts of land use and land cover changes(LUCC) on the terrestrial carbon budget, atmospheric CO2 concentration, and CO2-related climatic change are important to understand the environmental effects of LUCC and provide information about the effects of historical carbon emissions. Using regional land cover reconstructions from historical records, with a bookkeeping model, we estimated the carbon sink changes caused by historical cropland expansion in Northeast China during the past 300 years. The conclusions are as follows:(1) There was a dramatic land reclamation of cropland during the past 300 years in Northeast China. Approximately 26% of the natural land was cultivated, and 38% of the grassland and 20% of the forest and shrubland were converted to cropland.(2) The carbon emission induced by cropland expansion between 1683 and 1980 was 1.06–2.55 Pg C, and the estimation from the moderate scenario was 1.45 Pg C. The carbon emissions of the soil carbon pool was larger than that from the vegetation carbon pool and comprised more than 2/3 of the total carbon emissions.(3) The carbon emissions of the three provinces in Northeast China were different. Heilongjiang Province had the largest carbon emissions, and Jilin Province had the second largest emissions.(4) The primary source of carbon emissions was forest reclamation(taking 60% of the total emissions in the moderate scenario), the secondary source was grassland cultivation(taking 27%), and the tertiary sources were shrubland and wetland reclamation(taking 13%). Examination on the data accuracy revealed that the high-resolution regional land cover data allowed the carbon budget to be evaluated at the county level and improved the precision of the results. The carbon emission estimation in this study was lower than those in previous studies because of the improved land use data quality and various types of land use change considered.
历史土地利用数据集是解释过去人类活动及其影响、定量评估土地利用变化对气候变化和碳循环影响、检验模型模拟结果的基础.SAGE("全球土地利用数据集")和HYDE("全球历史环境数据集")是国际上已被广泛应用的代表性的全球土地利用数据集,但其数据的准确性并未得到充分地评估.CNEC数据(Cropland data of Northeast China)重建的是东北三省过去300年耕地覆盖变化数据,该数据所利用的数据源为历史文献和调查、统计资料,并经过订正和归一化处理,能够较好地反映耕地变化的实际情况.以CNEC数据为参照,分别从东北地区、分省和网格3个空间尺度对SAGE和HYDE数据集中东北地区的耕地数据精度进行了评估.结果表明,SAGE和HYDE数据集中东北地区的耕地重建结果无论总量还是空间分布都存在显著的误差,其反映的耕地面积增长与空间变化过程也与研究区历史开垦过程相左.其中,SAGE数据集对东北历史耕地数据的估计普遍偏高,并且没有反映出东北地区耕地面积阶段性增长的特点;HYDE数据集的耕地总量虽然具有一定可比性,但是年代际变化和各时间断面空间分布与历史真实情形不符.全球数据集在东北地区的重建误差可能来源于两方面:一是依据现代耕地分布格局回推历史耕地分布的方法不能反映现实开垦进程,二是遥感资料提取的现代耕地数据存在误差。
