CAJ | 학술논문

草原是干旱区生态系统中重要的可再生资源。本文基于草本植被的结构特征，利用ASAR和TM数据，结合MIMICS模型，提出了一种估算干旱区草原地上植被生物量的方法。该方法将光学遥感数据容易反演的叶面积指数（LAI)作为反演生物量模型的参数之一，并利用LAI成功估算了单位面积内的草本植被密度。将地上生物量作为输入变量代入改进的MIMICS模型，利用查找表方法，计算出地上植被生物量。然后，将该方法应用于乌图美仁草原的地上植被生物量的反演。结果表明，该方法能够成功地反演干旱区草原草本植被地上生物量，精度达到R2=0.8562，RMSD=0.6263。最后，分析了该方法估算植被生物量的误差来源。
초원시간한구생태계통중중요적가재생자원。본문기우초본식피적결구특정，이용ASAR화TM수거，결합MIMICS모형，제출료일충고산간한구초원지상식피생물량적방법。해방법장광학요감수거용역반연적협면적지수（LAI)작위반연생물량모형적삼수지일，병이용LAI성공고산료단위면적내적초본식피밀도。장지상생물량작위수입변량대입개진적MIMICS모형，이용사조표방법，계산출지상식피생물량。연후，장해방법응용우오도미인초원적지상식피생물량적반연。결과표명，해방법능구성공지반연간한구초원초본식피지상생물량，정도체도R2=0.8562，RMSD=0.6263。최후，분석료해방법고산식피생물량적오차래원。
Grasslands are important renewable resources in ecosystems of arid areas. As one of the important components of prairie ecosystems, aboveground biomass is the key indicator of health status of the prairie eco-systems. Comparing to the significant limitations of the traditional method of biomass, the satellite remote sens-ing provides a unique effective and efficient means in biomass monitoring and assessment. In this paper, we pro-posed a retrieval methodology for herbaceous vegetation biomass based on the vegetation structure characteristic and Michigan Microwave Canopy Scattering (MIMICS) model using the ASAR and TM data. A two-layer cano-py reflectance model (ACRM) was used to inverse the Leaf Area Index (LAI) which can be easily retrieved from the optical remote sensing data. Then LAI was used to get the number of plants per unit area. The aboveground biomass served as an input parameter was input into the adaption of MIMICS model which adopted for character-izing the backscatter from herbaceous vegetation by deleting the scatter component associated with ground-trunks. Then, based on established equations which used the dual-polarized radar data, the aboveground biomass was calculated using the lookup table. The method was applied to retrieve the biomass of Wutumeiren prairie, Qinghai Province. Results indicated that the method was of the operational potential in aboveground bio-mass of the herbaceous vegetation in arid region. And a good accuracy of the biomass retrieval was achieved (R2=0.8562, RMSD=0.6263). Finally, we analyzed the error sources of biomass estimation using this method. The sources of error might be come from two aspects, i.e. error of the input model parameters, and the ill-posed inversion problem, for which, the mean value was used as the inverse results when the solution is not unique.