地理研究

地理研究 ›› 2020, Vol. 39 ›› Issue (3): 735-748.doi: 10.11821/dlyj020190164

• 研究论文 • 上一篇    下一篇

基于MODIS的秦巴山地气温估算与山体效应分析

刘俊杰1,2, 潘自武1,2, 秦奋1,2,3, 顾江岩1, 朱明阳1, 赵芳1()   

  1. 1. 河南大学环境与规划学院,开封 475004
    2. 黄河中下游数字地理技术教育部重点实验室,开封 475004
    3. 河南省时空大数据产业技术研究院,郑州 450000
  • 收稿日期:2019-03-06 修回日期:2019-05-23 出版日期:2020-03-20 发布日期:2020-05-20
  • 通讯作者: 赵芳
  • 作者简介:刘俊杰（1995-）,女,河南周口人,硕士,主要从事山地GIS、山地地理研究。E-mail: junjieliu555@163.com
  • 基金资助:
    国家自然科学基金项目(41601091);中国南北过渡带综合科学考察项目(2017FY100900);国家科技平台建设项目(2005DKA32300)

Estimation of air temperature based on MODIS and analysis of mass elevation effect in the Qinling-Daba Mountains

LIU Junjie1,2, PAN Ziwu1,2, QIN Fen1,2,3, GU Jiangyan1, ZHU Mingyang1, ZHAO Fang1()   

  1. 1. College of Environment and Planning, Henan University, Kaifeng 475004, Henan, China
    2. Key Laboratory of Geospatial Technology for the Middle and Lower Yellow River Regions, Kaifeng 475004, Henan, China
    3. Henan Industrial Technology Academy of Spatio-Temporal Big Data, Henan University, Zhengzhou 450000, China
  • Received:2019-03-06 Revised:2019-05-23 Online:2020-03-20 Published:2020-05-20
  • Contact: ZHAO Fang

RichHTML

PDF (PC)

赞

摘要：

秦巴山地作为横亘在中国南北过渡带的巨大山脉,其山体效应对中国中部植被和气候的非地带性分布产生了重要的影响,山体内外同海拔的温差是表征山体效应大小较为理想的指标。本研究结合MODIS地表温度（LST）数据、STRM-1 DEM数据和秦巴山地的118个气象站点的观测数据,分别采用普通线性回归（OLS）和地理加权回归（GWR）两种分析方法对秦巴山地的气温进行估算,在此基础上将秦巴山地各月气温转换为同海拔（1500 m,秦巴山地平均海拔）气温,对比分析秦巴山地的山体效应。结果表明：① 相比OLS分析,GWR分析方法的精度更高,各月回归模型的R 2均在0.89以上,均方根误差（RMSE）在0.68~0.98 ℃之间。② 利用GWR估算得到的同海拔气温,从东向西随海拔升高呈现了明显的升高的趋势,秦岭西部山地比东段升高约6 ℃和4.5 ℃;大巴山西部山地年均和7月份同海拔的气温较东段升高约8 ℃和5 ℃。③ 从南向北,以汉江为分界,秦岭与大巴山的同海拔的气温均呈现出由山体边缘向内部升高的趋势。④ 秦巴山地西部大起伏高山,秦岭大起伏高中山和大巴山大起伏中山,相比豫西汉中中山谷地,各月均同海拔气温分别升高了约3.85~9.28 ℃、1.49~3.34 ℃和0.43~3.05 ℃,平均温差约为3.50 ℃,说明秦巴山地大起伏中高山的山体效应十分明显。

关键词: 山体效应, 秦巴山地, MODIS地表温度, 气温估算, 地理回归加权

Abstract:

As a huge mountain range in the North-South boundary of China, the Qinling-Daba Mountains are characterized by prominent mass elevation effect (MEE) and play an important role in the azonality pattern of climate and ecology in central China. The essence of the MEE is the warming effect of mountains, as huge mountain and plateau absorbs more solar radiation compared with the free atmosphere of the same altitude and then releases in the form of long-wave radiation external heat, making the internal mountain temperature higher than the external in the same altitude of free atmosphere. Therefore, the temperature difference between the mountain interior and the periphery has been suggested as an appropriate indicator to quantify the MEE. To analyze MEE of the Qinling-Daba Mountains, MODIS land surface temperature (LST) data, STRM-1 DEM data and observation data from 118 meteorological stations were combined to estimate monthly mean air temperature by ordinary linear regression (OLS) and geographical weighted regression (GWR) methods in the Qinling-Daba Mountains. Air temperature at an altitude of 1500 m (the average elevation of the Qinling-Daba Mountains) in the interior of the Qinling-Daba Mountains was calculated by a fixed lapse rate and compared with that in the periphery. The results show that: (1) Compared with OLS method, the GWR method has higher accuracy with R 2 > 0.89 and the root mean squared error (RMSE) = 0.68-0.98 ℃. (2) The monthly mean temperature at the altitude of 1500 m estimated by GWR presents a gradual upward trend from east to west. In the western Qinling Mountains, the annual average temperature and temperature in July at the altitude of 1500 m increase about 6 ℃ and 4.5 ℃ compared with the eastern flank, while in the Daba Mountains, they are about 8°C and 5 ℃ higher in the west than in the east. (3) From south to north, with the Hanjiang River as the boundary, the monthly mean temperature at the altitude of 1500 m tends to rise from the rim of the mountains to the ridge. (4) Compared with the lower valleys in Hanzhong and western Henan, the monthly mean temperatures at the altitude of 1500 m are approximately 3.85-9.28 ℃, 1.49-3.34 ℃ and 0.43-3.05 ℃ higher those in the great undulating high mountains in the western Qinling-Daba Mountains, the great undulating middle-high mountains in the Qinling Mountains and the great undulating middle mountains in the Daba Mountains, respectively, and the average temperature difference is about 3.50 ℃. This shows that the MEE of the Qinling-Daba Mountains is obvious and its impact on the distribution patterns of mountain climate and ecology needs to be further studied.

Key words: mass elevation effect, Qinling-Daba Mountains, MODIS land surface temperature, temperature estimation, geographical weighted regression