论文总字数：27411字

摘 要

降水是组成地球水循环的最重要部分，它也是组成世界的水和能量循环过程中基本要素之一，精确熟悉降水信息有利于充分研究大气循环和水循环的过程。同时在新疆省，尤其是北疆地区许多河流的源头大部分来自于天山积雪覆盖区，而新疆干旱区农业极其依赖积雪融水所提供的水资源,该地区降雨变化引起的降雪变化会影响河道在春、夏季的径流情况，不利于干旱区经济的发展和生态环境的保护，甚至会引发该地区干旱与洪涝等自然灾害频发。因此，研究天山地区降水情况有利于当地的农、牧业、经济的可持续发展。本文以天山北坡区域为研究区域，利用多个统计指标对CMFD产品数据进行精度评价，验证其适用性，并以天山的DEM为基础，提取其地形因子，结合ArcGIS空间分析和空间插值法，旨在探索天山冰雪区域的降雨特征，以及研究降雨和积雪的时空相关性。

结果表明：（1）CMFD产品逐月、逐季降水量与地面实际降水量线性拟合相关系数分别为R=0.932和0.841、0.903、0.822、0.780，拟合精度较高。说明两者呈现较好的线性相关性，即逐月、逐季产品均具有较高精度，可以适用于天山北坡体现降雨量的年内和季节内时空变化特征；逐年降水量产品的相关系数略低，但仍相对准确，可以用于研究区年际变化特征分析。同时，随时间尺度增大，CMFD数据与地面观测降水数据的相关系数减小，精度降低。（2）天山北坡降水分布不均匀，总趋势为东段少于西段，南部低于北部，西段内部高于外缘；终年积雪的高海拔区域降水多。年际间降水变动幅度不大，保持稳定，但变化趋势明显的地区是降水丰富和缺水的区域。降水量与积雪面积在融雪期内表现负相关，即降水有略增加，但积雪减少的趋势。（3）天山北坡坡向变化复杂、坡向以北为主，降水受经纬度、海拔影响较大，随海拔增大先增大后减小；受坡度、坡向等影响较小，降水随坡度增加而增大，在坡向上分布复杂。

关键词：天山北坡 CMFD降水再分析数据 精度评估 时空变化特征分析

Accuracy Evaluation and Spatio-temporal variation of CMFD precipitation reanalysis data in the North Slope of Tianshan Mountain

ABSTRACT

Precipitation is the most important part of the earth's water cycle, and it is also one of the basic elements of the water and energy cycle that make up the world.Precise familiarity with precipitation information is conducive to the full study of the atmospheric cycle and water cycle. At the same time, the sources of many rivers in Xinjiang Province, especially in northern Xinjiang, are mostly from the Tianshan snow-covered area, and agriculture in the arid area of Xinjiang is extremely dependent on the water resources provided by snowmelt and meltwater. Affecting the runoff of rivers in spring and summer is not conducive to the economic development and ecological environment protection of arid areas, and even triggers frequent natural disasters such as droughts and floods in the area.This article uses multiple statistical indicators to evaluate the accuracy of the CMFD product data on the northern slope of the mountain area and verify its applicability. Based on the DEM of Tianshan, it extracts its topographic factors and combines ArcGIS spatial analysis and spatial interpolation methods The aim is to explore the characteristics of rainfall in the icy and snowy areas of the Tianshan Mountains, and to study the spatial and temporal correlation of rainfall and snow.

The results showed that：(1) The monthly and seasonal precipitation of CMFD products and the actual ground precipitation were R = 0.932, 0.841, 0.903, 0.822 and 0.780, respectively, and the fitting precision is high. It shows that the two have a good linear correlation, that is, the monthly and seasonal products have higher accuracy, which can be applied to the northern and southern slopes of the Tianshan Mountains to reflect the characteristics of the annual and intra-seasonal changes in rainfall; the correlation coefficient of the annual precipitation products is slightly Low, but still relatively accurate, and can be used to analyze the characteristics of inter-annual changes in the study area. At the same time, as the time scale increases, the correlation coefficient between CMFD data and surface observation precipitation data decreases, and the accuracy decreases.

(2)The precipitation is unevenly distributed.The general trend is that the eastern section is less than the western section, the southern section is lower than the northern section, and the internal section is higher than the outer edge; Inter-annual precipitation has not changed much and remained stable, but the areas where the change trend is obvious are areas with abundant precipitation and lack of water. The amount of precipitation is negatively correlated with the area of snow cover during the snowmelt period, that is, the precipitation slightly increases, but the snow cover decreases.

(3)Precipitation is greatly affected by latitude, longitude, and altitude, and increases first and then decreases as the altitude increases. The slope increases and increases, and the distribution on the slope is complicated.

Key words：Tianshan Mountains；CMFD precipitation reanalysis data；Precision evaluation；Analysis of the characteristics of spatio-temporal variation

目录

摘要 I

ABSTRACT II

目录 IV

第一章 引言 1

1.1选题背景与依据 1

1.1.1选题背景 1

1.1.2选题依据 1

1.2国内外研究进展 2

1.2.1 降水产品介绍 2

1.2.2 研究处理方法 3

1.2.2.1不同降水产品处理方法 3

1.2.2.2 不确定性评估方法及应用 3

1.2.2.3 针对TRMM数据的研究方法 3

1.2.3 针对山区降水现有研究的分析结果 4

1.2.4 研究切入点 4

1.3 研究内容与技术路线 5

1.3.1研究内容 5

1.3.2 技术路线 5

第二章 研究区与数据资料 7

2.1 研究区概况 7

2.1.1 自然地理概况 7

2.1.2 积雪分布情况 7

2.2数据资料 8

第三章 CMFD降水数据的多尺度精度检验 11

3.1 降水数据空间插值 11

3.2精度检验方法 12

3.2.1相关系数 (R) 12

3.2.2相对偏差（BIAS） 12

3.2.3均方根误差（RMSE） 12

3.3.4散点斜率法 13

3.3多时空尺度降雨量精度检验结果 13

3.3.1 月尺度降水量精度检验 13

3.3.2 季尺度降水量精度检验 14

3.3.3 年尺度降水量精度检验 15

3.4 CMFD降水数据在天山北坡的适用性分析 16

第四章 CMFD降雨量的时空分布特征 18

4.1降雨量空间分布特征 18

4.1.1逐年年均降水量的空间分布 18

4.1.2 天山北坡降水的海拔变化分析 19

4.1.3 降水的坡度、坡向的变化分析 20

请支付后下载全文，论文总字数：27411字