冰川冻土, 2021, 43(2): 650-661 doi: 10.7522/j.issn.1000-0240.2021.0049

冰冻圈技术

基于多源遥感数据的疏勒河上游山区流域VIC-CAS模型积雪模拟效果评估

郭佳锴,1,2, 李哲1,2, 李飞1,2, 张世强,1,2

1.西北大学 陕西省地表系统与环境承载力重点实验室,陕西 西安 710127

2.西北大学 城市与环境学院,陕西 西安 710127

Evaluation on snow coverage and snow depth simulated by VIC-CAS model based on multi-source remote sensing data in mountainous upper reach of the Shule River basin

GUO Jiakai,1,2, LI Zhe1,2, LI Fei1,2, ZHANG Shiqiang,1,2

1.Shaanxi Key Laboratory of Earth Surface System and Environmental Carrying Capacity,Northwest University,Xi’an 710127,China

2.College of Urban and Environmental Sciences,Northwest University,Xi’an 710127,China

通讯作者: 张世强,教授,主要从事寒区流域各水循环要素的观测、遥感反演与模拟研究. E-mail: zhangsq@lzb.ac.cn

编委: 武俊杰

收稿日期: 2020-10-03   修回日期: 2021-04-09  

基金资助: 国家自然科学基金项目.  41671056.  41730751

Received: 2020-10-03   Revised: 2021-04-09  

作者简介 About authors

郭佳锴,硕士研究生,主要从事冰冻圈积雪研究.E-mail:gjk18561723579@163.com , E-mail:gjk18561723579@163.com

摘要

积雪积累和消融过程是冰冻圈水文模型的重要组成部分,利用多源遥感数据对水文模型模拟的积雪分布和深度进行评估是进一步增强融雪过程模拟的物理基础,也是提高模拟可靠性的重要手段。基于2002—2013年疏勒河上游山区流域MODIS地表反射率数据集和中国雪深长时间序列数据集,对VIC-CAS模型模拟的逐日积雪覆盖度和雪深进行了综合评估。结果表明:从不同降雪年份来看,VIC-CAS模型可以较好地模拟多雪年(2008年)疏勒河上游山区流域积雪的覆盖度,在平雪年(2004年)和少雪年(2013年)模型模拟精度相对较低。从不同海拔的模拟结果来看,在流域占比最高的4 000~5 000 m高程带精度最高,2 000~3 000 m高程带精度最低;对比模拟雪深与中国雪深产品发现,多雪年的一致性较高,平雪年和少雪年的一致性较低。这表明VIC-CAS模型对疏勒河上游日尺度的积雪覆盖度和雪深的模拟精度相对较低,特别在低海拔处和薄雪情况下,其原因可能是对积雪再分布和风吹雪过程的模拟算法和参数化存在较大的不确定性,需要进一步改进。

关键词: VIC-CAS模型 ; 积雪覆盖度 ; 雪深 ; 精度评价 ; 疏勒河上游山区流域

Abstract

The accumulation and melting processes of snow are the important parts of the cryospheric hydrological model. Generally, the simulated snow cover and snow depth time-series by distributed hydrological model were difficult to evaluate, which mainly calibrated and validated by observed runoff data. The multi-source remote sensing snow cover and snow depth products give a good choice for evaluating the spatial-temporal patterns of snow cover and snow depth of simulation, which probably help to enhance the physical basement of model. However, due to the cryospheric components include glacier, snow cover, permafrost always distributed together in one basin, the evaluation on snow cover in one basin should base on that there are enough precipitation observation data in alpine areas of the basin, and glacier meltwater were reasonable simulated, which indicated that the large uncertainties of simulated cover were removed. There are more than 20 precipitation observation instruments were installed since 2009 in the upper reach of the Shule River basin (URSRB) with average elevation above 4 000 m a.s.l., and the annual precipitation gradient were obtained with 14.654 mm·(100m)-1. The river runoff of Changmabao in URSRB was successfully simulated by VIC-CAS model, which coupled with glacier modules with VIC-3L model, and the simulated single glacier area changes were compared with that observed from multiple temporal remote sensing data. The simulation suggested that it well represent the glacier meltwater and glacier change. Thus, the simulated snow coverage and snow depth by VIC-CAS model were evaluated by remote sensing products in URSRB. Based on 4 383 daily MODIS surface reflectance datasets and the Long-term Snow Depth Dataset of China from 2002 to 2013, in which the snow depth algorithm developed by Che and Dai of Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, the daily snow cover and snow depth simulated by VIC-CAS model were comprehensively compared in URSRB. All the forcing data, parameters, and the calibration and validation processes of VIC-CAS model in URSRB are the same with previous published literature. The comparison of stats in monthly average snow coverage and spatial pattern of annual average snow coverage in 182 sub-basins suggested that VIC-CAS model can better simulate the snow cover a in more snow year such as 2008, which has the higher relative coefficient (r) by 0.67 and lower root mean square error (RMSE) by 0.12. The simulation accuracy of VIC-CAS model is relatively lower in normal year such as 2004 with r by 0.37 and RMSE by 0.13 and less snow year such as 2013, which has r by 0.52 and RMSE by 0.09. The spatial distribution of annual average simulated snow cover has the similar patterns with that from remote sensing data, especially in more snow year, although the simulated snow coverage is less than observed by remote sensing data. The best simulation of snow coverage is located at 4 000~5 000 m a.s.l. altitude zone, which has high r by 0.44, 0.66, 0.60, and RMSE by 0.15, 0.12, 0.11 in normal year, more snow year, and less snow year, respectively, while that is the worst at 2 000~3 000 m a.s.l. altitude zone, which has r by -0.1. The consistency of snow depth between simulated and observed is high in 2008, while is low in other years. The simulated annual average snow depth is less than observed by remote sensing data. The spatial distribution of simulated snow depth suggested that it has more relationship with altitude than snow depth product, due to the later has coarse resolution. These results indicate that VIC-CAS model has a lower accuracy in the low altitude area or with thin snow, which probably comes from the related algorithm and parameterizations in snow redistribution and wind-blown snow process, which need to further enhance the observation and simulation in the future. This study provides some clues for further improving the simulation ability of hydrological model in alpine cryospheric basins.

Keywords: VIC-CAS model ; snow coverage ; snow depth ; accuracy assessment ; mountainous upper reach of the Shule River basin

PDF (4397KB) 元数据 多维度评价 相关文章 导出 EndNote| Ris| Bibtex  收藏本文

本文引用格式

郭佳锴, 李哲, 李飞, 张世强. 基于多源遥感数据的疏勒河上游山区流域VIC-CAS模型积雪模拟效果评估[J]. 冰川冻土, 2021, 43(2): 650-661 doi:10.7522/j.issn.1000-0240.2021.0049

GUO Jiakai, LI Zhe, LI Fei, ZHANG Shiqiang. Evaluation on snow coverage and snow depth simulated by VIC-CAS model based on multi-source remote sensing data in mountainous upper reach of the Shule River basin[J]. Journal of Glaciology and Geocryology, 2021, 43(2): 650-661 doi:10.7522/j.issn.1000-0240.2021.0049

0 引言

在全球水循环中,积雪的积累和消融过程调节着水的年内再分配,是干旱半干旱地区春季最重要的淡水资源1。积雪作为冰冻圈的重要组成部分,是气候系统中最敏感的变化因子之一2。积雪是全球气候变化过程中的重要因素,积雪变化对地表反照率、全球和地区能量平衡有重要影响3-4。同时,积雪分布和深度是天气数值预报模式的重要参数5

积雪覆盖和雪深可以由地面观测、遥感反演和模型模拟获得4-5。地面观测雪深资料是目前研究积雪长期变化特征最广泛、最可靠的资料,具有时间尺度长,连续性好的优点,但也存在着空间分辨率低,空间分布不均匀,观测投入较大的不足6。近年来,遥感积雪数据的不断发展,为寒区水文模型提供了丰富的数据源,改善了以往地面观测资料分辨率低,成本高的缺点7。目前,常见的积雪覆盖遥感产品有Landsat和SPOT8,AVHRR9,MODIS积雪产品10及SMMR、SSM/I11-12和AMSR-E13-14等微波积雪产品。MODIS数据因其较高的时空分辨率和光谱分辨率在积雪研究与动态变化监测领域中广泛应用15-16。张颖等17研究发现MODIS逐日积雪覆盖率产品在青藏高原地区精度高于MODIS标准积雪产品。在积雪深度监测方面,被动微波遥感数据被认为是监测大尺度积雪深度时空变化的有效手段。现有雪深产品在青藏高原地区表现出较大的不确定性18。Dai等19基于青藏高原的台站雪深观测数据和近年来对地面雪深的大规模调查资料,对利用SSM I/S和AMSR-E亮温数据反演雪深的算法进行了校准,进而生产了中国雪深数据集,这是目前青藏高原地区精度最高的雪深数据产品。

积雪的积累和消融过程是寒区水文模型的重要组成部分。寒区水文模型模拟积雪一般可分为度日因子和能量平衡两种方法20。其中,度日因子法是基于冰雪消融和气温之间的关系计算21,目前SRM22和SWAT23融雪模块均采用过这种方法。能量平衡模型是模拟积雪-大气,积雪-土壤及积雪内的物质和能量平衡21,常用的模型有ISNOBAL24、UEB25等,一些分布式水文模型也出现了融雪模块,如VIC26、DHSVM27等。近年来,已经出现了多个利用卫星遥感获得的积雪分布来驱动水文模型提高精度的成功案例728-30,表明利用卫星遥感积雪数据有助于改善以往水文模型仅仅利用出山口径流资料进行率定和检验的不确定性。如赵军等28在疏勒河流域上游将SRM融雪径流模型与MODIS积雪产品结合,取得了较好的模拟效果。Andreadis等29利用EnKF方法将遥感积雪面积数据和雪水当量数据同化到VIC模型中,对中低海拔,融雪期和浅层积雪的积雪变量模拟精度取得显著提升。Che等30利用EnKF方法将被动微波遥感雪深数据同化到陆面模型中,提高了积累期的雪深估算精度。VIC-CAS31分布式水文模型中很好地考虑了单条冰川的融水过程和冰川变化,其模拟的疏勒河流域的冰川变化与遥感监测的单条冰川变化32具有很好的可比性,其径流模拟也取得了很好的效果33,为评估模型中对积雪的积累和消融过程的合理性提供了很好的基础。

因此,本研究选取了在降水等观测资料较为丰富的疏勒河上游山区流域,基于中分辨率成像光谱仪(Moderate Resolution Imaging Spectroradiometer,简称MODIS)逐日积雪覆盖资料和中国长序列遥感反演雪深资料,选取对冰川融水模拟较好的VIC-CAS模型模拟的积雪覆盖和雪深进行评估,并对比模型在不同高程带的模拟效果差异,为进一步改进模型中的积雪过程提供支撑。

1 研究区与数据

1.1 研究区概况

疏勒河是河西走廊三大内陆河之一,发源于祁连山腹地的疏勒南山北坡,其出山口昌马水文站以上为疏勒河上游(96.6°~99.0° E,38.2°~40.0° N)34-35图1)。疏勒河上游地区地形主要由疏勒南山、托勒南山和疏勒河谷地组成,山区地势高峻、地形陡峭,谷地地形相对低缓。疏勒河上游面积约1.14×104 km2,属于高原大陆性气候,研究区海拔介于2 100~5 750 m之间,平均海拔为3 900 m36-37,其中2 000~3 000 m高程带所占区域占整个研究区面积的7%,主要是河流出山口的河谷区;3 000~4 000 m高程带所占区域占整个研究区面积的45%,主要为疏勒河上游谷地中段,河谷两侧的山前缘区;4 000~5 000 m高程带所占区域占整个研究区面积的47%;5 000~6 000 m高程带所占区域占整个研究区面积的2%37。研究区分布大量冻土,积雪和冰川38,多年平均气温约-4 ℃,多年平均降雨量约378.4 mm,降水主要集中在5—9月,降水量约占全年总降雨量的90%以上,冬季和春季以降雪为主39

图1

图1   疏勒河上游山区流域气象站、雨量计和高程带分布及VIC-CAS模型中的182个子流域单元

Fig.1   Spatial distribution of meteorological stations and rain gauges, elevation zones, and 182 sub-basins of VIC-CAS in mountainous upper reach of the Shule River basin


1.2 数据及预处理

1.2.1 MODIS积雪覆盖数据

本研究中的积雪覆盖观测数据利用美国国家雪冰中心(National Snow and Ice Data Center,NSIDC)网站下载的TERRA MODIS MOD09GA地表反射率产品计算获得。MOD09GA的时间分辨率为日,空间分辨率为500 m。本研究收集了2002—2013年全年共计4 383天的MOD09产品1~7通道的数字影像。在选择不同版本的MODIS积雪面积产品时,考虑到在青藏高原区域MOD09GA数据经过处理后的积雪覆盖与同时相Landsat TM/ETM+影像获取的积雪覆盖的相关系数r达到0.85,高于MOD10A1的0.74,最终选择MOD09GA作为检验数据17

对MOD09GA影像的预处理包括对卫星影像进行坐标变换。将正弦曲线投影转换为地理坐标,椭球体选为WGS84。选用双线性法重采样,将图像文件转换为GeoTIFF格式,并利用研究区边界进行剪裁。根据雪的反射率特点,利用NDSI算法对MODIS数据对每个像元的积雪覆盖度进行计算40,其公式为

NDSI=Band4-Band6Band4+Band6

式中:Band4、Band6分别为MODIS波段4和波段6的反射率。

根据高扬等41在青藏高原地区不同土地覆盖类型NDSI阈值积雪判别的研究结果,选取0.33作为最佳阈值,并根据波段2和波段4排除水体和暗物质干扰。本研究利用MOD10A1产品中的云覆盖数据确定含云像元,在进行对比时对含云像元不作对比。不含云的各像元的积雪覆盖度,采用Salomoson等42提出的NDSI与真实亚像元积雪覆盖度之间的经验公式获得。

FSC=0.06+1.21×NDSI

式中:FSC为每个像元的积雪覆盖度;NDSI为每个像元的NDSI值。对逐日数据进行计算,最终得到2002—2013年疏勒河上游山区流域MODIS各子流域的日积雪覆盖度。

1.2.2 雪深数据

本研究中2002—2013年疏勒河上游山区流域的雪深数据从国家冰川冻土沙漠科学数据中心网站(www.crensed.ac.cn)下载的中国雪深长时间序列数据集(the Long-term Snow Depth Dataset of China)中提取。该数据集中2002—2007年的雪深数据用AMSR-E反演获得,2008—2016年的雪深采用SSMI/S传感器的亮度温度反演,其时间分辨率为日,空间分辨率均为0.25°。用于反演该雪深数据集的原始数据来自美国国家雪冰数据中心(NSIDC)处理的AMSR-E(2002—2007年)和SSMI/S(2008—2019年)逐日被动微波亮温数据,其首先通过对不同传感器的亮温进行交叉定标提高亮温在时间上的一致性,然后利用在针对中国地区修正的Chang算法进行雪深反演11

本研究将下载的ASCII文件用Python语言批量转成栅格,利用研究区边界剪裁得到2002—2013年疏勒河上游山区流域逐月和逐年平均的积雪深度栅格文件。根据遥感观测雪深分别将最接近平均值的2004年作为平雪年的代表,雪深最大的2008年作为多雪年代表,雪深最小的2013年作为少雪年代表。

1.2.3 VIC-CAS模型模拟的积雪覆盖度和雪深

VIC-CAS模型是在VIC-3L基础上改进的分布式水文模型43,其计算每个网格单元的能量和水分平衡,并考虑到积雪、融雪和土壤冻融过程26。VIC模型可以由日或日内尺度气象数据驱动,包括降水量、最高和最低气温以及风速数据,模型通过内置的插值程序生成各格网的气象驱动数据,模型参数包括土壤和植被参数,以及数字高程模型等44。VIC-CAS模型增加了冰川模块,将单条冰川作为子流域,进而将单条冰川细分为100 m间距高程带,利用度日因子逐高程带计算冰川消融量和物质平衡,并考虑了冰川面积的长期变化45。VIC-CAS通过分别采用单层和双层模型计算冠层和地面积雪过程和融雪过程,其中在双层模型中考虑了积雪的积累和升华过程,并考虑了不同高程带积雪积累和消融的空间异质性46。VIC-CAS积雪算法中的雪深与雪龄有关,考虑了积雪的密实化过程以及新雪的影响47。VIC-CAS模型假定子流域内的积雪均匀分布,通过阈值分段考虑了积雪覆盖面积与雪深之间的关系,从而考虑了薄雪覆盖率较低的特征29

Zhang等33利用VIC-CAS模型模拟了疏勒河上游山区流域的冰川径流和河川径流,与遥感获得的不同时期的单条冰川变化32和出山口观测径流对比表明,VIC-CAS取得了很好的模拟效果。本研究采用Zhang等33在疏勒河上游山区率定和验证的VIC-CAS模型参数,与遥感获取的积雪覆盖度和积雪深度对比,分析了VIC-CAS模型在疏勒河上游山区流域模拟的积雪覆盖度和积雪深度模拟效果。模型中的年降水梯度为根据1 139~4 156 m的降水观测计算所得的14.654 mm·(100m)-1,模型的关键参数、率定过程、冰川模拟结果和河流径流结果详见文献[33]。本研究中直接选取了2002—2013年疏勒河上游山区流域182个子流域的积雪覆盖度和雪深日模拟结果,其中,每个子流域的高程通过子流域中各高程带所占面积比率与高程的乘积累加得到。随后用属性表的子流域编号作为关键字,对VIC-CAS模型结果数据文件中的积雪覆盖度和积雪深度两个变量的数据与疏勒河上游山区流域子流域矢量边界进行连接,从而将MODIS积雪覆盖数据和中国雪深长时间序列数据与子流域编号链接统计。

考虑到遥感反演的雪深具有较大的不确定性,本研究中对模型的评价重点放在积雪覆盖度指标,对雪深指标主要对比其一致性,不作为评价重点。

1.3 统计分析
1.3.1 相关分析

采用相关系数(r)评价VIC-CAS模型模拟与观测的积雪覆盖度和积雪深度值之间的相关性。

r=i=1n(Oi-O¯)×(Mi-M¯)i=1n(Oi-O¯)2×i=1n(Mi-M¯)2

式中:Oi 为VIC-CAS模型模拟的积雪覆盖度和积雪深度;O¯为VIC-CAS模型模拟的日平均积雪覆盖度和日积雪深度的平均值;Mi 为MODIS观测的积雪覆盖度和中国雪深长时间序列数据集的积雪深度;M¯为MODIS影像积雪覆盖度和中国雪深长时间序列数据集积雪深度的平均值;n为总月数。

根据文献[48],将r划分为6个区间:r≤0为负相关,0<r≤0.2为不相关或极弱相关,0.2<r≤0.4为弱相关,0.4<r≤0.6为中相关,0.6<r≤0.8为强相关,0.8<r≤1为极强相关。

1.3.2 均方根误差

采用均方根误差(RMSE)分析方法评价VIC-CAS模型模拟与观测的积雪覆盖度和积雪深度之间的离散程度。

RMSE=i=1n(Mi-Oi)2n

式中:Mi为MODIS影像积雪覆盖度和中国雪深长时间序列数据集的积雪深度;Oi为VIC-CAS模型模拟的积雪覆盖度和积雪深度的平均值;n为总月数。

1.4 评估流程

疏勒河上游山区VIC-CAS模拟和遥感观测的积雪覆盖和雪深评估流程如图2所示。数据预处理包括对VIC-CAS模型输出文件分别计算各子流域的月平均和年平均积雪覆盖度和积雪深度,对中国雪深长时间序列数据集进行栅格转换,提取研究区月平均和年平均积雪深度。对MOD09GA数据的处理过程包括NDSI计算,NDSI与积雪覆盖度之间的经验线性回归方程计算,重采样,提取研究区范围。数据分析包含利用ArcPy语言对处理后的数据使用以表格显示分区统计工具统计182个子流域中积雪覆盖度和积雪深度的平均值,并计算r和RMSE。

图2

图2   疏勒河上游山区VIC-CAS模型模拟和遥感观测的积雪覆盖度和雪深评估流程

Fig.2   Flow chart of the evaluation on snow coverage and snow depth simulated by VIC-CAS model and observed by remote sensing in mountainous upper reach of the Shule River basin


2 结果与分析

2.1 积雪覆盖度评估

2.1.1 平雪年、多雪年和少雪年积雪覆盖度

在疏勒河上游山区流域VIC-CAS模型模拟和MODIS观测的月平均积雪覆盖度在平雪年,多雪年和少雪年的对比如图3所示。r在多雪年最大,为0.67,属于强相关;少雪年为0.52,属于中相关;平雪年为0.37,属于弱相关。RMSE在三个年份相差不大,多雪年为0.12,少雪年为0.09,平雪年为0.13。总体来看,VIC-CAS模型在多雪年的模拟效果最好,少雪年和平雪年精度相对较低,且存在大量模型模拟为0但MODIS积雪观测有值的点,表明薄雪的模拟结果可能较差。

图3

图3   VIC-CAS模型模拟和MODIS观测的月平均积雪覆盖度在疏勒河上游山区流域不同降雪年份的对比

Fig.3   Comparisons in monthly averaged snow coverage between simulated by VIC-CAS model and observed by MODIS in mountainous upper reach of the Shule River basin in normal snow year (a), more snow year (b) and less snow year (c)


疏勒河上游山区流域模拟和观测的年平均积雪覆盖度在平雪年,多雪年和少雪年的空间分布如图4所示。从空间分布看,VIC-CAS模型模拟的积雪覆盖度偏小,多雪年模拟数据与MODIS产品表现出最为相似的空间分布特征,即研究区中部高山区积雪覆盖度高,西北部河谷区积雪覆盖度低,表明其对积雪的空间分布模拟较好,少雪年和平雪年积雪空间分布模拟的特征相似性较差。

图4

图4   VIC-CAS模型模拟和MODIS观测的年平均积雪覆盖度在不同降雪年份的空间分布

Fig.4   Spatial distribution of yearly averaged snow coverage between simulated by VIC-CAS model and observed by MODIS in different years (a, b, c are simulated by VIC-CAS model; d, e, f are observed by MODIS)


2.1.2 不同高程带积雪覆盖度

VIC-CAS模型模拟和MODIS观测的月平均积雪覆盖度在不同高程带2 000~3 000 m,3 000~4 000 m,4 000~5 000 m的对比如图5所示。VIC-CAS模型在海拔4 000~5 000 m的模拟效果最好,其在平雪年,多雪年和少雪年中r分别为0.41,0.66和0.60,RMSE分别为0.15,0.12,0.11,模拟精度较高。海拔2 000~3 000 m模拟效果不佳,在平雪年出现了负相关-0.10。总的来看,VIC-CAS模型在占比最高的4 000~5 000 m高程带模拟精度最高,3 000~4 000 m其次,在较低海拔的2 000~3 000 m积雪覆盖度模拟精度较低。

图5

图5   VIC-CAS模型模拟和MODIS观测的月平均积雪覆盖度在疏勒河上游山区流域不同高程带的对比

Fig.5   Comparisons in monthly averaged snow coverage between simulated by VIC-CAS model and observed by MODIS at different elevation zones in mountainous upper reach of the Shule River basin


2.2 雪深对比
2.2.1 平雪年、多雪年和少雪年雪深

VIC-CAS模型模拟和中国雪深长时间序列数据集观测的月平均积雪深度在平雪年,多雪年和少雪年的对比如图6所示。VIC-CAS模型在多雪年的r为0.44,属于中相关;在平雪年和少雪年相关程度较低,其中平雪年出现负相关-0.22。总体来看,VIC-CAS模型在不同降雪年份雪深一致性较低。平雪年和少雪年存在大量模拟积雪深度为0但积雪产品中深度不为0的情况,说明薄雪仍然是其主要差异所在。

图6

图6   VIC-CAS模型模拟和中国雪深长时间序列数据集观测的月平均积雪深度在疏勒河上游山区流域不同降雪年份的对比

Fig.6   Comparisons in monthly averaged snow depth between simulated by VIC-CAS model and observed by the Long-term Snow Depth Dataset of China in mountainous upper reach of the Shule River basin in normal snow year (a), more snow year (b) and less snow year (c)


模拟和观测的年平均积雪深度在平雪年,多雪年和少雪年的空间分布如图7所示。可以看出VIC-CAS模型模拟的雪深总体偏小,但很好地反映了雪深随高程变化的特点;观测的雪深数据与高程带的关系刻画不明显,主要是较粗的空间分辨率所致。

图7

图7   VIC-CAS模型模拟和中国雪深长时间序列数据集观测的年平均积雪深度在不同降雪年份的空间分布

Fig.7   Spatial distribution of yearly averaged snow depth between simulated by VIC-CAS model and observed by the Long-term Snow Depth Dataset of China in different years (a, b, c are simulated by VIC-CAS model; d, e, f are observed by the Long-term Snow Depth Dataset of China)


2.2.2 不同高程带雪深

VIC-CAS模型模拟和中国雪深长时间序列数据集观测的月平均积雪深度在2 000~3 000 m、3 000~4 000 m、4 000~5 000 m高程带的对比如图8所示。可以看出,VIC-CAS模拟值与观测的积雪深度的r与高程带无关,其均在多雪年呈现中相关,少雪年无相关性,平雪年呈现负相关的特点,其可能仍与观测的积雪深度分辨率较粗有关。

图8

图8   疏勒河上游山区流域VIC-CAS模型模拟和中国雪深长时间序列数据集观测的月平均积雪深度在不同高程带下的对比

Fig.8   Comparisons in monthly averaged snow depth between simulated by VIC-CAS and observed by the Long-term Snow Depth Dataset of China at different elevation zones in mountainous upper reach of the Shule River basin


3 讨论

传统积雪产品大多采用地面气象站观测值空间插值和人工观测获得,如Bulygina等49利用856个观测站的积雪深度数据研究俄罗斯的积雪覆盖和雪深的年际变化。这种方法需要耗费大量人力物力,且难以获得连续的,大面积的积雪信息50。在疏勒河上游山区流域不存在有关积雪的气象台站数据,无法提供有效的地面实测资料进行交叉验证的数据支持。利用遥感获取积雪覆盖和雪深能够有效地获取积雪的空间分布1。因此,本研究证明了采用遥感反演的积雪对整个研究区的积雪模拟效果进行评估是有效可行的。

从不同积雪覆盖的遥感产品的精度看,张颖等17研究表明,MOD10A1产品对于积雪破碎区的信息提取较差,山体阴影的漏分现象更为严重,较小区域的积雪识别误差大,精度低。王雪璐等51研究表明,MOD10A1在青海省的积雪分类精度明显低于在其他研究区的精度验证结果,经过NDSI阈值调整后的MOD09GA产品的精度高于同时期MOD10A1产品。本研究中采用了这一算法,因此积雪覆盖产品应具有较高的精度。

目前来看,雪深数据产品空间分辨率仍较粗,难以反映高程分布的影响。Dai等18在2016年,基于SSMI/S(F17)对所有传感器的亮温数据进行校正,最终得到长期的中国积雪深度数据,并在2019年52开发了一种适用于被动微波的积雪识别方法,提升了产品的精度。尽管本研究中所采用的积雪产品是目前在中国精度最高的雪深数据产品,但依旧存在空间分辨率低和不确定性较大的问题,所以本研究中的评估以积雪覆盖结果为主,雪深结果重在对比两者之间的一致性。

需要指出的是,遥感反演的积雪覆盖和雪深也具有较大的不确定性。MODIS积雪覆盖数据的不确定性主要在于MOD09GA产品进行归一化处理的时候受到云,水体等影响以及处理过程中NDSI与积雪覆盖度的线性回归算法的误差。Salomonson等42建立了NDSI与真实亚像元积雪覆盖度之间的关系,该回归方程是在西伯利亚地区积雪观测的基础上建立的,对其他地区的适用性可能存疑。本研究中存在大量观测值为0的数据,可能是由于线性回归算法得到的积雪覆盖度剔除了小于0.1的数据产生的误差42。遥感雪深反演数据的不确定性主要在于产生中国地区雪深长时间序列数据集的被动微波卫星遥感资料空间分辨率很低,反演过程时积雪的空间异质性问题难以解决。

从空间分布对比看,VIC-CAS模型模拟的积雪覆盖度和雪深偏小,自2009年起在研究区内开始布设较多的降水观测设备,获取了较为准确的降水梯度,模型的冰川融水和径流模拟效果均较好33,为对比积雪的相关结果提供了很好的基础。本研究发现VIC-CAS模型和遥感反演的薄雪覆盖度的差异较大,在不同高程带对积雪的模拟精度差别不大,但在2 000~3 000 m模拟精度较低,说明在低海拔区存在较大不确定性,这可能与模型中对积雪再分布和风吹雪的算法和参数化方案有关,特别是在薄雪情况下,其可能存在较大的不确定性。因此,需要在流域尺度上进一步加强对风吹雪分布、过程和参数化方案的观测和模拟优化。

4 结论

本文基于2002—2013年疏勒河上游山区流域MODIS积雪覆盖度产品和中国雪深长时间序列数据集,对VIC-CAS模型模拟的积雪覆盖度和雪深进行了评估,主要结论如下:

(1) 总体来看,VIC-CAS模型对薄雪和低海拔区的模拟效果相对较差,这可能与模型中地形对降雪的影响,以及风吹雪造成的积雪再分布等过程可能考虑不够有关,需进一步加强对降雪和风吹雪过程的观测和模拟。

(2) 从不同降雪年份看,积雪覆盖度和雪深都表现为多雪年模拟效果较好,多雪年积雪覆盖度r为0.67,RMSE为0.12;少雪年和平雪年模拟效果都较差,说明VIC-CAS对薄雪模拟存在较大的不确定性。

(3) 从不同海拔看,VIC-CAS模拟的积雪覆盖度在4 000~5 000 m处精度最高,r平均为0.56,RMSE平均为0.13,在2 000~3 000 m处精度最低,说明VIC-CAS模拟的积雪在低海拔区不确定性较大。

参考文献

Che TaoHao XiaohuaDai Liyunet al.

Snow cover variation and its impacts over the Qinghai-Tibet Plateau

[J]. Bulletin of Chinese Academy of Sciences, 20193411): 1247-1253.

[本文引用: 2]

车涛郝晓华戴礼云.

青藏高原积雪变化及其影响

[J]. 中国科学院院刊, 20193411): 1247-1253.

[本文引用: 2]

Wang YetangHe YongHou Shugui.

Analysis of the temporal and spatial variations of snow cover over the Tibetan Plateau based on MODIS

[J]. Journal of Glaciology and Geocryology, 2007296): 855-861.

[本文引用: 1]

王叶堂何勇侯书贵.

2000-2005年青藏高原积雪时空变化分析

[J]. 冰川冻土, 2007296): 855-861.

[本文引用: 1]

Gao YangXie HongjieYao Tandonget al.

Integrated assessment on multi-temporal and multi-sensor combinations for reducing cloud obscuration of MODIS snow cover products of the Pacific Northwest USA

[J]. Remote Sensing of Environment, 20101148): 1662-1675.

[本文引用: 1]

Zhou WeiLiu LilongHuang Liangkeet al.

A new GPS SNR-based combination approach for land surface snow depth monitoring

[J/OL]. Scientific Reports, 201991) [2021-04-07]. .

[本文引用: 2]

Zhang Haijun.

Study on spatiotemporal variations of snow from 2000 to 2009 in Northeast China

[D]. ChangchunJilin University2010.

[本文引用: 2]

张海军.

2000-2009年东北地区积雪时空变化研究

[D]. 长春吉林大学2010.

[本文引用: 2]

Li DongliangWang Chunxue.

Research progress of snow cover and its influence on China climate

[J]. Transactions of Atmospheric Sciences, 2011345): 627-636.

[本文引用: 1]

李栋梁王春学.

积雪分布及其对中国气候影响的研究进展

[J]. 大气科学学报, 2011345): 627-636.

[本文引用: 1]

Yan YunaChe TaoLi Hongyiet al.

Using snow remote sensing data to improve the simulation accuracy of spring snowmelt runoff: take Babao River basin as an example

[J]. Journal of Glaciology and Geocryology, 2016381): 211-221.

[本文引用: 2]

闫玉娜车涛李弘毅.

使用积雪遥感面积数据改善山区春季融雪径流模拟精度

[J]. 冰川冻土, 2016381): 211-221.

[本文引用: 2]

Dankers Rde Jong S M.

Monitoring snow-cover dynamics in Northern Fennoscandia with SPOT VEGETATION images

[J]. International Journal of Remote Sensing, 20042515): 2933-2949.

[本文引用: 1]

Wang SiyuanYin HangYang Qichunet al.

Spatiotemporal patterns of snow cover retrieved from NOAA-AVHRR LTDR: a case study in the Tibetan Plateau, China

[J]. International Journal of Digital Earth, 2016105): 504-521.

[本文引用: 1]

Hall D KRiggs G ASalomonson V Vet al.

MODIS snow-cover products

[J]. Remote Sensing of Environment, 200283181-194.

[本文引用: 1]

Che TaoLi XinGao Feng.

Estimation of snow water equivalent in the Tibetan Plateau using passive microwave remote sensing data (SSM/I)

[J]. Journal of Glaciology and Geocryology, 2004263): 363-368.

[本文引用: 2]

车涛李新高峰.

青藏高原积雪深度和雪水当量的被动微波遥感反演

[J]. 冰川冻土, 2004263): 363-368.

[本文引用: 2]

Che TaoLi Xin.

Spatial distribution and temporal variation of snow water resources in China during

1993-2002[J]. Journal of Glaciology and Geocryology, 2005271): 64-67.

[本文引用: 1]

车涛李新.

1993-2002年中国积雪水资源时空分布与变化特征

[J]. 冰川冻土, 2005271): 64-67.

[本文引用: 1]

Yu HuiFeng QishengZhang Xuetonget al.

An approach for monitoring snow depth based on AMSR-E data in the pastoral area of northern Xinjiang

[J]. Acta Prataculturae Sinica, 2009184): 210-216.

[本文引用: 1]

于慧冯琦胜张学通.

基于AMSR-E信息的北疆牧区雪深遥感监测模型方法初探

[J]. 草业学报, 2009184): 210-216.

[本文引用: 1]

Feng QishengZhang XuetongLiang Tiangang.

Dynamic monitoring of snow cover based on MOD

10A1 and AMSR-E in the north of Xinjiang[J]. Acta Prataculturae Sinica, 2009181): 125-133.

[本文引用: 1]

冯琦胜张学通梁天刚.

基于MOD10A1和AMSR-E的北疆牧区积雪动态监测研究

[J]. 草业学报, 2009181): 125-133.

[本文引用: 1]

Yan Hao.

A comparison of MODIS and passive microwave snow mapping

[J]. Journal of Glaciology and Geocryology, 2005274): 515-519.

[本文引用: 1]

延昊.

利用MODIS和AMSR-E进行积雪制图的比较分析

[J]. 冰川冻土, 2005274): 515-519.

[本文引用: 1]

Hou HuishuYang Hongye.

A general introduction to MODIS snow products and its researching application

[J]. Remote Sensing Technology and Application, 2009242): 252-256.

[本文引用: 1]

侯慧姝杨宏业.

MODIS积雪产品及研究应用概述

[J]. 遥感技术与应用, 2009242): 252-256.

[本文引用: 1]

Zhang YingHuang XiaodongWang Weiet al.

Validation and algorithm redevelopment of MODIS daily fractional snow cover products

[J]. Arid Zone Research, 2013305): 808-841.

[本文引用: 3]

张颖黄晓东王玮.

MODIS逐日积雪覆盖率产品验证及算法重建

[J]. 干旱区研究, 2013305): 808-814.

[本文引用: 3]

Dai LiyunChe TaoDing Yongjianet al.

Evaluation of snow cover and snow depth on the Qinghai-Tibetan Plateau derived from passive microwave remote sensing

[J]. The Cryosphere, 2017114): 1933-1948.

[本文引用: 2]

Dai LiyunChe TaoDing Yongjian.

Inter-calibrating SMMR, SSM/I and SSMI/S data to improve the consistency of snow-depth products in China

[J]. Remote Sensing, 201576): 7212-7230.

[本文引用: 1]

Qiao PengQin YanLiu Zhihui.

A spatially distributed snowmelt model based on energy balance

[J]. Journal of China Hydrology, 2011313): 22-26.

[本文引用: 1]

乔鹏秦艳刘志辉.

基于能量平衡的分布式融雪径流模型

[J]. 水文, 2011313): 22-26.

[本文引用: 1]

Zhao QiudongLiu ZhihuiQin Rongmaoet al.

Research advance of snowmelt model

[J]. Xinjiang Agricultural Sciences, 2007446): 734-739.

[本文引用: 2]

赵求东刘志辉秦荣茂.

融雪模型研究进展

[J]. 新疆农业科学, 2007446): 734-739.

[本文引用: 2]

Liu JunfengYang JianpingChen Renshenget al.

The simulation of snowmelt runoff model in the Dongkemadi River basin, headwater of the Yangtze River

[J]. Acta Geographica Sinica, 20066111): 1149-1159.

[本文引用: 1]

刘俊峰杨建平陈仁升.

SRM融雪径流模型在长江源区冬克玛底河流域的应用

[J]. 地理学报, 20066111): 1149-1159.

[本文引用: 1]

Meng XianyongJi XiaonanLiu Zhihuiet al.

Research on improvement and application of snowmelt module in SWAT

[J]. Journal of Natural Resources, 2014293): 528-539.

[本文引用: 1]

孟现勇吉晓楠刘志辉.

SWAT模型融雪模块的改进与应用研究

[J]. 自然资源学报, 2014293): 528-539.

[本文引用: 1]

Andrew R HDanny MScott Het al.

Direct insertion of NASA airborne snow observatory-derived snow depth time series into the iSnobal energy balance snow model

[J]. Water Resources Research, 20185410): 8045-8063.

[本文引用: 1]

Tarboton D GLuce C H.

Utah energy balance snow accumulation and melt model (UEB)

[R]. Logan, Utah, USAWater Research Laboratory & USDA Forest Service Intermountain Research Station1996.

[本文引用: 1]

Ling XuLettenmaier D PWood E F.

Surface soil moisture parameterization of the VIC-2L model: elevation and modification

[J]. Global Plant Change, 1996131): 195-206.

[本文引用: 2]

Wigmosta M SVail L WLettenamaier D P.

A distributed hydrological-vegetation model for complex terrain

[J]. Water Resource Research, 1994306): 1665-1679.

[本文引用: 1]

Zhao JunHuang YongshengSong Geqinget al.

Application of snowmelt runoff model in upper stream of Shule River basin

[J]. Journal of Water Resources & Water Engineering, 2015261): 72-76.

[本文引用: 2]

赵军黄永生宋阁庆.

SRM融雪径流模型在疏勒河流域上游的应用

[J]. 水资源与水工程学报, 2015261): 72-76.

[本文引用: 2]

Andreadis K MLettenmaier D P.

Assimilating remotely sensed snow observations into a macroscale hydrology model

[J]. Advances in Water Resources, 2006296): 872-886.

[本文引用: 2]

Che TaoLi XinJin Ruiet al.

Assimilating passive microwave remote sensing data into a land surface model to improve the estimation of snow depth

[J]. Remote Sensing of Environment, 201414354-63.

[本文引用: 2]

Zhao QiudongYe BaishengDing Yongjianet al.

Coupling a glacier melt model to the Variable Infiltration Capacity (VIC) model for hydrological modeling in Northwestern China

[J]. Environmental Geology, 2013681): 87-101.

[本文引用: 1]

Zhang XiaowenLi HaojieZhang Zhihuaet al.

Recent glacier mass balance and area changes from DEMs and Landsat images in upper reach of Shule River basin, northeastern edge of Tibetan Plateau during 2000 to 2015

[J/OL]. Water, 2018106) [2021-04-07]. .

[本文引用: 2]

Zhang ZhihuaDeng ShifanZhao Qiudonget al.

Projected glacier meltwater and river run-off changes in the upper reach of the Shule River basin, northeastern edge of the Tibetan Plateau

[J]. Hydrological Processes, 2019337): 1059-1074.

[本文引用: 5]

Xu HaojieYang TaibaoZhang Xiaoxiao.

Climate change in upstream area of Shule River and its effects on runoff yield in last 50 years

[J]. Bulletin of Soil and Water Conservation, 2014344): 39-45.

[本文引用: 1]

徐浩杰杨太保张晓晓.

近50年来疏勒河上游气候变化及其对地表径流的影响

[J]. 水土保持通报, 2014344): 39-45.

[本文引用: 1]

Chang YapingZhong DanLi Haojieet al.

Land surface temperature retrieved from HJ-1B satellite data in the upper reaches of the Shule River

[J]. Journal of Glaciology and Geocryology, 2015374): 954-962.

[本文引用: 1]

苌亚平种丹李浩杰.

基于HJ-1B卫星数据的疏勒河上游流域地表温度反演

[J]. 冰川冻土, 2015374): 954-962.

[本文引用: 1]

Chen ShengyunLiu WenjieYe Baishenget al.

Species diversity of vegetation in relation to biomass and environmental factors in the upper area of the Shule River

[J]. Acta Prataculturae Sinica, 2011203): 70-83.

[本文引用: 1]

陈生云刘文杰叶柏生.

疏勒河上游地区植被物种多样性和生物量及其与环境因子的关系

[J]. 草业学报, 2011203): 70-83.

[本文引用: 1]

Sheng YuLi JingWu Jichunet al.

Distribution patterns of permafrost in the upper area of Shule River with the application of GIS technique

[J]. Journal of China University of Mining & Technology, 2010391): 32-39.

[本文引用: 2]

盛煜李静吴吉春.

基于GIS的疏勒河流域上游多年冻土分布特征

[J]. 中国矿业大学学报, 2010391): 32-39.

[本文引用: 2]

Xie XiaYang GuojingWang Zengruet al.

Landscape pattern change in mountainous areas along an altitude gradient in the upper reaches of Shule River

[J]. Chinese Journal of Ecology, 2010297): 1420-1426.

[本文引用: 1]

谢霞杨国靖王增如.

疏勒河上游山区不同海拔梯度的景观格局变化

[J]. 生态学杂志, 2010297): 1420-1426.

[本文引用: 1]

Wu HaoYe BaishengWu Jinkuiet al.

Observations and study on the CO2 flux in an alpine meadow ecosystem in the upper reaches of the Shule River basin

[J]. Acta Prataculturae Sinica, 2013224): 18-26.

[本文引用: 1]

吴灏叶柏生吴锦奎.

疏勒河上游高寒草甸生态系统CO2通量观测研究

[J]. 草业学报, 2013224): 18-26.

[本文引用: 1]

Salomonson V VAppel I.

Development of the Aqua MODIS NDSI fractional snow cover algorithm and validation results

[J]. IEEE Transactions on Geoscience & Remote Sensing, 2006447): 1747-1756.

[本文引用: 1]

Gao YangHao XiaohuaHe Dongcaiet al.

Snow cover mapping algorithm in the Tibetan Plateau based on NDSI threshold optimization of different land cover types

[J]. Journal of Glaciology and Geocryology, 2019415): 1162-1172.

[本文引用: 1]

高扬郝晓华和栋材.

基于不同土地覆盖类型NDSI阈值优化下的青藏高原积雪判别

[J]. 冰川冻土, 2019415): 1162-1172.

[本文引用: 1]

Salomonson V VAppel I.

Estimating fractional snow cover from MODIS using the normalized difference snow index

[J]. Remote Sensing of Environment, 2003893): 351-360.

[本文引用: 3]

Liang XuLettenmaier D PWood E Fet al.

A simple hydrologically based model of land surface water and energy fluxes for general circulation models

[J]. Journal of Geophysical Research: Atmospheres, 199499D7): 14415-14428.

[本文引用: 1]

Lohmann DRaschke ENijssen Bet al.

Regional scale hydrology: I. Formulation of the VIC-2L model coupled to a routing model

[J]. Hydrological Sciences Journal, 1998431): 131-141.

[本文引用: 1]

Zhao QiudongDing YongjianWang Jianet al.

Projecting climate change impacts on hydrological processes on the Tibetan Plateau with model calibration against the glacier inventory data and observed streamflow

[J]. Journal of Hydrology, 201957360-81.

[本文引用: 1]

Cherkauer K ALettenmaier D P.

Simulation of spatial variability in snow and frozen soil

[J/OL]. Journal of Geophysical Research: Atmospheres, 2003108(D22) [2021-04-07]. .

[本文引用: 1]

Cherkauer K ALettenmaier D P.

Hydrologic effects of frozen soils in the upper Mississippi River basin

[J]. Journal of Geophysical Research: Atmospheres, 1999104D16): 19599-19610.

[本文引用: 1]

Xia ZhiSong YuexinMa Jinet al.

Research on the Pearson correlation coefficient evaluation method of analog signal in the process of unit peak load regulation

[C]// 13th IEEE International Conference on Electronic Measurement & Instruments (ICEMI). New YorkIEEE2017522-527.

[本文引用: 1]

Bulygina O NRazuvaev V NKorshunova N N.

Changes in snow cover over Northern Eurasia in the last few decades

[J/OL]. Environmental Research Letters, 200944) [2021-04-07]. .

[本文引用: 1]

Chen XiuxueLi XiaofengWang Guangruiet al.

Based on snow cover survey data of accuracy verification and analysis of passive microwave snow cover remote sensing products in Northeast China

[J]. Remote Sensing Technology and Application, 2019346): 1181-1189.

[本文引用: 1]

陈秀雪李晓峰王广蕊.

基于积雪调查数据的东北地区被动微波积雪遥感产品精度验证与分析

[J]. 遥感技术与应用, 2019346): 1181-1189.

[本文引用: 1]

Wang XueluWang WeiFeng Qishenget al.

A snow cover mapping algorithm based on MODIS data in Qinghai Province

[J]. Acta Prataculturae Sinica, 2012214): 293-299.

[本文引用: 1]

王雪璐王玮冯琦胜.

基于MODIS数据的青海省积雪覆盖范围监测算法探索

[J]. 草业学报, 2012214): 293-299.

[本文引用: 1]

Dai LiyunChe TaoWang Jianet al.

Snow depth and snow water equivalent estimation from AMSR-E data based on a priori snow characteristics in Xinjiang, China

[J]. Remote Sensing of Environment, 20121271): 14-29.

[本文引用: 1]

/