考虑土壤水分影响的比辐射率方法在地表温度反演中的应用

doi:10.18402/resci.2017.08.15

摘要/Abstract

摘要：

地表比辐射率是确定地表长波能量平衡的一个关键参数,也是影响地表温度反演的主要因素,因此比辐射率的精确测定具有重要意义。地表比辐射率除了受地表覆盖类型的影响,与土壤水分含量也密切相关。本文针对MODIS通用分裂窗算法和Landsat TM/ETM+单窗算法,根据 Mira等建立的土壤含水量和土壤比辐射率的经验模型,利用SMEX04试验中Arizona研究区的遥感数据和地面观测数据,探讨考虑土壤水分影响后的比辐射率信息在地表温度反演中是否能够提高其反演精度。研究结果表明：利用考虑土壤水分影响后的比辐射率所反演的地表温度平均误差（ME）和均方根误差（RMSE）均低于比辐射率未考虑土壤水分影响反演的地表温度,其中通用分裂窗算法反演的地表温度ME降低了1.0~1.5K,RMSE降低了0.4~0.8K;单窗算法反演的地表温度ME降低了0.7K,RMSE降低了0.9K。因此,基于土壤比辐射率与土壤水分关系模型的比辐射率修正方法能够提高地表温度的反演精度,并且敏感性分析的结果表明目前土壤水分遥感数据0.04cm³/cm³的误差对本文使用的考虑土壤水分获取地表比辐射率进而反演地表温度的方法影响不明显。

关键词: MODIS通用分裂窗算法, Landsat TM/ETM+单窗算法, 比辐射率, SMEX04, 地表温度, 土壤水分

Abstract:

The emissivity of natural surfaces is a major parameter determining land surface temperature（LST）. In addition,the surface cover type influences emissivity and soil moisture is closely related to emissivity. Here,methods for obtaining the emissivity of bare soil using the MODIS generalized split-window algorithm and Landsat mono-window algorithm were improved. According to the empirical logarithm linear formula between soil moisture and soil emissivity,based on remote sensing data and ground observation data from the Soil Moisture Experiment 2004 （SMEX04）- Arizona study area,we discuss if the accuracy of land surface temperature retrieval can be improved when surface emissivity acquisition methods consider effects of soil moisture. We found that the accuracy of both improved algorithms considering soil moisture effects were better than algorithms not considering soil moisture effects. The mean error of LST retrieved by the improved MODIS generalized split-window algorithm reduced 1.0~1.5K,and the root mean square error reduced by 0.4~0.8K. Moreover,the mean error of LST retrieved by the improved Landsat mono-window algorithm reduced by 0.7K,and the root mean square error reduced by 0.9K. As a whole,the accuracy of land surface temperature retrieval can be improved when surface emissivity acquisition methods consider the effects of soil moisture,especially areas where vegetation coverage is less. Sensitivity analysis results show that the influence of remote sensing soil moisture data with a 0.04cm³/cm³ error on LST retrieval algorithms considering soil moisture effects is not obvious.

Key words: MODIS generalized split-window algorithm, Landsat TM/ETM+ mono-window algorithm, emissivity, SMEX04, Land surface temperature, soil moisture

董雪, 田静, 吴骅, 刘素华. 考虑土壤水分影响的比辐射率方法在地表温度反演中的应用[J]. 资源科学, 2017, 39(8): 1592-1604.

Xue DONG, Jing TIAN, Hua WU, Suhua LIU. Application of the emissivity method considering the effects of soil moisture for retrieving land surface temperature[J]. Resources Science, 2017, 39(8): 1592-1604.

图/表 16

表1

CE-312不同通道、不同质地的土壤比辐射率与土壤含水量关系模型各系数值"

	$ε i = a i + b i θ v + c i ln (θ v)$
CE-312通道	Channel 1（8~13.3μm）			Channel 2（11.5~12.4μm）			Channel 3（10.2~11.3μm）
对应传感器波段	TM/ETM+ 6			MODIS 32			MODIS 31
	a	b	c	a	b	c	a	b	c
壤土	0.963	-	0.009	0.965	0.040	0.004	0.958	0.052	0.004
砂壤	0.987	-	0.016	0.980	0.020	0.010	0.970	0.030	0.009
粉壤	0.968	-	0.012	1.010	-0.050	0.019	1.004	-0.04	0.018
粘壤	0.970	-	0.012	1.023	-0.068	0.025	0.996	-0.034	0.016

表1

图1

图2

图3

表2

图4

图5

表3

表4

表5

表6

表7

图6

表8

表9

图7

参考文献 31

[1]	江东,王乃斌,杨小唤,等. 地面温度的遥感反演:理论、推导及应用[J]. 甘肃科学学报,2001,13(4):36-40.
	[ Jiang D,Wang N B,Yang X H,et al. Retrieval of land surface temperature:Theory,deduction and application[J]. Journal of Gansu Sciences,2001,13(4):36-40.]
[2]	甘甫平,陈伟涛,张绪教,等. 热红外遥感反演陆地表面温度研究进展[J]. 国土资源遥感,2006,(1):6-11.
	[Gan F P,Chen W T,Zhang X J,et al. The progress in the study of thermal infrared remote sensing for retrieving land surface temperature[J]. Remote Sensing for Land & Resources,2006,(1):6-11.]
[3]	李琴,陈曦,包安明,等.干旱/半干旱区MODIS地表温度反演与验证研究[J]. 遥感技术与应用,2008,23(6):643-647.
	[Li Q,Chen X,Bao A M,et al. Validation of land surface temperatures retrieving in arid/semi-arid regions[J]. Remote Sensing Tech-nology and Application,2008,23(6):643-647.]
[4]	Snyder W C,Wan Z M,Zhang Y,et al. Classification-based emissivity for land surface temperature measurement from space[J]. International Journal of Remote Sensing,1998,19(14):2753-2774.
[5]	Li Z L,Petitcolin F,Zhang R H.A physically based algorithm for land surface emissivity retrieval from combined mid-infrared and thermal infrared data[J]. Science in China Series E:Technological Sciences,2000,43(1):23-33.
[6]	Peres L F,Dacamara C C.Land surface temperature and emis-sivity estimation based on the two-temperature method:Sen-sitivity analysis using simulated MSG/SEVIRI data[J]. Remote Sensing of Environment,2004,91(3-4):377-389.
[7]	Jiang G M,Li Z L,Nerry F.Land surface emissivity retrieval from combined mid-infrared and thermal infrared data of MSG-SEVIRI[J]. Remote Sensing of Environment,2006,105(4):326-340.
[8]	Peres L F,Dacamara C C,Trigo I F,et al. Synergistic use of the two-temperature and split-window methods for land-surface temperature retrieval[J]. International Journal of Remote Sensing,2010,31(16):4387-4409.
[9]	Wang N,Wu H,Nerry F,et al. Temperature and emissivity retrievals from hyperspectral thermal infrared data using linear spectral emissivity constraint[J]. IEEE Transactions on Geo-science and Remote Sensing,2011,49(4):1291-1303.
[10]	Van De Griend A A,Owe M. On the relationship between thermal emissivity and the normalized difference vegetation index for natural surfaces[J]. International Journal of Remote Sensing,1993,14(6):1119-1131.
[11]	Sobrino J A,Raissouni N,Li Z L.A comparative study of land surface emissivity retrieval from NOAA data[J]. Remote Sensing of Environment,2001,75(2):256-266.
[12]	田静,米素娟,何洪林,等. 土壤水分及粗糙度对比辐射率的影响[J]. 遥感学报,2016,20(4):561-569.
	[Tian J,Mi S J,He H L,et al. Effect of soil water content and soil roughness on the thermal infrared emissivity of bare soil[J]. Journal of Remote Sensing,2016,20(4):561-569. ]
[13]	肖青,柳钦火,李小文,等.热红外发射率光谱的野外测量方法与土壤热红外发射率特性研究[J]. 红外与毫米波学报,2003,22(5):373-378.
	[Xiao Q,Liu Q H,Li X W,et al. A field measurement method of spectral emissivity and research on the feature of soil thermal infrared emissivity[J]. Journal of Infrared and Millimeter Waves,2003,22(5):373-378.]
[14]	Xu L,Wang Q.Retrieval of soil water content in saline soils from emitted thermal infrared spectra using partial linear squares regression[J]. Remote Sensing,2015,7(11):14646-14662.
[15]	Mira M,Valor E,Caselles V,et al. Soil moisture effect on thermal infrared (8-13-μm)emissivity[J]. IEEE Transactions on Geoscience & Remote Sensing,2010,48(5):2251-2260.
[16]	Sanchez J M,French A N,Mira M,et al. Thermal infrared emissivity dependence on soil moisture in field conditions[J]. IEEE Transactions on Geoscience & Remote Sensing,2011,49(11):4652-4659.
[17]	Wang H S,Xiao Q,Li H,et al. Investigating the impact of soil moisture on thermal infrared emissivity using ASTER data[J]. Geoscience & Remote Sensing Letters IEEE,2015,12(2):294-298.
[18]	Qin Z H,Karnieli A,Berliner P.A mono-window algorithm for retrieving land surface temperature from Landsat TM data and its application to the Israel-Egypt border region[J]. International Journal of Remote Sensing,2001,22(18):3719-3746.
[19]	Wan Z M,Dozier J.A generalized split-window algorithm for retrieving land-surface temperature from space[J]. IEEE Trans-actions on Geoscience & Remote Sensing,1996,34(4):892-905.
[20]	覃志豪,李文娟,徐斌,等. 陆地卫星TM6波段范围内地表比辐射率的估计[J]. 国土资源遥感,2004,16(3):28-32.
	[Qin Z H,Li W J,Xu B,et al. The estimation of land surface emissivity for Landsat TM6[J]. Remote Sensing for Land Resources,2004,16(3):28-32.]
[21]	Sobrino J A,Jiménez-Muñoz J C,Paolini L. Land surface temperature retrieval from LANDSAT TM 5[J]. Remote Sensing of Environment,2004,90(4):434-440.
[22]	Mira M,Valor E,Boluda R,et al. Influence of soil water content on the thermal infrared emissivity of bare soils. Implication for land surface temperature determination[J]. Journal of Geo-physical Research Atmospheres,2007,112(F4):117-131.
[23]	Keefer T,Goodrich D,Moran S.SMEX04 Surface Temperature Data:Arizona[DB/OL]. Boulder,Colorado USA:NASA DAAC at the National Snow and Ice Data Center,2009.
[24]	Cosh M H,White W A.SMEX04 Soil Characteristics Data:Arizona[DB/OL]. Boulder,Colorado USA:NASA DAAC at the National Snow and Ice Data Center,2009.
[25]	Jackson T J,Bindlish R,Gasiewski A J,et al. SMEX04 Aircraft Polarimetric Scanning Radiometer(PSR)Data[DB/OL]. Boulder,Colorado USA:NASA DAAC at the National Snow and Ice Data Center,2009.
[26]	Hunt E R,Jackson T J,Yilmaz M T.SMEX04 Land Use Classi-fication Data[DB/OL]. Boulder,Colorado USA:NASA DAAC at the National Snow and Ice Data Center,2009.
[27]	Cosh M,Hunt E R,Jackson T J,et al. SMEX04 Landsat TM/ETM+ NDVI and NDWI[DB/OL]. Boulder,Colorado USA:NASA DAAC at the National Snow and Ice Data Center,2009.
[28]	Keefer T,Goodrich D,Moran S.SMEX04 Meteorological Network Data:Arizona[DB/OL]. Boulder,Colorado USA:NASA DAAC at the National Snow and Ice Data Center,2009.
[29]	Zeng J,Chen K S,Bi H,et al. A preliminary evaluation of the SMAP radiometer soil moisture product over United States and Europe using ground-based measurements[J]. IEEE Trans-actions on Geoscience & Remote Sensing,2016,54(8):4929-4940.
[30]	Sanchez N,Martinez-Fernandez J,Scaini A,et al. Validation of the SMOS L2 soil moisture data in the REMEDHUS network (Spain)[J]. IEEE Transactions on Geoscience & Remote Sensing,2012,50(5):1602-1611.
[31]	Jackson T J,Bindlish R,Cosh M H,et al. Validation of soil moisture and ocean salinity (SMOS)soil moisture over watershed networks in the U.S.[J]. IEEE Transactions on Geo-science & Remote Sensing,2012,50(5):1530-1543.

时间	ME			RMSE
时间	LST_GSW	LST_Weighed	LST_addSM_M	LST_GSW	LST_Weighed	LST_addSM_M
第221天	-1.81	-1.27	-0.42	2.25	1.77	1.52
第222天	-5.01	-4.50	-3.73	5.31	4.82	4.15
第223天	-0.70	-0.18	0.73	1.96	1.74	1.78
第226天	-0.71	-0.29	0.72	2.34	2.24	2.38
第238天	-4.33	-3.78	-2.62	5.24	4.77	3.91
第239天	-1.40	-0.84	0.35	2.56	2.28	2.12
Mean	-2.33	-1.81	-0.83	3.57	3.22	2.83

	地表类型	裸土	灌木地	草地	稀疏林地	常绿林地
地表比辐射率	Emi_GSW	0.971	0.974	0.975	0.977	0.979
	Emi_Weighed	0.966	0.971	0.982	0.982	0.982
	Emi_addSM_M	0.950	0.969	0.980	0.979	0.978
地表温度/K	LST_GSW	319.56	314.48	314.39	314.00	311.90
	LST_Weighed	319.88	314.68	314.08	313.70	311.76
	LST_addSM_M	320.67	314.73	314.16	313.82	311.80

	地表类型	裸土	灌木地	草地	稀疏林地	常绿林地
地表比辐射率	Emi_Qin	0.972	0.982	0.974	0.974	0.977
	Emi_addSM_L	0.931	0.969	0.960	0.940	0.957
地表温度/K	LST_Qin	305.71	306.01	304.90	303.22	301.63
	LST_addSM_L	307.97	306.67	305.70	305.04	302.69

	LST_obser	LST_Qin	LST_addSM_L
RG46	305.90	305.02	306.99
RG92	307.28	308.60	309.38
RG100	305.67	301.61	303.94
ME		-1.21	0.49
RMSE		2.52	1.69

输入参数	变化幅度	引起的MOIDS第31波段比辐射率平均误差	引起的MOIDS第32波段比辐射率平均误差	引起的地表温度平均误差 /K
θ_v（cm³/cm³）	+0.20	0.016	0.015	-0.94
	+0.16	0.014	0.013	-0.80
	+0.12	0.011	0.011	-0.64
	+0.08	0.008	0.008	-0.48
	+0.04	0.005	0.005	-0.28
	-0.04	-0.005	-0.005	0.27
	-0.08	-0.010	-0.010	0.52
	-0.12	-0.013	-0.013	0.58
	-0.16	-0.016	-0.017	0.75