An evaluation of a remote sensing method based on optimized triangular vegetation index （TVI） for aboveground shrub biomass estimation in shrub-encroached grassland

doi:10.11686/cyxb2022488

Abstract

Abstract:

Shrub-encroachment onto grassland is becoming an important ecological problem in grassland ecosystems， and accurate estimation of the shrub above-ground biomass （AGB） in shrub-encroached grassland vegetation plays a significant role in research into regional ecosystem carbon cycles. Due to the dual effects of soil background noise and differences in vegetation growth structure characteristics， the traditional vegetation indices are extremely unstable for model-building involving shrub-encroached grassland AGB estimation. To solve this problem， in this study we developed a novel way by optimizing the triangular vegetation index （TVI） using Sentinel-2 remote sensing data for shrub-encroached grassland AGB estimation. The results showed that： 1） In the area dominated by herbaceous vegetation， TVI calculated using a combination of green， red-edge and near-infrared （ $R 560$ ， $R 705$ and $R 865$ ） performed best with an R² of 0.684； in the area dominated by shrub vegetation， the TVI again performed best with R²= 0.368. 2） When analyzing the sensitivity of the 12 commonly used vegetation indexes to soil noise， the enhanced vegetation index （EVI） was the most sensitive to soil noise in the area dominated by herbaceous vegetation； in the area dominated by shrub vegetation， the modified soil adjusted vegetation index （MSAVI） was the most sensitive. 3） In the area dominated by herbaceous vegetation， the optimized vegetation index grassland triangular vegetation index （GTVI） performed better than TVI with the value of $R C V 2$ （coefficient of determination cross validation） increased by 0.153 and the value of $R M S E C V$ decreased by 12.222 g·m^-2； in the area dominated by shrub vegetation， GTVI performed better than TVI and the $R C V 2$ value increased 0.029， while the $R M S E C V$ （root mean square error cross validation） decreased 1.684 g·m^-2. 4） The estimation results acquired by GTVI showed the highest accuracy when compared with the results estimated by the commonly used 12 vegetation indices. The results of this study are expected to provide a scientific basis and reference AGB estimation in shrub-encroached grassland using vegetation indices extracted from remote sensing data.

Key words: shrub-encroached grassland, aboveground biomass of vegetation, vegetation index method, optimize triangular vegetation index

Zheng-yong XU, Bin SUN, Wang-fei ZHANG, Yi-fu LI, Zi-yu YAN, Wei YUE, Si-han TENG. An evaluation of a remote sensing method based on optimized triangular vegetation index （TVI） for aboveground shrub biomass estimation in shrub-encroached grassland[J]. Acta Prataculturae Sinica, 2023, 32(10): 1-14.

Figures/Tables 13

Fig.1 Study area location and sample distribution

Fig.2 Photos of field measurement

Fig.3 Schematic diagram of triangular vegetation index （TVI） construction

Table 1 Formula for calculating TVI

TVI组合Triangular vegetation index （TVI） combination

计算公式Formula

绿、红和红边组合Green， red and red-edge combination

$T V I = 0.5 × 105 × R 705 - R 560 - 145 × R 665 - R 560$

$T V I = 0.5 × 105 × R 740 - R 560 - 145 × R 665 - R 560$

$T V I = 0.5 × 105 × R 783 - R 560 - 223 × R 665 - R 560$

绿、红和近红组合Green， red and near-infrared combination

$T V I = 0.5 × 105 × R 842 - R 560 - 282 × R 665 - R 560$

$T V I = 0.5 × 105 × R 865 - R 560 - 305 × (R 665 - R 560)$

绿、红边和近红组合Green， red-edge and near-infrared combination

$T V I = 0.5 × 145 × R 842 - R 560 - 282 × R 705 - R 560$

$T V I = 0.5 × 145 × R 865 - R 560 - 305 × R 705 - R 560$

$T V I = 0.5 × 180 × R 842 - R 560 - 282 × R 740 - R 560$

$T V I = 0.5 × 180 × R 865 - R 560 - 305 × R 740 - R 560$

$T V I = 0.5 × 223 × R 842 - R 560 - 282 × R 783 - R 560$

$T V I = 0.5 × 223 × R 865 - R 560 - 305 × R 783 - R 560$

Table 1 Formula for calculating TVI

TVI组合Triangular vegetation index （TVI） combination

计算公式Formula

绿、红和红边组合Green， red and red-edge combination

$T V I = 0.5 × 105 × R 705 - R 560 - 145 × R 665 - R 560$

$T V I = 0.5 × 105 × R 740 - R 560 - 145 × R 665 - R 560$

$T V I = 0.5 × 105 × R 783 - R 560 - 223 × R 665 - R 560$

绿、红和近红组合Green， red and near-infrared combination

$T V I = 0.5 × 105 × R 842 - R 560 - 282 × R 665 - R 560$

$T V I = 0.5 × 105 × R 865 - R 560 - 305 × (R 665 - R 560)$

绿、红边和近红组合Green， red-edge and near-infrared combination

$T V I = 0.5 × 145 × R 842 - R 560 - 282 × R 705 - R 560$

$T V I = 0.5 × 145 × R 865 - R 560 - 305 × R 705 - R 560$

$T V I = 0.5 × 180 × R 842 - R 560 - 282 × R 740 - R 560$

$T V I = 0.5 × 180 × R 865 - R 560 - 305 × R 740 - R 560$

$T V I = 0.5 × 223 × R 842 - R 560 - 282 × R 783 - R 560$

$T V I = 0.5 × 223 × R 865 - R 560 - 305 × R 783 - R 560$

Table 2 Vegetation index formula

植被指数 Vegetation index	计算公式 Formula	参考文献 Reference
归一化植被指数Normalized vegetation index （NDVI）	$N D V I = (ρ n i r - ρ r e d) / (ρ n i r + ρ r e d)$	［19］
差值植被指数Difference vegetation index （DVI）	$D V I = ρ n i r - ρ r e d$	［19］
比值植被指数Rational vegetation index （RVI）	$R V I = ρ n i r / ρ r e d$	［19］
增强型植被指数Enhanced vegetation index （EVI）	$E V I = 2.5 × (ρ n i r - ρ r e d) / (ρ n i r + 6.0 × ρ r e d - 7.5 × ρ b l u e + 1)$	［20］
绿色归一化植被指数Green normalized difference vegetative index （GNDVI）	$G N D V I = (ρ n i r - ρ g r e e n) / (ρ n i r + ρ g r e e n)$	［21］
绿色叶绿素指数Green chlorophyll index （CI_green）	$C I g r e e n = ρ n i r ρ g r e e n - 1$	［22］
归一化差异红外指数Normalized difference infrared index （NDII）	$N D I I = (ρ n i r - ρ s w i r) / (ρ n i r + ρ s w i r)$	［23］
优化土壤调节植被指数Optimized soil adjusted vegetation index （OSAVI）	$O S A V I = (ρ n i r - ρ r e d) / (ρ n i r + ρ r e d + 0.16)$	［24］
土壤调节植被指数Soil adjusted vegetation index （SAVI）	$S A V I = ρ n i r - ρ r e d × (1 + L) / (ρ n i r + ρ g r e e n - L)$	［25］
修正土壤调节植被指数Modified soil adjusted vegetation index （MSAVI）	$M S A V I = 2 × ρ n i r + 1 - 2 × ρ n i r + 1 2 - 8 × (ρ n i r - ρ r e d) 2$	［26］
绿红植被指数Green red vegetation index （GRVI）	$G R V I = (ρ g r e e n - ρ r e d) / (ρ g r e e n + ρ r e d)$	［27］
全球植被湿度指数Global vegetation moisture index （GVMI）	$G V M I = ρ n i r + 0.1 - ρ s w i r + 0.02 ρ n i r + 0.1 + ρ s w i r + 0.02$	［28］

Table 2 Vegetation index formula

植被指数 Vegetation index	计算公式 Formula	参考文献 Reference
归一化植被指数Normalized vegetation index （NDVI）	$N D V I = (ρ n i r - ρ r e d) / (ρ n i r + ρ r e d)$	［19］
差值植被指数Difference vegetation index （DVI）	$D V I = ρ n i r - ρ r e d$	［19］
比值植被指数Rational vegetation index （RVI）	$R V I = ρ n i r / ρ r e d$	［19］
增强型植被指数Enhanced vegetation index （EVI）	$E V I = 2.5 × (ρ n i r - ρ r e d) / (ρ n i r + 6.0 × ρ r e d - 7.5 × ρ b l u e + 1)$	［20］
绿色归一化植被指数Green normalized difference vegetative index （GNDVI）	$G N D V I = (ρ n i r - ρ g r e e n) / (ρ n i r + ρ g r e e n)$	［21］
绿色叶绿素指数Green chlorophyll index （CI_green）	$C I g r e e n = ρ n i r ρ g r e e n - 1$	［22］
归一化差异红外指数Normalized difference infrared index （NDII）	$N D I I = (ρ n i r - ρ s w i r) / (ρ n i r + ρ s w i r)$	［23］
优化土壤调节植被指数Optimized soil adjusted vegetation index （OSAVI）	$O S A V I = (ρ n i r - ρ r e d) / (ρ n i r + ρ r e d + 0.16)$	［24］
土壤调节植被指数Soil adjusted vegetation index （SAVI）	$S A V I = ρ n i r - ρ r e d × (1 + L) / (ρ n i r + ρ g r e e n - L)$	［25］
修正土壤调节植被指数Modified soil adjusted vegetation index （MSAVI）	$M S A V I = 2 × ρ n i r + 1 - 2 × ρ n i r + 1 2 - 8 × (ρ n i r - ρ r e d) 2$	［26］
绿红植被指数Green red vegetation index （GRVI）	$G R V I = (ρ g r e e n - ρ r e d) / (ρ g r e e n + ρ r e d)$	［27］
全球植被湿度指数Global vegetation moisture index （GVMI）	$G V M I = ρ n i r + 0.1 - ρ s w i r + 0.02 ρ n i r + 0.1 + ρ s w i r + 0.02$	［28］

Table 3 Characteristics of vegetation and herbaceous sample area size

研究区 Study area	窗口大小（像元数） Window size （pixel number）	植被覆盖度 Vegetation coverage （%）	地表植被特点 Characteristics of vegetation
样区1 Sample area 1	26×26	7.62	沙质地表，有零星天然草本植被覆盖。Sandy surface with scattered natural herbaceous vegetation.
样区2 Sample area 2	26×26	29.83	沙质地表，稀疏天然草本植被覆盖。Sandy surface， sparsely covered by natural herbaceous vegetation.
样区3 Sample area 3	26×26	35.76	不均匀天然草本植被覆盖，有少量裸露地表。Uneven natural herb vegetation coverage， with a small amount of bare ground.
样区4 Sample area 4	26×26	51.52	3/4均匀天然草本植被覆盖，1/4不均匀少量裸露地表覆盖。3/4 uniform natural herbaceous vegetation coverage， 1/4 uneven small amount of bare ground coverage.
样区5 Sample area 5	26×26	69.22	人工草本植被覆盖为主，分布不均匀，且有大量空隙。Artificial herbaceous vegetation is mainly covered， with uneven distribution and a large number of gaps.
样区6 Sample area 6	26×26	84.87	人工草本植被覆盖为主，且有少量不均匀空隙。Artificial herbaceous vegetation is mainly covered with a small amount of uneven gaps.

Table 4 Characteristics of vegetation and shrubs sample area size

研究区 Study area	窗口大小（像元数） Window size （pixel number）	植被覆盖度 Vegetation coverage （%）	地表植被特点 Characteristics of vegetation
样区1 Sample area 1	26×26	6.04	沙质地表，有零星灌丛覆盖。Sandy surface， covered by scattered shrubs.
样区2 Sample area 2	26×26	12.58	沙质地表，少量不均匀灌丛覆盖。Sandy surface， covered by a small amount of uneven shrubs.
样区3 Sample area 3	26×26	15.58	1/2沙质地表，1/2不均匀灌丛覆盖。1/2 sandy surface， 1/2 uneven shrub coverage.
样区4 Sample area 4	26×26	19.42	3/4不均匀灌丛覆盖，1/4不均匀灌草混合覆盖。3/4 uneven shrub coverage， 1/4 uneven shrub grass mixed coverage.
样区5 Sample area 5	26×26	24.92	分布较均匀的灌丛覆盖为主，掺杂少量草本植被覆盖，且有大量空隙。It is mainly covered by bushes with relatively uniform distribution， mixed with a small amount of herbaceous vegetation， and has a large number of gaps.
样区6 Sample area 6	26×26	33.09	分布较均匀的灌草混合分布，且有少量空隙。Evenly distributed shrubs and grasses are mixed with a small amount of gaps.

Table 5 The aboveground biomass of herbaceous vegetation and shrub vegetation was linearly fitted with different TVI combinations

TVI组合 TVI combination	描述 Description	决定系数Coefficient of determination （R²）
TVI组合 TVI combination	描述 Description	草本植被Herbaceous vegetation	灌丛植被Shrub vegetation
绿、红和红边组合 Green， red and red-edge combination	TVI （ $R 560$ ， $R 665$ and $R 705$ ）	0.376	0.175
	TVI （ $R 560$ ， $R 665$ and $R 740$ ）	0.571	0.088
	TVI （ $R 560$ ， $R 665$ and $R 783$ ）	0.423	0.088
绿、红和近红组合 Green， red and near-infrared combination	TVI （ $R 560$ ， $R 665$ and $R 842$ ）	0.616	0.053
绿、红和近红组合 Green， red and near-infrared combination	TVI （ $R 560$ ， $R 665$ and $R 865$ ）	0.468	0.098
绿、红边和近红组合 Green， red-edge and near-infrared combination	TVI （ $R 560$ ， $R 705$ and $R 842$ ）	0.623	0.029
	TVI （ $R 560$ ， $R 705$ and $R 865$ ）	0.684	0.032
	TVI （ $R 560$ ， $R 740$ and $R 842$ ）	0.048	0.299
	TVI （ $R 560$ ， $R 740$ and $R 865$ ）	0.033	0.042
	TVI （ $R 560$ ， $R 783$ and $R 842$ ）	0.327	0.368
	TVI （ $R 560$ ， $R 783$ and $R 865$ ）	0.316	0.054

Table 5 The aboveground biomass of herbaceous vegetation and shrub vegetation was linearly fitted with different TVI combinations

TVI组合 TVI combination	描述 Description	决定系数Coefficient of determination （R²）
TVI组合 TVI combination	描述 Description	草本植被Herbaceous vegetation	灌丛植被Shrub vegetation
绿、红和红边组合 Green， red and red-edge combination	TVI （ $R 560$ ， $R 665$ and $R 705$ ）	0.376	0.175
	TVI （ $R 560$ ， $R 665$ and $R 740$ ）	0.571	0.088
	TVI （ $R 560$ ， $R 665$ and $R 783$ ）	0.423	0.088
绿、红和近红组合 Green， red and near-infrared combination	TVI （ $R 560$ ， $R 665$ and $R 842$ ）	0.616	0.053
绿、红和近红组合 Green， red and near-infrared combination	TVI （ $R 560$ ， $R 665$ and $R 865$ ）	0.468	0.098
绿、红边和近红组合 Green， red-edge and near-infrared combination	TVI （ $R 560$ ， $R 705$ and $R 842$ ）	0.623	0.029
	TVI （ $R 560$ ， $R 705$ and $R 865$ ）	0.684	0.032
	TVI （ $R 560$ ， $R 740$ and $R 842$ ）	0.048	0.299
	TVI （ $R 560$ ， $R 740$ and $R 865$ ）	0.033	0.042
	TVI （ $R 560$ ， $R 783$ and $R 842$ ）	0.327	0.368
	TVI （ $R 560$ ， $R 783$ and $R 865$ ）	0.316	0.054

Fig.4 Variation of soil noises 2σ in extracting vegetation information with the vegetation coverage

Fig.5 Leave-one-out cross validation results of optimized vegetation index GTVI and biomass modeling

Table 6 Cross validation result of different index models in herbaceous

植被指数 Vegetation index	拟合模型 Fit model	F	P	留一交叉验证结果Leave-one-out cross validation result
植被指数 Vegetation index	拟合模型 Fit model	F	P	$R C V 2$	$R M S E C V$
草地三角形植被指数GTVI	$y = 2.1613 x + 55.45$	178.964	0.000	0.773	41.083
绿色叶绿素指数CI_green	$y = 231.22 x - 88.551$	48.036	0.000	0.461	64.608
差值植被指数DVI	$y = 0.111 x - 58.268$	41.271	0.000	0.419	65.801
增强型植被指数EVI	$y = 128.63 x - 4.286$	149.495	0.000	0.736	44.165
绿色归一化植被指数GNDVI	$y = 974.36 x - 176.45$	42.791	0.000	0.422	65.690
绿红植被指数GRVI	$y = 1537 x + 173.83$	99.978	0.000	0.648	51.230
修正土壤调节植被指数MSAVI	$y = 521.13 x - 94.975$	58.020	0.000	0.514	60.252
归一化差异红外指数NDII	$y = 643.37 x + 86.327$	94.345	0.000	0.643	51.772
归一化植被指数NDVI	$y = 665.08 x - 58.357$	68.197	0.000	0.551	57.870
比值植被指数RVI	$y = 163.69 x - 174.83$	87.103	0.000	0.609	53.961
土壤调节植被指数SAVI	$y = 465 x - 70.313$	61.084	0.000	0.523	59.659
全球植被湿度指数GVMI	$y = 643.38 x + 86.32$	94.346	0.000	0.641	51.786
优化土壤调节植被指数OSAVI	$y = 665.1 x - 58.357$	68.197	0.000	0.552	57.875

Table 6 Cross validation result of different index models in herbaceous

植被指数 Vegetation index	拟合模型 Fit model	F	P	留一交叉验证结果Leave-one-out cross validation result
植被指数 Vegetation index	拟合模型 Fit model	F	P	$R C V 2$	$R M S E C V$
草地三角形植被指数GTVI	$y = 2.1613 x + 55.45$	178.964	0.000	0.773	41.083
绿色叶绿素指数CI_green	$y = 231.22 x - 88.551$	48.036	0.000	0.461	64.608
差值植被指数DVI	$y = 0.111 x - 58.268$	41.271	0.000	0.419	65.801
增强型植被指数EVI	$y = 128.63 x - 4.286$	149.495	0.000	0.736	44.165
绿色归一化植被指数GNDVI	$y = 974.36 x - 176.45$	42.791	0.000	0.422	65.690
绿红植被指数GRVI	$y = 1537 x + 173.83$	99.978	0.000	0.648	51.230
修正土壤调节植被指数MSAVI	$y = 521.13 x - 94.975$	58.020	0.000	0.514	60.252
归一化差异红外指数NDII	$y = 643.37 x + 86.327$	94.345	0.000	0.643	51.772
归一化植被指数NDVI	$y = 665.08 x - 58.357$	68.197	0.000	0.551	57.870
比值植被指数RVI	$y = 163.69 x - 174.83$	87.103	0.000	0.609	53.961
土壤调节植被指数SAVI	$y = 465 x - 70.313$	61.084	0.000	0.523	59.659
全球植被湿度指数GVMI	$y = 643.38 x + 86.32$	94.346	0.000	0.641	51.786
优化土壤调节植被指数OSAVI	$y = 665.1 x - 58.357$	68.197	0.000	0.552	57.875

Table 7 Cross validation result of different index models in shrubs

植被指数 Vegetation index	拟合模型 Fit model	F	P	留一交叉验证结果Leave-one-out cross validation result
植被指数 Vegetation index	拟合模型 Fit model	F	P	$R C V 2$	$R M S E C V$
草地三角形植被指数GTVI	$y = - 0.0013 x - 20.712$	14.238	0.001	0.338	69.912
绿色叶绿素指数CI_green	$y = 147.01 x + 152.94$	3.561	0.074	0.052	86.587
差值植被指数DVI	$y = 0.0662 x + 172.75$	2.566	0.125	0.011	87.775
增强型植被指数EVI	$y = 121.32 x + 198.67$	2.321	0.143	0.009	88.182
绿色归一化植被指数GNDVI	$y = 665.32 x + 84.066$	3.722	0.068	0.054	85.422
绿红植被指数GRVI	$y = 782.4 x + 321.2$	1.569	0.225	0.008	89.673
修正土壤调节植被指数MSAVI	$y = 309.5 x + 162.25$	2.818	0.109	0.019	87.235
归一化差异红外指数NDII	$y = 429.95 x + 270$	2.655	0.119	0.016	87.195
归一化植被指数NDVI	$y = 396.23 x + 184.4$	2.956	0.101	0.028	86.781
比值植被指数RVI	$y = 98.24 x + 113.74$	2.946	0.102	0.036	87.853
土壤调节植被指数SAVI	$y = 276.88 x + 176.32$	2.985	0.099	0.028	86.735
全球植被湿度指数GVMI	$y = 429.95 x + 270$	2.655	0.119	0.015	87.197
优化土壤调节植被指数OSAVI	$y = 396.24 x + 184.4$	2.959	0.101	0.029	86.780

Table 7 Cross validation result of different index models in shrubs

植被指数 Vegetation index	拟合模型 Fit model	F	P	留一交叉验证结果Leave-one-out cross validation result
植被指数 Vegetation index	拟合模型 Fit model	F	P	$R C V 2$	$R M S E C V$
草地三角形植被指数GTVI	$y = - 0.0013 x - 20.712$	14.238	0.001	0.338	69.912
绿色叶绿素指数CI_green	$y = 147.01 x + 152.94$	3.561	0.074	0.052	86.587
差值植被指数DVI	$y = 0.0662 x + 172.75$	2.566	0.125	0.011	87.775
增强型植被指数EVI	$y = 121.32 x + 198.67$	2.321	0.143	0.009	88.182
绿色归一化植被指数GNDVI	$y = 665.32 x + 84.066$	3.722	0.068	0.054	85.422
绿红植被指数GRVI	$y = 782.4 x + 321.2$	1.569	0.225	0.008	89.673
修正土壤调节植被指数MSAVI	$y = 309.5 x + 162.25$	2.818	0.109	0.019	87.235
归一化差异红外指数NDII	$y = 429.95 x + 270$	2.655	0.119	0.016	87.195
归一化植被指数NDVI	$y = 396.23 x + 184.4$	2.956	0.101	0.028	86.781
比值植被指数RVI	$y = 98.24 x + 113.74$	2.946	0.102	0.036	87.853
土壤调节植被指数SAVI	$y = 276.88 x + 176.32$	2.985	0.099	0.028	86.735
全球植被湿度指数GVMI	$y = 429.95 x + 270$	2.655	0.119	0.015	87.197
优化土壤调节植被指数OSAVI	$y = 396.24 x + 184.4$	2.959	0.101	0.029	86.780

Table 8 Analysis of GTVI precision improvement in medium vegetation coverage area

植被指数 Vegetation index	中等植被覆盖区 Moderate vegetation coverage area		低植被覆盖区与高植被覆盖区 Low vegetation coverage area and high vegetation coverage area
植被指数 Vegetation index	$R C V 2$	$R M S E C V$	$R C V 2$	$R M S E C V$
草地三角形植被指数GTVI	0.695	25.847	0.786	51.684
增强型植被指数EVI	0.580	30.339	0.764	54.200
绿红植被指数GRVI	0.492	33.372	0.656	58.105
全球植被湿度指数GVMI	0.333	38.247	0.693	61.911

Table 8 Analysis of GTVI precision improvement in medium vegetation coverage area

植被指数 Vegetation index	中等植被覆盖区 Moderate vegetation coverage area		低植被覆盖区与高植被覆盖区 Low vegetation coverage area and high vegetation coverage area
植被指数 Vegetation index	$R C V 2$	$R M S E C V$	$R C V 2$	$R M S E C V$
草地三角形植被指数GTVI	0.695	25.847	0.786	51.684
增强型植被指数EVI	0.580	30.339	0.764	54.200
绿红植被指数GRVI	0.492	33.372	0.656	58.105
全球植被湿度指数GVMI	0.333	38.247	0.693	61.911

References 34

1	Hu Z M， Fan J W， Zhong H P， et al. Progress on grassland underground biomass researches in China. Chinese Journal of Ecology， 2005（9）： 1095-1101.
	胡中民，樊江文，钟华平，等. 中国草地地下生物量研究进展. 生态学杂志， 2005（9）： 1095-1101.
2	Wang W， Peng S S， Fang J Y. Biomass distribution of natural grasslands and it response to climate change in North China. Arid Zone Research， 2008， 25（1）： 90-97.
	王娓，彭书时，方精云. 中国北方天然草地的生物量分配及其对气候的响应. 干旱区研究， 2008， 25（1）： 90-97.
3	Shao J. Monitoring of shrub encroachment into grassland by utilizing Sentinel-1 and Sentinel-2 jointly-Taking Xilingol grassland as an example. Beijing： University of Chinese Academy of Sciences， 2020.
	邵京. 联合Sentinel-1与Sentinel-2的草地灌丛化监测研究-以锡林郭勒盟草原为例. 北京：中国科学院大学， 2020.
4	Kawamura K， Akiyama T， Yokota H， et al. Monitoring of forage conditions with MODIS imagery in the Xilingol steppe， Inner Mongolia. International Journal of Remote Sensing， 2005， 26（7）： 1423-1436.
5	Ren H， Zhou G， Zhang F. Using negative soil adjustment factor in soil-adjusted vegetation index （SAVI） for aboveground living biomass estimation in arid grasslands. Remote Sensing of Environment， 2018， 209（2）： 439-445.
6	Guo C F， Chen W J， Niu M Y， et al. Collaborative estimation of aboveground biomass in grassland based on multiple vegetation index models. Agricultural Research in the Arid Areas， 2022， 40（4）： 206-213.
	郭超凡，陈雯璟，牛明艳，等. 基于多植被指数模型的草地地上生物量协同估算. 干旱地区农业研究， 2022， 40（4）： 206-213.
7	Zhao Y H， Hou M J， Feng Q S， et al. Estimation of aboveground biomass in Menyuan grassland based on Landsat 8 and random forest approach. Acta Prataculturae Sinica， 2022， 31（7）： 1-14.
	赵翊含，侯蒙京，冯琦胜，等. 基于Landsat 8和随机森林的青海门源天然草地地上生物量遥感估算. 草业学报， 2022， 31（7）： 1-14.
8	Guo Q X， Zhang F. Study on estimation of forest biomass based on remote sensing data. Journal of Northeast Forestry University， 2003， 31（2）： 13-16.
	国庆喜，张锋. 基于遥感信息估测森林的生物量. 东北林业大学学报， 2003， 31（2）： 13-16.
9	Wu Lan T Y， Bao G， Wu Yun D J， et al. Hyper-spectral remote sensing estimates of aboveground biomass of grassland. Journal of Inner Mongolia Normal University （Natural Science Edition）， 2015， 44（5）： 660-666.
	乌兰吐雅，包刚，乌云德吉，等. 草地地上生物量高光谱遥感估算研究. 内蒙古师范大学学报（自然科学汉文版）， 2015， 44（5）： 660-666.
10	Wang G X， Jing C Q， Dong P， et al. Study on biomass estimation and influencing factors of desert grassland in Xinjiang. Acta Agrestia Sinica， 2022， 30（7）： 1862-1872.
	王公鑫，井长青，董萍，等. 新疆荒漠草地生物量估算及影响因素研究. 草地学报， 2022， 30（7）： 1862-1872.
11	Ding R J. Intelligent recognition of targets in specific areas based on satellite remote sensing images. Guilin： Guangxi Normal University， 2022.
	丁锐进. 基于卫星遥感图像的特定地区目标智能识别. 桂林：广西师范大学， 2022.
12	Sun B， Li Z Y， Guo Z， et al. Comparison of sparse vegetation information estimation based on GF-1 and Landsat multi-spectral data. Remote Sensing Information， 2015， 30（5）： 48-56.
	孙斌，李增元，郭中，等. 高分一号与Landsat TM数据估算稀疏植被信息对比. 遥感信息， 2015， 30（5）： 48-56.
13	Yu H， Wu Y F， Jin Y， et al. Retrieving and spatiotemporal variation of aboveground biomass of grassland in arid area based on MODIS SWIR data. Remote Sensing Technology and Application， 2017， 32（3）： 524-530.
	于惠，吴玉锋，金毅，等. 基于MODIS SWIR数据的干旱区草地地上生物量反演及时空变化研究. 遥感技术与应用， 2017， 32（3）： 524-530.
14	Zhi R， Chen M M， Yan M， et al. Research on the influence of herdsmen’s family income under the grassland subsidy policy-Take Xilingol League， Inner Mongolia as an example. Acta Agrestia Sinica， 2022， 30（12）： 3392-3401.
	智荣，陈梅梅，闫敏，等. 草原补奖政策下牧户家庭收入的影响研究-以内蒙古锡林郭勒盟为例. 草地学报， 2022， 30（12）： 3392-3401.
15	Xu Q C. Effects of grassland carbon cycle on drought in Xilingol League. Qufu： Qufu Normal University， 2021.
	徐清宸. 锡林郭勒盟草地碳循环对干旱的响应. 曲阜：曲阜师范大学， 2021.
16	Rong R， Sun B， Wu Z T， et al. Study on above-ground biomass measurement of Caragana microphylla in shrub-encroached grassland. Acta Prataculturae Sinica， 2023， 32（1）： 36-47.
	戎荣，孙斌，武志涛，等. 灌丛化草原小叶锦鸡儿灌丛地上生物量测量方法研究. 草业学报， 2023， 32（1）： 36-47.
17	Jesús D， Jochem V， Luis A， et al. Evaluation of Sentinel-2 red-edge bands for empirical estimation of green LAI and chlorophyll content. Sensors， 2011， 11（7）： 7063-7081.
18	Xie E H， Wu J E， Yang K. Mass concentration inversion for chlorophyll a in Erhai lake based on Sentinel-2. Chinese Journal of Environmental Engineering， 2022， 16（9）： 3058-3069.
	谢恩弘，吴骏恩，杨昆. 基于Sentinel-2影像的洱海叶绿素a质量浓度反演. 环境工程学报， 2022，16（9）： 3058-3069.
19	Broge N H， Leblanc E. Comparing prediction power and stability of broadband and hyperspectral vegetation indices for estimation of green leaf area index and canopy chlorophyll density. Remote Sensing of Environment， 2001， 76（2）： 156-172.
20	Jiang Z Y， Huete A R， Didan K， et al. Development of a two-band enhanced vegetation index without a blue band. Remote Sensing of Environment， 2008， 112（10）： 3833-3845.
21	Huang J X， Wu J， Li C B， et al. Remote sensing retrieval of grassland above-ground biomass in Tianzhu County based on Sentinel-2 and Landsat 8 data. Acta Agrestia Sinica， 2021， 29（9）： 2023-2030.
	黄家兴，吴静，李纯斌，等. 基于Sentinel-2和Landsat 8数据的天祝县草地地上生物量遥感反演. 草地学报， 2021， 29（9）： 2023-2030.
22	Wu C， Niu Z， Gao S. The potential of the satellite derived green chlorophyll index for estimating midday light use efficiency in maize， coniferous forest and grassland. Ecological Indicators， 2011， 14（1）： 66-73.
23	Hardisky M A， Daiber F C， Roman C T， et al. Remote sensing of biomass and annual net aerial primary productivity of a salt marsh. Remote Sensing of Environment， 1984， 16（2）： 91-106.
24	Steven M D. The sensitivity of the OSAVI vegetation index to observational parameters. Remote Sensing of Environment， 1998， 63（1）： 49-60.
25	Huete A R. A soil-adjusted vegetation index （SAVI）. Remote Sensing of Environment， 1988， 25（3）： 295-309.
26	Qi J， Chehbouni A， Huete A R， et al. A modified soil adjusted vegetation index. Remote Sensing of Environment， 1994， 48（2）： 119-126.
27	Ren Z P， Gao R， Wang D Q. Identification and classification of rice lodging based on Sentinel-2 multispectral image. Water Saving Irrigation， 2022（7）： 44-50.
	任志鹏，高睿，王大庆. 基于哨兵2号多光谱影像的水稻倒伏识别与分类. 节水灌溉， 2022（7）： 44-50.
28	Ceccato P， Flasse S， Grégoire J M. Designing a spectral index to estimate vegetation water content from remote sensing data. Remote Sensing of Environment， 2022， 82（2）： 198-207.
29	Geisser S A. A predictive approach to the random effect model. Biometrika， 1974， 61（1）： 101-107.
30	Wen B M， Zhao L W， Huang L. Proof of the asymptotic equivalence between AIC criterion and LOOCV. Statistics & Decision， 2022， 38（6）： 40-43.
	文冰梅，赵联文，黄磊. AIC准则与留一法交叉验证渐近等价的证明. 统计与决策， 2022， 38（6）： 40-43.
31	Gao Z H， Li Z Y， Wei H D， et al. Study on the suitability of vegetation indices（Ⅵ） in arid area. Journal of Desert Research， 2006， 26（2）： 243-248.
	高志海，李增元，魏怀东，等. 干旱地区植被指数（Ⅵ）的适宜性研究. 中国沙漠， 2006， 26（2）： 243-248.
32	Zhang X X. A comparative study on different vegetation indices in capabilities for eliminating soil background noise and detecting vegetation coverage. Highlights of Sciencepaper Online， 2016， 9（2）： 126-132.
	张秀霞. 不同植被指数抵抗土壤背景噪声及探测植被覆盖能力的对比分析. 中国科技论文在线精品论文， 2016， 9（2）： 126-132.
33	Dong L X， Wu B F， Tang S H. Estimation of forest aboveground biomass by integrating GLAS and ETM data. Acta Scientiarum Naturalium Universitatis Pekinensis， 2011， 47（4）： 703-710.
	董立新，吴炳方，唐世浩. 激光雷达GLAS与ETM联合反演森林地上生物量研究. 北京大学学报（自然科学版）， 2011，47（4）： 703-710.
34	Ma C， Shi Y， Li M H， et al. Research of grassland biomass model based on NDVI in loess hilly and gully region-A case of Pengyang County， Ningxia. Ningxia Engineering Technology， 2017， 16（1）： 19-23.
	马超，石云，李梦华，等. 基于NDVI的黄土丘陵沟壑区草地生物量模型研究-以宁夏彭阳县为例. 宁夏工程技术， 2017， 16（1）： 19-23.