RELATIONSHIP BETWEEN CANOPY HYPERSPECTRA PARAMETER AND LEAF NITROGEN CONCENTRATION IN COTTON

doi:10.17521/cjpe.2007.0114

Abstract

Abstract:

Aims It is very important to develop a nondestructive method for estimating nitrogen status and growth characters in cotton. We carried out two experiments to investigate the quantitative relationship between leaf nitrogen concentration and the canopy hyperspectral parameter in cotton.

Methods Based on the canopy hyperspectral reflectance and derived hyperspectral parameter, we investigated the quantitative relationship between leaf nitrogen concentration and canopy reflectance spectra in different cotton cultivars under different N rates and put forward the sensitive parameters and monitoring equations for leaf nitrogen concentration. Experiment 1 was conducted with two cultivars (‘Sumian 12’ and ‘Zhongmian 29’) and four N application levels (0, 150, 300 and 450 kg·hm^-2) in 2004. Experiment 2 included two cultivars (‘Kemian 1’ and ‘Meimian 33B’) with three nitrogen levels (0, 240 and 480 kg·hm^-2) in 2005.

Important findings Leaf nitrogen concentration and canopy hyperspectral reflectance significantly changed with levels of nitrogen fertilization. The sensitive bands of leaf nitrogen concentration were 600-700 nm of visible light and 750-900 nm of near infrared light, and the quantitative relationships between leaf nitrogen concentration and ratio vegetation index (RVI) [average (760-850), 700] of canopy were significant with an average R² of 0.70 in four cultivars. An integrated regression equation could be used for describing the dynamic change pattern of leaf nitrogen concentration with hyperspectral parameters in different varieties, growing stages and nitrogen levels in cotton. These results provide a technical approach for monitoring plant nitrogen status and guiding precision nitrogen management in cotton production.

Key words: cotton, leaf, nitrogen concentration, hyperspectrum, canopy reflectance, ratio vegetation index, quantitative relationship

WU Hua-Bing, ZHU Yan, TIAN Yong-Chao, YAO Xia, LIU Xiao-Jun, ZHOU Zhi-Guo, CAO Wei-Xing. RELATIONSHIP BETWEEN CANOPY HYPERSPECTRA PARAMETER AND LEAF NITROGEN CONCENTRATION IN COTTON[J]. Chin J Plant Ecol, 2007, 31(5): 903-909.

Add to citation manager EndNote|Ris|BibTeX

URL: https://www.plant-ecology.com/EN/10.17521/cjpe.2007.0114

https://www.plant-ecology.com/EN/Y2007/V31/I5/903

Figures/Tables 7

Table 1 Algorithm of different hyperspectral parameters

高光谱参数 Hyperspectral parameter	计算公式 Algorithm formula	文献 Reference
吸收谷深度 Depth of absorption valley (VD_i)	VD_i=1- $R Ci (R Si + (R Ei - R Si) * (λ Ci - λ Si) / (λ Ei - λ Si))$	Baret et al. (1989)
吸收谷特征面积 Feature area of absorption valley (V-Area_i)	V-Area_i= $∫ λ Si λ Ei$ (R_line(λ)-R(λ))dλ = $∫ λ Si λ Ei$ ((R_Si+ $R Ei - R Si λ Ei - λ Si$ (λ-λ_Si*))-R(λ))dλ	Baret et al. (1989)
归一化吸收深度 Normalized depth of valley (NVD_i)	NVD_i=VD_i/V-Area_i	Baret et al. (1989)
比值植被指数 Ratio vegetation index (RVI)	RVI=IR/R	Pearson & Miller ( 1972)
光化学反射指数 Photochemical reflectance index (PRI)	PRI=(R₅₇₀-R₅₃₁)/(R₅₇₀+R₅₃₁)	Pe?uelas et al. (1995)
绿度归一化差值植被指数 Green normalized difference vegetation index (GNDVI)	GNDVI=(R₇₅₀-R₅₅₀)/(R₇₅₀+R₅₅₀)	Gitelson & Merzlyak ( 1996)

Table 1 Algorithm of different hyperspectral parameters

高光谱参数 Hyperspectral parameter	计算公式 Algorithm formula	文献 Reference
吸收谷深度 Depth of absorption valley (VD_i)	VD_i=1- $R Ci (R Si + (R Ei - R Si) * (λ Ci - λ Si) / (λ Ei - λ Si))$	Baret et al. (1989)
吸收谷特征面积 Feature area of absorption valley (V-Area_i)	V-Area_i= $∫ λ Si λ Ei$ (R_line(λ)-R(λ))dλ = $∫ λ Si λ Ei$ ((R_Si+ $R Ei - R Si λ Ei - λ Si$ (λ-λ_Si*))-R(λ))dλ	Baret et al. (1989)
归一化吸收深度 Normalized depth of valley (NVD_i)	NVD_i=VD_i/V-Area_i	Baret et al. (1989)
比值植被指数 Ratio vegetation index (RVI)	RVI=IR/R	Pearson & Miller ( 1972)
光化学反射指数 Photochemical reflectance index (PRI)	PRI=(R₅₇₀-R₅₃₁)/(R₅₇₀+R₅₃₁)	Pe?uelas et al. (1995)
绿度归一化差值植被指数 Green normalized difference vegetation index (GNDVI)	GNDVI=(R₇₅₀-R₅₅₀)/(R₇₅₀+R₅₅₀)	Gitelson & Merzlyak ( 1996)

Table 2 Leaf nitrogen concentrations of different cotton varieties at flowering, full-blossoming, early opening boll, opening boll and full opening boll stages under different N rates

年份 Year	品种 Variety	施氮水平 N rate (kg N·hm^-2)	代号 Code	叶片氮含量 Leaf nitrogen concentration (mg·g^-1)
年份 Year	品种 Variety	施氮水平 N rate (kg N·hm^-2)	代号 Code	开花 Flowering	盛花 Full-blossoming	始絮 Early opening boll	吐絮 Opening boll	盛絮 Full opening boll
2004	‘苏棉12’ ‘Sumian 12’	0	N₀	20.50	24.52	22.95	22.61	21.70
		150	N₁	24.65	30.97	29.74	28.44	23.47
		300	N₂	27.16	33.67	32.46	29.51	29.10
		450	N₃	32.82	34.88	31.72	31.91	30.37
		F值 F Value		53.07^**	17.45^*	9.20	24.82^*	38.91^**
	‘中棉29’ ‘Zhongmian 29’	0	N₀	21.17	25.94	24.01	24.52	16.95
		150	N₁	25.04	30.04	27.27	25.37	23.70
		300	N₂	28.46	34.75	31.62	27.99	28.62
		450	N₃	30.59	33.70	33.37	31.19	29.92
		F值 F Value		9.44^*	14.80^*	22.50^*	98.66^**	19.01^*
2005	‘科棉1号’ ‘Kemian 1’	0	N₀	-	26.38	23.00	23.92	22.16
		240	N₄	-	33.65	30.37	28.46	27.74
		480	N₅	-	35.34	33.16	34.16	32.18
		F值 F Value		-	40.43^*	223.45^**	47.85^*	45.57^*
	‘美棉33B’ ‘Meimian 33B’	0	N₀	-	25.39	21.48	23.25	23.90
		240	N₄	-	33.57	30.73	27.95	25.54
		480	N₅	-	37.50	36.82	36.35	28.64
		F值 F Value		-	28.55^*	169.91^**	222.30^**	22.82^*

Fig.1 Hyperspectral reflectance of cotton canopy under different N rates (‘Kemian 1’,2005)

Fig.2 Correlation of leaf nitrogen concentration to canopy hyperspectral reflectance with different cotton cultivars

Table 3 Quantitative relationships of leaf nitrogen concentration (y) to main hyperspectral parameter (x) in different cotton cultivars

品种类型 Cultivar type	高光谱参数 Hyperspectral parameter	回归方程 Regression equation	决定系数 R²	样本数 n
‘科棉1号’ `Kemian 1'	Average (760~850)/R₇₀₀	y=0.234 4x-1.208 1	0.807^**	24
	Average (760~850)/R₅₅₀	y=0.162 6x+1.036 7	0.761^**	24
	NVD672	y=0.019 4x-0.265 7	0.760^**	24
	Average (760~850)/Average (510~560)	y=0.195 9x+1.342	0.749^**	24
	PRI	y=-0.003 1x+0.084 5	0.724^**	24
	GNDVI	y=0.007 4x+0.470 2	0.723^**	24
‘美棉33B’ `Meimian 33B'	Average (760~850)/R₇₀₀	y=0.2x-0.107 9	0.804^**	23
	PRI	y=-0.002 6x+0.069 1	0.772^**	23
	NVD672	y=0.015 9x-0.207 1	0.750^**	23
	Average (760~850)/Average (510~560)	y=0.168 4x+2.228 9	0.737^**	23
	Average (760~850)/R₅₅₀	y=0.136 3x+1.912 2	0.734^**	23
	GNDVI	y=0.005 8x+0.517 8	0.659^**	23
‘中棉29’ `Zhongmian 29'	NVD672	y=0.021 4x-0.279 9	0.614^**	50
	Average (760~850)/R₇₀₀	y=0.160 8x-0.347 1	0.544^**	50
	PRI	y=-0.003 6x+0.121	0.528^**	49
‘苏棉12’ `Sumian 12'	PRI	y=-0.004 5x+0.145 4	0.679^**	39
	NVD672	y=-0.000 1x+0.011 5	0.626^**	39
	Average (760~850)/R₇₀₀	y=0.231 6x-2.115 7	0.616^**	39

Fig.3 Relationships of leaf nitrogen concentration to ratio vegetation index (RVI [average (760-850),700]) of hyperspectral index with different cotton cultivars

Fig.4 Relationships of leaf nitrogen concentration to ratio vegetation index (RVI [average (760-850), 700]) of hyperspectral index with combined cotton cultivars

References 21

[1]	Baret F, Guyot G, Major DJ (1989). TSAVI: a vegetation index which minimizes soil brightness effects on LAI and APAR estimation. In: Proceedings of the 12th Canadian Symposium on Remote Sensing and IGARSS'89. Geoscience and Remote Sensing Society of Institute of Electrical and Electronics Engineers. Vancouver, Canada, 3,1355-1358.
[2]	Blackmer TM, Schepers JS, Meyer GE (1994). Remote sensing to detect nitrogen deficiency in corn. In: Robert PC, Rust RH, Larson WEeds. Proceedings of the Second International Conference on Site-specific Management for Agricultural Systems. ASA/CSSA/SSSA, Madison, WI,505-512.
[3]	Blackmer TM, Schepers JS, Varvel GE, Walter-Shea EA (1996). Nitrogen deficiency detection using reflected shortwave radiation from irrigated corn canopies. Agronomy Journal, 88,1-5. DOI URL
[4]	Bronson KF, Chua TT, Booker JD, Keeling JW, Lascano RJ (2003). In-season nitrogen status sensing in irrigated cotton.Ⅱ.Leaf nitrogen and biomass. Soil Science Society of America Journal, 67,1439-1448.
[5]	Gitelson AA, Kaufman YJ, Stark R, Rundquist D (2002). Novel algorithms for remote estimation of vegetation fraction. Remote Sensing of Environment, 80,76-87.
[6]	Gitelson AA, Merzlyak MN (1996). Signature analysis of leaf reflectance spectra: algorithm development for remote sensing of chlorophyll. Journal of Plant Physiology, 148,494-500.
[7]	Jiang GY (蒋桂英), Li SK (李少昆), Wang DW (王登伟), Chen YF (陈永芳), Lei YW (雷永雯) (2002). Studies and advances on application of remote sensing technique to cotton. Journal of Xinjiang Agricultural University (新疆农业大学学报), 25(3),76-79. (in Chinese with English abstract)
[8]	Li Q (李强), Zhao W (赵伟) (2001). Date Processing and Application in MATLAB (MATLAB数据处理与应用). National Defense Industry Press, Beijing,1.
[9]	Maas SJ (1997). Structure and reflectance of irrigated cotton leaf canopies. Agronomy Journal, 89,54-59.
[10]	Ma YQ (马亚琴), Bao AM (包安明), Wang DW (王登伟), Sun L (孙莉), Huang CY (黄春燕), Feng XW (冯宪伟), Xiao CH (肖春华), Yang XJ (杨新军) (2003). Estimating of the total nitrogen content (TN) in cotton canopy by using hyperspectral remote sensing under water stress. Arid Land Geography (干旱区地理), 4,26-30. (in Chinese with English abstract)
[11]	Mutanga O, Skidmore AK, van Wieren S (2003). Discriminating tropical grass ( Cenchrus ciliaris) canopies grown under different nitrogen treatments using spectroradiometry. ISPRS Journal of Photogrammetry and Remote Sensing, 57,263-272.
[12]	Pearson RL, Miller DL (1972). Remote mapping of standing crop biomass for estimation of the productivity of the short-grass prairie, Pawnee national grasslands, Colorado. In: Proceeding of the 8th International Symposium on Remote Sensing of Environment. Environment Research Institute of Michigan, Ann Arbor, Michigan, USA,1357-1381.
[13]	Peñuelas J, Filella I, Gamon JA (1995). Assessment of photosynthetic radiation-use-efficiency with spectral reflectance. New Phytologist, 131, 291-296.
[14]	Stone ML, Solie JB, Raun WR, Whitney RW, Taylor SL, Ringer JD (1996). Use of spectral radiance for correcting in-season fertilizer nitrogen deficiencies in winter wheat. Transactions of the American Society of Agricultural Engineers, 39,1623-1631.
[15]	Thomas JR, Gausman HW (1977). Leaf reflectance vs. leaf chlorophyll and carotenoid concentration for eight crops. Agronomy Journal, 69,799-802.
[16]	Wang DW (王登伟), Li SK (李少昆), Tian QJ (田庆玖), Huang CY (黄春燕), Cao LP (曹连莆), Xiao CH (肖春华), Ma YQ (马亚琴), Yang XJ (杨新军) (2003). Estimating of main cultivation physiology parameters of cotton by using hyperspectral remote sensing. Scientia Agricultura Sinica (中国农业科学), 36,770-774. (in Chinese with English abstract)
[17]	Wang RC (王人潮), Chen MZ (陈铭臻), Jiang HX (蒋亨显) (1993). Studies on agronomic mechanism of the rice yield estimation by remote sensing.Ⅰ. The rice reflectance characteristics of different nitrogen levels and the selection of their sensitive bands. Journal of Zhejiang University(Agriculture and Life Sciences) (浙江农业大学学报(农业与生命科学版)), 19(Suppl.),7-14. (in Chinese with English abstract)
[18]	Xue LH (薛利红), Cao WX (曹卫星), Luo WH (罗卫红), Jiang D (姜东), Meng YL (孟亚利), Zhu Y (朱艳) (2003). Diagnosis of nitrogen status in rice leaves with the canopy spectral reflectance. Scientia Agricultura Sinica (中国农业科学), 36,807-812. (in Chinese with English abstract)
[19]	Xue LH (薛利红), Cao WX (曹卫星), Luo WH (罗卫红), Zhang X (张宪) (2004). Correlation between leaf nitrogen status and canopy spectral characteristics in wheat. Acta Phytoecologica Sinica (植物生态学报), 28,172-177. (in Chinese with English abstract)
[20]	Zhao DL, Reddy KR, Kakani VG, Read JJ, Koti S (2005). Selection of optimum reflectance ratios for estimating leaf nitrogen and chlorophyll concentrations of field-grown cotton. Agronomy Journal, 97,89-98.
[21]	Zhou QF (周启发), Wang RC (王人潮) (1993). A preliminary study on the relationship between the nitrogen levels and the spectral characteristics of early rice leaves. Journal of Zhejiang University(Agriculture and Life Sciences) (浙江农业大学学报(农业与生命科学版)), 19(Suppl.),40-46. (in Chinese with English abstract)