Estimation on seasonal dynamics of alpine grassland aboveground biomass using phenology camera-derived NDVI

doi:10.17521/cjpe.2020.0076

Abstract

Abstract:

Aims Accurate assessment of plant aboveground biomass is important for optimizing grassland resource management and for understanding the balance of carbon, water and energy fluxes in grassland ecosystems. This study constructed the optimal empirical models by near-surface remote sensing normalized difference vegetation index (NDVI) data, and then estimated plant aboveground biomass in an alpine grassland on the Qingzang Plateau.
Methods Using the dataset of both the field-measured aboveground biomass and the NDVI_RS observed by plant canopy spectrometer (RapidSCAN), we constructed the empirical models for estimating aboveground biomass in different phases of the growing season across 2018 and 2019. Using the NDVI_Cam time series observed by phenology camera and the estimated models, we simulated seasonal dynamics of aboveground biomass in 2018.
Important findings (1) The seasonal dynamics of NDVI_Cam, NDVI_RS and aboveground biomass exhibited a similar unimodal pattern; however, the timing of peak NDVI (August) preceded that of peak aboveground biomass (July). (2) The best model for estimating aboveground biomass is the power function in May, July and September, and the quadratic equation in June and August. The estimation accuracy ranged from 0.29 to 0.77. (3) The estimation of aboveground biomass based on the models in different phases of growing season (R² = 0.91) showed a higher accuracy compared to that based on the model at a single time (September)(R² = 0.49). Our results suggest that the near-surface remote sensing is an effective approach for estimating alpine grassland aboveground biomass, and further investigation on the seasonal growth of plants will help accurately evaluate grassland resources.

Key words: phenology camera, near-surface remote sensing, normalized difference vegetation index (NDVI), aboveground biomass, alpine grassland, Qingzang Plateau

CHEN Zhe, WANG Hao, WANG Jin-Zhou, SHI Hui-Jin, LIU Hui-Ying, HE Jin-Sheng. Estimation on seasonal dynamics of alpine grassland aboveground biomass using phenology camera-derived NDVI[J]. Chin J Plant Ecol, 2021, 45(5): 487-495.

Add to citation manager EndNote|Ris|BibTeX

URL: https://www.plant-ecology.com/EN/10.17521/cjpe.2020.0076

https://www.plant-ecology.com/EN/Y2021/V45/I5/487

Figures/Tables 7

Fig. 1 Seasonal dynamics of air temperature and precipitation, and the alpine grassland vegetation growth in different phases of the growing season at Haibei Station. A, Mean monthly air temperature and precipitation in 2018 and 2019. B-F, Pictures of plant growth from May to September as photographed by phenology camera. NIR, near-infrared images; RGB, red-green-blue images.

Fig. 2 Diurnal (A) and seasonal patterns (B) of normalized difference of vegetation index measured by NetCam (NDVICam) from May to September in 2018 at Haibei Station. The shaded part in A indicates that the NDVI of the phenology camera from 10:00-14:00 is the most stable.

Fig. 3 Dynamics of the normalized difference of vegetation index (NDVI) measured by RapidSCAN and aboveground biomass (mean ± SE) of alpine grassland in the growing seasons of 2018 (A) and 2019 (B) at Haibei Station.

Table 1 Pearson correlation coefficients between normalized difference of vegetation index measured by RapidSCAN and aboveground biomass of alpine grassland in different months of the growing season

月份 Month	样本数 No. of samples	r	p
5	56	0.80	<0.001
6	34	0.88	<0.001
7	64	0.67	<0.001
8	51	0.31	0.03
9	63	0.78	<0.001

Table 2 Fitted regression equations between normalized difference of vegetation index measured by RapidSCAN (NDVIRS)(x) and aboveground biomass (y) of alpine grassland across the growing seasons of 2018 and 2019

月份 Month	模型 Model	回归方程 Regression equation	R²	RMSE	RMSE_r (%)
5月 May (n = 56)	线性 Linear	y = 397.3x - 130.6	0.67	13.62	25.38
	对数 Logarithm	y = 186.0ln(x) + 198.3	0.65	14.32	26.68
	指数 Exponent	y = 2.0e^6.9^x	0.65	13.03	24.28
	乘幂 Power	y = 615.1x^3.3	0.67	12.45	23.19
	多项式 Quadratic	y = 491.7x² - 83.7x - 14.7	0.65	12.88	24.00
6月 June (n = 34)	线性 Linear	y = 1080.5x - 595.7	0.75	24.99	21.12
	对数 Logarithm	y = 730.4ln(x) + 423.1	0.73	26.82	22.66
	指数 Exponent	y = 0.7e^7.7^x	0.74	22.25	18.80
	乘幂 Power	y = 939.7x^5.2	0.74	21.95	18.55
	多项式 Quadratic	y = 3363.7x² - 3537.5x + 979.2	0.77	20.04	16.94
7月 July (n = 64)	线性 Linear	y = 952.0x - 508.5	0.72	39.94	18.54
	对数 Logarithm	y = 730.2ln(x) + 416.7	0.71	40.35	18.73
	指数 Exponent	y = 8.4e^4.2^x	0.73	38.00	17.64
	乘幂 Power	y = 517.6x^3.3	0.73	37.65	17.48
	多项式 Quadratic	y = 1422.3x² - 1248.1x + 339.51	0.72	38.95	18.08
8月 August (n = 51)	线性 Linear	y = 532.0x - 105.1	0.18	38.72	13.43
	对数 Logarithm	y = 375.4ln(x) + 403.0	0.16	39.01	13.53
	指数 Exponent	y = 81.9e^1.7^x	0.16	34.73	12.04
	乘幂 Power	y = 402.3x^1.2	0.14	35.28	12.23
	多项式 Quadratic	y = 5968.0x² - 8287.3x + 3134.3	0.29	31.70	10.99
9月 September (n = 63)	线性 Linear	y = 507.3x - 85.5	0.61	38.17	19.30
	对数 Logarithm	y = 292.5ln(x) + 371.7	0.61	38.18	19.31
	指数 Exponent	y = 48.0e^2.5^x	0.62	37.92	19.18
	乘幂 Power	y = 448.2x^1.4	0.63	37.68	19.06
	多项式 Quadratic	y = -48.0x² + 563.9x - 101.8	0.61	37.87	19.15

Fig. 4 Seasonal dynamics of alpine grassland biomass estimated by the normalized difference of vegetation index measured by NetCam time series and the models in different phases of growing season (A) and at a single time (September)(B) in 2018.

Fig. 5 Estimation of alpine grassland aboveground biomass using models in different phases of growing season (A) and at a single time (September)(B). The actual biomass is the averaged biomass for each measurement during the growing season of 2018 (n = 10). The estimated biomass is calculated by the optimal model and normalized difference of vegetation index measured by NetCam.

References 42

[1]	Beck PSA,Atzberger C,Arild Høgda K,Johansen B,Skidmore AK(2006).Improved monitoring of vegetation dynamics at very high latitudes: a new method using MODIS NDVI.Remote Sensing of Environment,100, 321-334. DOI URL
[2]	Busetto L,Colombo R,Migliavacca M,Cremonese E,Meroni M,Galvagno M,Rossini M,Siniscalco C,Morra Di Cella U,Pari E(2010).Remote sensing of larch phenological cycle and analysis of relationships with climate in the Alpine region.Global Change Biology,16, 2504-2517.
[3]	Chen DL,Xu BQ,Yao TD,Guo ZT,Cui P,Chen FH,Zhang RH,Zhang XZ,Zhang YL(2015).Assessment of past, present and future environmental changes on the Tibetan Plateau.Chinese Science Bulletin,60, 3023-3035.
	[陈德亮,徐柏青,姚檀栋,郭正堂,崔鹏,陈发虎,张人禾,张宪洲,张镱锂(2015).青藏高原环境变化科学评估: 过去、现在与未来.科学通报,60, 3023-3035.]
[4]	Chen XQ,Wang LH(2009).Progress in remote sensing phenological research.Progress in Geography,28, 33-40.
	[陈效逑,王林海(2009).遥感物候学研究进展.地理科学进展,28, 33-40.]
[5]	Chu D,Deji YZ,Pubu CR,Ji QM,Tang H(2013a).Aboveground biomass in the North Tibet and estimate model using remote sensing data.Journal of Natural Resources,28, 2000-2011.
	[除多,德吉央宗,普布次仁,姬秋梅,唐洪(2013a).藏北草地地上生物量及遥感监测模型研究.自然资源学报,28, 2000-2011.]
[6]	Chu D,Pubu CR,Deji YZ,Ji QM,Tang H(2013b).Aboveground biomass estimate methods of grassland in the Central Tibet.Journal of Mountain Science,31, 664-671.
	[除多,普布次仁,德吉央宗,姬秋梅,唐洪(2013b).西藏高原中部草地地上生物量遥感估算方法.山地学报,31, 664-671.]
[7]	Craine JM,Nippert JB,Elmore AJ,Skibbe AM,Hutchinson SL,Brunsell NA(2012).Timing of climate variability and grassland productivity.Proceedings of the National Academy of Sciences of the United States of America,109, 3401-3405.
[8]	Fan DQ,Zhao XS,Zhu WQ,Zheng ZT(2016).Review of influencing factors of accuracy of plant phenology monitoring based on remote sensing data.Progress in Geography,35, 304-319.
	[范德芹,赵学胜,朱文泉,郑周涛(2016).植物物候遥感监测精度影响因素研究综述.地理科学进展,35, 304-319.]
[9]	Filippa G,Cremonese E,Migliavacca M,Galvagno M,Sonnentag O,Humphreys E,Hufkens K,Ryu Y,Verfaillie J,Morra di Cella U,Richardson AD(2018).NDVI derived from near-infrared-enabled digital cameras: applicability across different plant functional types.Agricultural and Forest Meteorology,249, 275-285. DOI URL
[10]	Ge J,Meng BP,Yang SX,Gao JL,Yin JP,Zhang RP,Feng QS,Liang TG(2017).Monitoring of above-ground biomass in alpine grassland based on agricultural digital camera and MODIS remote sensing data: a case study in the Yellow River Headwater Region.Acta Prataculturae Sinica,26, 23-34.
	[葛静,孟宝平,杨淑霞,高金龙,殷建鹏,张仁平,冯琦胜,梁天刚(2017).基于ADC和MODIS遥感数据的高寒草地地上生物量监测研究——以黄河源区为例.草业学报,26, 23-34.]
[11]	He JS,Wang Z,Wang X,Schmid B,Zuo W,Zhou M,Zheng C,Wang M,Fang J(2006).A test of the generality of leaf trait relationships on the Tibetan Plateau.New Phytologist,170, 835-848. DOI URL
[12]	Huete A,Didan K,Miura T,Rodriguez EP,Gao X,Ferreira LG(2002).Overview of the radiometric and biophysical performance of the MODIS vegetation indices.Remote Sensing of Environment,83, 195-213. DOI URL
[13]	Jin Y,Yang X,Qiu J,Li J,Gao T,Wu Q,Zhao F,Ma H,Yu H,Xu B(2014).Remote sensing-based biomass estimation and its spatio-temporal variations in temperate grassland, northern China.Remote Sensing,6, 1496-1513. DOI URL
[14]	Jin YX,Xu B,Yang XC,Li JY,Wang DL,Ma HL(2011).Remote sensing dynamic estimation of grass production in Xilinguole, Inner Mongolia.Scientia Sinica Vitae,41, 1185-1195. DOI URL
	[金云翔,徐斌,杨秀春,李金亚,王道龙,马海龙(2011).内蒙古锡林郭勒盟草原产草量动态遥感估算.中国科学: 生命科学,41, 1185-1195.]
[15]	Klosterman S,Hufkens K,Richardson AD(2018).Later springs green-up faster: the relation between onset and completion of green-up in deciduous forests of North America.International Journal of Biometeorology,62, 1645-1655. DOI PMID
[16]	Liang TG,Yang SX,Feng QS,Liu BK,Zhang RP,Huang XD,Xie HJ(2016).Multi-factor modeling of above-ground biomass in alpine grassland: a case study in the Three-River Headwaters Region, China.Remote Sensing of Environment,186, 164-172. DOI URL
[17]	Liu A,Liu DF(2005).A review of grassland biomass research in China.Inner Mongolia Prataculture,17, 7-11.
	[刘艾,刘德福(2005).我国草地生物量研究概述.内蒙古草业,17, 7-11.]
[18]	Liu F,Wang CK,Wang XC(2018).Application of near-surface remote sensing in monitoring the dynamics of forest canopy phenology.Chinese Journal of Applied Ecology,29, 1768-1778.
	[刘帆,王传宽,王兴昌(2018).近地遥感在森林冠层物候动态监测中的应用.应用生态学报,29, 1768-1778.]
[19]	Main-Knorn M,Cohen WB,Kennedy RE,Grodzki W,Pflugmacher D,Griffiths P,Hostert P(2013).Monitoring coniferous forest biomass change using a Landsat trajectory-‌based approach.Remote Sensing of Environment,139, 277-290. DOI URL
[20]	Parton WJ,Scurlock JMO,Ojima DS,Schimel DS,Hall DO,Scopegram Group Members(1995).Impact of climate change on grassland production and soil carbon worldwide.Global Change Biology,1, 13-22. DOI URL
[21]	Petach AR,Toomey M,Aubrecht DM,Richardson AD(2014).Monitoring vegetation phenology using an infrared-enabled security camera.Agricultural and Forest Meteorology, 195-196, 143-151.
[22]	Piao S,Fang J,Zhou L,Tan K,Tao S(2007).Changes in biomass carbon stocks in Chinaʼs grasslands between 1982 and 1999.Global Biogeochemical Cycles,21, GB2002. DOI:10.1029/2005GB002634. DOI
[23]	Piao SL,Liu Q,Chen AP,Janssens IA,Fu YS,Dai JH,Liu LL,Lian X,Shen MG,Zhu XL(2019).Plant phenology and global climate change: current progresses and challenges.Global Change Biology,25, 1922-1940. DOI URL
[24]	Qiu J(2008).China: the third pole.Nature,454, 393-396. DOI URL
[25]	Scurlock JMO,Johnson K,Olson RJ(2002).Estimating net primary productivity from grassland biomass dynamics measurements.Global Change Biology,8, 736-753. DOI URL
[26]	Shen M,Piao S,Dorji T,Liu Q,Cong N,Chen X,An S,Wang S,Wang T,Zhang G(2015).Plant phenological responses to climate change on the Tibetan Plateau: research status and challenges.National Science Review,2, 454-467. DOI URL
[27]	Sonnentag O,Hufkens K,Teshera-Sterne C,Young AM,Friedl M,Braswell BH,Milliman T,O’Keefe J,Richardson AD(2012).Digital repeat photography for phenological research in forest ecosystems.Agricultural and Forest Meteorology,152, 159-177. DOI URL
[28]	Wang H,Liu H,Cao G,Ma Z,Li Y,Zhang F,Zhao X,Zhao X,Zhao X,Jiang L,Sanders NJ,Classen AT,He JS(2020).Alpine grassland plants grow earlier and faster but biomass remains unchanged over 35 years of climate change.Ecology Letters,23, 701-710. DOI PMID
[29]	Wang WF,Lei YC,Wang XF,Zhao HY(2008).A review of forest biomass models.Journal of Northwest Forestry University,23, 58-63.
	[王维枫,雷渊才,王雪峰,赵浩彦(2008).森林生物量模型综述.西北林学院学报,23, 58-63.]
[30]	Wang ZX,Liu C,Huete A(2003).From AVHRR-NDVI to MODIS-EVI: advances in vegetation index research.Acta Ecologica Sinica,23, 979-987.
	[王正兴,刘闯,Huete A(2003).植被指数研究进展: 从AVHRR-NDVI到MODIS-EVI.生态学报,23, 979-987.]
[31]	Wingate L,Ogée J,Cremonese E,Filippa G,Mizunuma T,Migliavacca M,Moisy C,Wilkinson M,Moureaux C,Wohlfahrt G,Hammerle A,Hörtnagl L,Gimeno C,Porcar- Castell A,Galvagno M,et al.(2015).Interpreting canopy development and physiology using a European phenology camera network at flux sites.Biogeosciences,12, 5995-6015. DOI URL
[32]	Xia CF,Li J,Liu QH(2013).Review of advances in vegetation phenology monitoring by remote sensing.Journal of Remote Sensing,17, 1-16.
	[夏传福,李静,柳钦火(2013).植被物候遥感监测研究进展.遥感学报,17, 1-16.]
[33]	Xu B,Yang XC,Tao WG,Miao JM,Yang Z,Liu HQ,Jin YX,Zhu XH,Qin ZH,Lv HY,Li JY(2013).MODIS-based remote-sensing monitoring of the spatiotemporal patterns of China’s grassland vegetation growth.International Journal of Remote Sensing,34, 3867-3878. DOI URL
[34]	Xu W,Zhu MY,Zhang Z,Ma ZZ,Liu HH,Chen LT,Cao GM,Zhao XQ,Schmid B,He JS(2018).Experimentally simulating warmer and wetter climate additively improves rangeland quality on the Tibetan Plateau.Journal of Applied Ecology,55, 1486-1497. DOI URL
[35]	Yahdjian L,Sala OE(2006).Vegetation structure constrains primary production response to water availability in the Patagonian Steppe.Ecology,87, 952-962. PMID
[36]	Yang J,Guo N,Huang LN,Jia JH(2008).Analyses on MODIS-NDVI index saturation in northwest China.Plateau Meteorology,27(4), 198-205.
	[杨嘉,郭铌,黄蕾诺,贾建华(2008).西北地区MODIS-NDVI指数饱和问题分析.高原气象,27(4), 198-205.]
[37]	Yang SX,Feng QS,Meng BP,Gao JL,Ge J,Liang TG(2018).Temporal and spatial dynamics of alpine grassland biomass in the Three-River Headwater Region.Pratacultural Science,35, 956-968.
	[杨淑霞,冯琦胜,孟宝平,高金龙,葛静,梁天刚(2018).三江源地区高寒草地地上生物量时空动态变化.草业科学,35, 956-968.]
[38]	Yang YH(2018).Ecological processes in alpine ecosystems under changing environment.Chinese Journal of Plant Ecology,42, 1-5. DOI URL
	[杨元合(2018).全球变化背景下的高寒生态过程.植物生态学报,42, 1-5.]
[39]	Zhang B,Zhang L,Xie D,Yin X,Liu C,Liu G(2016).Application of synthetic NDVI time series blended from Landsat and MODIS data for grassland biomass estimation.Remote Sensing,8, 10. DOI:10.3390/rs8010010. DOI URL
[40]	Zhang XC,Zhu HZ,Zhong HP,Cheng YD,Jin GL,Shao XM(2015).Assessment of above-ground biomass of grassland using remote sensing, Yili, Xinjiang.Acta Prataculturae Sinica, (6), 25-34.
	[张旭琛,朱华忠,钟华平,程耀东,靳瑰丽,邵小明(2015).新疆伊犁地区草地植被地上生物量遥感反演.草业学报, (6), 25-34.]
[41]	Zhou YT,Fu G,Shen ZX,Zhang XZ,Wu JS,Li YL,Yang PW(2013).Estimation model of aboveground biomass in the Northern Tibet Plateau based on remote sensing date.Acta Prataculturae Sinica,22, 120-129.
	[周宇庭,付刚,沈振西,张宪洲,武建双,李云龙,杨鹏万(2013).藏北典型高寒草甸地上生物量的遥感估算模型.草业学报,22, 120-129.]
[42]	Zhu X,Liu D(2015).Improving forest aboveground biomass estimation using seasonal Landsat NDVI time-series.ISPRS Journal of Photogrammetry and Remote Sensing,102, 222-231. DOI URL