祁连山国家公园森林地上碳密度遥感估算

doi:10.12118/j.issn.1000–6060.2021.04.17

摘要/Abstract

摘要：

准确、及时地绘制森林地上碳密度图是了解全球碳循环的必要条件。虽然星载激光雷达（如ICESat/GLAS）数据已被广泛用于估算大尺度的森林地上碳密度分布,但地形坡度对GLAS提取冠层高度精度的影响往往限制了其在山区森林的应用。通过以祁连山国家公园为研究区域,结合GLAS数据、Landsat OLI（Operational land imagery）数据、样地调查数据,对祁连山地区进行区域性的森林地上碳密度估算。首先通过改进后的地形校正模型减小坡度对GLAS数据提取森林冠层高度精度的影响,使得更多的GLAS数据可用于后续的研究;其次将所建立的不同类型森林地上生物量估算模型与相关碳含量转换系数结合,得到GLAS激光光斑（脚印点）的森林地上碳密度;最后利用非参数化算法最大熵（MaxEnt）模型得到祁连山国家公园2018年森林地上碳密度分布图。结果表明：祁连山国家公园2018年平均森林地上碳密度为40.72±6.72 t·hm^-2,总蓄积量为28.58±4.72 Tg,海拔2770~3770 m区域的森林植被碳储量最大,且阴坡的碳储量明显高于阳坡。采用森林资源清查数据独立验证估算结果的准确性,模型估测均方根误差（Root mean square error,RMSE）为18.946 t·hm^-2。本研究结果可以为监测区域乃至国家尺度的森林碳储量变化以及制定可持续的森林管理措施提供依据。此外,本文所采用的方法在山区森林碳储量估算方面也具有较大的潜力。

关键词: 星载激光雷达, Landsat OLI, 森林地上碳密度, 非参数化算法, 祁连山

Abstract:

An accurate, up-to-date, and spatially explicit estimate of the forest aboveground carbon (AGC) density is indispensable for understanding the global carbon cycle and implementing measures to mitigate climate change such as reducing emissions from deforestation and forest degradation (REDD+). Geoscience Laser Altimeter System (GLAS) data have proven to be a powerful candidate for estimating the forest AGC density distribution over large-scale areas because of its global-scale sampling strategy and freely available data. However, the influence of terrain slopes on the accuracy of the GLAS-derived canopy height often limits its utility in mountainous regions. Nevertheless, mountain forests play a key role in carbon storage because of the relatively small influence of human activities, and the carbon sequestration capacities of forests differ according to the region. In this study, we integrated GLAS data, Landsat 8 OLI imagery, and field inventory data to estimate the forest AGC density distribution of Qilian Mountains National Park in northwestern China, which is an essential ecological barrier and carbon sequestration site in western China. We aimed to address the underestimation of carbon storage by mountain forests in previous studies by improving the estimation accuracy of the GLAS-derived canopy height and in turn the forest AGC density in the study area. We first improved upon a physical terrain correction model to derive the canopy height from GLAS data according to five different slope gradient classes. The canopy height could be reliably estimated with a root-mean-squared error (RMSE) of 1.84-3.91 m on the basis of independent validation with forest inventory data. We subsequently calculated the AGC density of GLAS footprints located in different forest types as accurately as possible by using an aboveground biomass estimation model of different forest types and a carbon content conversion factor, as well as the stand density recorded in the forest inventory data. We took a nonparametric approach and applied a MaxEnt model to extrapolate AGC data at the GLAS footprint scale to the whole forest as extracted from Landsat 8 OLI imagery. We also made great efforts to spatiotemporally match all data sources in this study. The results showed that the average forest AGC density in Qilian Mountains National Park was 40.72±6.72 t·hm^-2 in 2018, and the total aboveground carbon storage was 28.58±4.72 Tg. Forest vegetation in areas at 2770-3770 m above sea level had the largest carbon storage, and shaded slopes had significantly greater carbon storage than sunny slopes. The accuracy of the estimation results was independently verified by comparison with the forest resource inventory data and resulted in an RMSE of 18.946 t·hm^-2, which meets the measuring, reporting, and verification guidelines of REDD+. The results of this study can provide a basis for monitoring regional- and even national-scale changes in forest carbon reserves and for formulating sustainable forest management policies. Additionally, the method used in this study can potentially be applied to estimating the carbon storage of forests in mountainous areas.

Key words: satellite LiDAR system, Landsat OLI, forest aboveground carbon density, non-parametric algorithm, Qilian Mountains

宋洁,刘学录. 祁连山国家公园森林地上碳密度遥感估算[J]. 干旱区地理, 2021, 44(4): 1045-1057.

SONG Jie,LIU Xuelu. Estimation of forest aboveground carbon density in Qilian Mountains National Park based on remote sensing[J]. Arid Land Geography, 2021, 44(4): 1045-1057.

图/表 10

图1

图2

图3

表1

图4

表2

祁连山地区相关树种异速方程和碳含量转换系数"

树种	异速生长方程	参考文献	碳含量转换系数
青海云杉 Picea crassifolia	$W S = 0.0478 D 2 H 0.8665$ $W B = 0.0122 D 2 H 0.8905$ $W L = 0.2650 D 2 H 0.4701$ $W = W S + W B + W L$	王金叶等^[26]	0.5200^[26]
祁连圆柏 Juniperus przewalskii	$W S = 0.2738 D 2 H 0.6912$ $W B = 0.0061 D 2 H 0.9455$ $W L = 0.0042 D 2 H 0.8986$ $W = W S + W B + W L$	王金叶等^[27]	0.4650^[28]
侧柏 Platycladus orientalis	$W S = 0.0427 D 2 H 0.8545$ $W B = 0.0128 D 2 H 0.8916$ $W L = 0.0258 D 2 H 0.7037$ $W = W S + W B + W L$	关晋宏等^[29]	0.5034^[30]
油松 Pinus tabulaeformis	$W S = e - 3.8828 D 2 H 0.9359$ $W B = e - 6.3807 D 2 H 1.1242$ $W L = e - 5.3277 D 2 H 0.8812$ $W = W S + W B + W L$	程堂仁等^[31]	0.5108^[32]
蒙古栎 Quercus mongolica	$W S = 0.0493 D 2 H 0.8514$ $W B = 0.0026 D 2 H 1.1887$ $W L = 0.0119 D 2 H 0.8350$ $W = W S + W B + W L$	关晋宏等^[29]	0.5004^[30]
山杨 Populus davidiana	$W S = e - 3.8023 D 2 H 0.9631$ $W B = e - 5.9070 D 2 H 1.0903$ $W L = e - 3.9108 D 2 H 0.6104$ $W = W S + W B + W L$	程堂仁等^[31]	0.4956^[30]
白桦 Betula platyphylla	$W S = e - 3.8023 D 2 H 0.9631$ $W B = e - 5.9070 D 2 H 1.0903$ $W L = e - 3.9108 D 2 H 0.6104$ $W = W S + W B + W L$	程堂仁等^[31]	0.5025^[32]

表2

图5

表3

图6

图7

参考文献 45

[1]	Pan Y, Birdsey R A, Fang J, et al. A large and persistent carbon sink in the world’s forests[J]. Science, 2011, 333(6045):988-993. doi: 10.1126/science.1201609
[2]	Houghton R A, House J I, Pongratz J, et al. Carbon emissions from land use and land-cover change[J]. Biogeosciences, 2012, 9(12):5125-5142. doi: 10.5194/bg-9-5125-2012
[3]	Shang X, Chazette P. Interest of a full-waveform flown UV Lidar to derive forest vertical structures and aboveground carbon[J]. Forests, 2014, 5(6):1454-1480. doi: 10.3390/f5061454
[4]	Narine L L, Popescu S, Neuenschwander A, et al. Estimating aboveground biomass and forest canopy cover with simulated ICESat-2 data[J]. Remote Sensing of Environment, 2019, 224:1-11. doi: 10.1016/j.rse.2019.01.037
[5]	Neigh C S R, Nelson R F, Ranson K J, et al. Taking stock of circumboreal forest carbon with ground measurements, airborne and spaceborne LiDAR[J]. Remote Sensing of Environment, 2013, 137:274-287. doi: 10.1016/j.rse.2013.06.019
[6]	Wulder M A, White J C, Nelson R F, et al. Lidar sampling for large-area forest characterization: A review[J]. Remote Sensing of Environment, 2012, 121:196-209. doi: 10.1016/j.rse.2012.02.001
[7]	Nelson R, Boudreau J, Gregoire T G, et al. Estimating Quebec provincial forest resources using ICESat/GLAS[J]. Canadian Journal of Forest Research, 2009, 39(4):862-881. doi: 10.1139/X09-002
[8]	Baccini A, Goetz S J, Walker W S, et al. Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps[J]. Nature Climate Change, 2012, 2(3):182-185. doi: 10.1038/nclimate1354
[9]	Tyukavina A, Stehman S V, Potapov P V, et al. National-scale estimation of gross forest aboveground carbon loss: A case study of the Democratic Republic of the Congo[J]. Environmental Research Letters, 2013, 8(4):044039, doi: 10.1088/1748-9326/8/4/044039. doi: 10.1088/1748-9326/8/4/044039
[10]	池泓, 黄进良, 邱娟, 等. GLAS星载激光雷达和Landsat/ETM+数据的森林生物量估算[J]. 测绘科学, 2018, 43(4):9-16, 23.
	[ Chi Hong, Huang Jinliang, Qiu Juan, et al. Estimation of forest aboveground biomass using ICESat/GLAS data and Landsat/ETM+ imagery[J]. Science of Surveying and Mapping, 2018, 43(4):9-16, 23. ]
[11]	Hilbert C, Schmullius C. Influence of surface topography on ICESat/GLAS forest height estimation and waveform shape[J]. Remote Sensing, 2012, 4(8):2210-2235. doi: 10.3390/rs4082210
[12]	Saatchi S S, Harris N L, Brown S, et al. Benchmark map of forest carbon stocks in tropical regions across three continents[J]. Proceedings of the National Academy of Sciences of the United States of America, 2011, 108(24):9899-9904. doi: 10.1073/pnas.1019576108 pmid: 21628575
[13]	Huang H, Liu C, Wang X, et al. Integration of multi-resource remotely sensed data and allometric models for forest aboveground biomass estimation in China[J]. Remote Sensing of Environment, 2019, 221:225-234. doi: 10.1016/j.rse.2018.11.017
[14]	Wang Y, Li G, Ding J, et al. A combined GLAS and MODIS estimation of the global distribution of mean forest canopy height[J]. Remote Sensing of Environment, 2016, 174:24-43. doi: 10.1016/j.rse.2015.12.005
[15]	程鹏, 孔祥伟, 罗汉, 等. 近60 a以来祁连山中部气候变化及其径流响应研究[J]. 干旱区地理, 2020, 43(5):1192-1201.
	[ Cheng Peng, Kong Xiangwei, Luo Han, et al. Climate change and its runoff response in the middle section of the Qilian Mountains in the past 60 years[J]. Arid Land Geography, 2020, 43(5):1192-1201. ]
[16]	马剑, 刘贤德, 李广, 等. 祁连山北麓中段青海云杉林土壤水热时空变化特征[J]. 干旱区地理, 2020, 43(4):1033-1040.
	[ Ma Jian, Liu Xiande, Li Guang, et al. Spatial and temporal variations of soil moisture and temperature of Picea crassifolia forest in north piedmont of central Qilian Mountains[J]. Arid Land Geography, 2020, 43(4):1033-1040. ]
[17]	Fang S, He Z, Du J, et al. Carbon mass change and its drivers in a boreal coniferous forest in the Qilian Mountains, China from 1964 to 2013[J]. Forests, 2018, 9(2):f9020057, doi: 10.3390/f9020057.
[18]	Yue J W, Guan J H, Deng L, et al. Allocation pattern and accumulation potential of carbon stock in natural spruce forests in northwest China[J]. Peerj, 2018(6):e4859, doi: 10.7717/peerj.4859.
[19]	Zwally H J, Schutz R, Hancock D, et al. GLAS/ICESat L2 global land surface altimetry data (HDF5), version 34[DB/OL]. [2014]. NASA National Snow and Ice Data Center Distributed Active Archive Center.
[20]	Chi H, Sun G, Huang J, et al. National forest aboveground biomass mapping from ICESat/GLAS data and MODIS imagery in China[J]. Remote Sensing, 2015, 7(5):5534-5564. doi: 10.3390/rs70505534
[21]	Wei L, Feng Q, Deo R C. Changes in climatic elements in the Pan-Hexi region during 1960—2014 and responses to global climatic changes[J]. Theoretical and Applied Climatology, 2018, 133(1-2):405-420. doi: 10.1007/s00704-017-2194-6
[22]	王晓学, 沈会涛, 林田苗, 等. 利用多信息源提高半干旱地区TM影像的森林类型制图精度: 以北京西部山区为例[J]. 自然资源学报, 2017, 32(7):1217-1228.
	[ Wang Xiaoxue, Shen Huitao, Lin Tianmiao, et al. Improving the forest type mapping accuracy in semiarid mountainous areas based on TM images: Take the west mountain of Beijing as an example[J]. Journal of Natural Resources, 2017, 32(7):1217-1228. ]
[23]	Park T, Kennedy R E, Choi S, et al. Application of physically-based slope correction for maximum forest canopy height estimation using waveform Lidar across different footprint sizes and locations: Tests on LVIS and GLAS[J]. Remote Sensing, 2014, 6(7):6566-6586. doi: 10.3390/rs6076566
[24]	Fick S E, Hijmans R J. WorldClim 2: New 1-km spatial resolution climate surfaces for global land areas[J]. International Journal of Climatology, 2017, 37(12):4302-4315. doi: 10.1002/joc.2017.37.issue-12
[25]	Lee S, Ni-Meister W, Yang W, et al. Physically based vertical vegetation structure retrieval from ICESat data: Validation using LVIS in White Mountain National Forest, New Hampshire, USA[J]. Remote Sensing of Environment, 2011, 115(11):2776-2785. doi: 10.1016/j.rse.2010.08.026
[26]	王金叶, 车克钧, 蒋志荣. 祁连山青海云杉林碳平衡研究[J]. 西北林学院学报, 2000(1):9-14.
	[ Wang Jinye, Che Kejun, Jiang Zhirong. A study on carbon balance of Picea crassifolia in Qilian Mountains[J]. Journal of Northwest Forestry College, 2000(1):9-14. ]
[27]	王金叶, 车克钧, 傅辉恩, 等. 祁连山水源涵养林生物量的研究[J]. 福建林学院学报, 1998(4):34-38.
	[ Wang Jinye, Che Kejun, Fu Hui’en, et al. Study on biomass of water conservation forest on north slope of Qilian Mountains[J]. Journal of FuJian College of Forestry, 1998(4):34-38. ]
[28]	罗艳, 唐才富, 辛文荣, 等. 青海省云杉属(Picea)和圆柏属(Sabina)乔木含碳率分析[J]. 生态环境学报, 2014, 23(11):1764-1768.
	[ Luo Yan, Tang Caifu, Xin Wenrong, et al. Carbon content rate of Picea and Sabina trees in Qinghai Province[J]. Ecology and Environmental Sciences, 2014, 23(11):1764-1768. ]
[29]	关晋宏, 杜盛, 程积民, 等. 甘肃省森林碳储量现状与固碳速率[J]. 植物生态学报, 2016, 40(4):304-317. doi: 10.17521/cjpe.2016.0017
	[ Guan Jinhong, Du Sheng, Cheng Jimin, et al. Current stocks and rate of sequestration of forest carbon in Gansu Province, China[J]. Chinese Journal of Plant Ecology, 2016, 40(4):304-317. ] doi: 10.17521/cjpe.2016.0017
[30]	Zhao M, Yue T, Zhao N, et al. Combining LPJ-GUESS and HASM to simulate the spatial distribution of forest vegetation carbon stock in China[J]. Journal of Geographical Sciences, 2014, 24(2):249-268. doi: 10.1007/s11442-014-1086-2
[31]	程堂仁, 马钦彦, 冯仲科, 等. 甘肃小陇山森林生物量研究[J]. 北京林业大学学报, 2007(1):31-36.
	[ Chen Tangren, Ma Qinyan, Feng Zhongke, et al. Research on forest biomass in Xiaolong Mountains, Gansu Province[J]. Journal of Beijing Forestry University, 2007(1):31-36. ]
[32]	程堂仁, 冯菁, 马钦彦, 等. 甘肃小陇山森林植被碳库及其分配特征[J]. 生态学报, 2008, 28(1):33-44.
	[ Cheng Tangren, Feng Jing, Ma Qinyan, et al. Carbon pool and allocation of forest vegetations in Xiaolong Mountains, Gansu Province[J]. Acta Ecologica Sinica, 2008, 28(1):33-44. ]
[33]	Sileshi G W. A critical review of forest biomass estimation models, common mistakes and corrective measures[J]. Forest Ecology and Management, 2014, 329:237-254. doi: 10.1016/j.foreco.2014.06.026
[34]	程堂仁. 甘肃小陇山森林生物量及碳储量研究[D]. 北京: 北京林业大学, 2007.
	[ Cheng Tangren. Research on the forest biomass and carbon storage in Xiaolong Mountains, Gansu Province[D]. Beijing: Beijing Forestry University, 2007. ]
[35]	Baccini A, Laporte N, Goetz S J, et al. A first map of tropical Africa’s above-ground biomass derived from satellite imagery[J]. Environmental Research Letters, 2008, 3(4):045011, doi: 10.1088/1748-9326/3/4/045011. doi: 10.1088/1748-9326/3/4/045011
[36]	Cao L, Pan J, Li R, et al. Integrating airborne LiDAR and optical data to estimate forest aboveground biomass in arid and semi-arid regions of China[J]. Remote Sensing, 2018, 10(4):532, doi: 10.3390/rs10040532. doi: 10.3390/rs10040532
[37]	Ei Hajj M, Baghdadi N, Fayad I, et al. Interest of integrating spaceborne LiDAR data to improve the estimation of biomass in high biomass forested areas[J]. Remote Sensing, 2017, 9(3):213, doi: 10.3390/rs9030213. doi: 10.3390/rs9030213
[38]	Phillips S J, Anderson R P, Schapire R E. Maximum entropy modeling of species geographic distributions[J]. Ecological Modelling, 2006, 190(3-4):231-259. doi: 10.1016/j.ecolmodel.2005.03.026
[39]	Peng J, Liu Z, Liu Y, et al. Trend analysis of vegetation dynamics in Qinghai-Tibet Plateau using Hurst exponent[J]. Ecological Indicators, 2012, 14(1):28-39. doi: 10.1016/j.ecolind.2011.08.011
[40]	Wagner B, Liang E, Li X, et al. Carbon pools of semi-arid Picea crassifolia forests in the Qilian Mountains (north-eastern Tibetan Plateau)[J]. Forest Ecology and Management, 2015, 343:136-143. doi: 10.1016/j.foreco.2015.02.001
[41]	巫明焱, 董光, 王艺积, 等. 川西米亚罗自然保护区森林地上碳储量遥感估算[J]. 生态学报, 2020, 40(2):621-628.
	[ Wu Mingyan, Dong Guang, Wang Yiji, et al. Estimation of forest aboveground carbon storage in Sichuan Miyaluo Nature Reserve based on remote sensing[J]. Acta Ecologica Sinica, 2020, 40(2):621-628. ]
[42]	穆喜云, 刘清旺, 庞勇, 等. 基于机载激光雷达的森林地上碳储量估测[J]. 东北林业大学学报, 2016, 44(11):52-56.
	[ Mu Xiyun, Liu Qingwang, Pang Yong, et al. Forest aboveground carbon storage using RF algorithmic model and airborne LiDAR data[J]. Journal of Northeast Forestry University, 2016, 44(11):52-56. ]
[43]	Wu J, Wang X, Zhang H, et al. Development of a forest canopy height estimation model using GLAS full waveform data over sloping terrain[J]. International Journal of Remote Sensing, 2018, 39(23):9073-9091. doi: 10.1080/01431161.2018.1506181
[44]	胡凯龙, 刘清旺, 庞勇, 等. 基于机载激光雷达校正的ICESat/GLAS数据森林冠层高度估测[J]. 农业工程学报, 2017, 33(16):88-95.
	[ Hu Kailong, Liu Qingwang, Pang Yong, et al. Forest canopy height estimation based on ICESat/GLAS data by airborne Lidar correction[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(16):88-95. ]
[45]	Liu K, Wang J, Zeng W, et al. Comparison and evaluation of three methods for estimating forest above ground biomass using TM and GLAS data[J]. Remote Sensing, 2017, 9(4):341, doi: 10.3390/rs9040341. doi: 10.3390/rs9040341

土地覆被类别	样本数量/个	栅格数量/个
农田	100	12937
灌木	102	9180
草地	132	11223
森林	115	8511
针叶林	157	20567
阔叶林	109	16459
针阔混交林	48	5769
建设用地	76	8360
裸地	103	8652
水域	157	15563

坡度等级	R²		RMSE/m
坡度等级	校正前	校正后	校正前	校正后
0°~5°	0.62	0.66	2.83	1.84
5°~15°	0.64	0.65	3.32	2.23
15°~25°	0.66	0.67	6.74	3.51
25°~35°	0.63	0.65	8.00	3.91
35°~58°	0.65	0.67	12.59	3.48