河北省北部华北落叶松人工林立地指数空间分布预测

doi:10.11707/j.1001-7488.20210308

摘要/Abstract

摘要：

目的: 探究河北省北部华北落叶松人工林立地生产力空间分布格局及其与环境因子的关系，为高效森林经营提供依据。方法: 基于研究区1 179块森林资源二类调查小班数据，建立以地形、气候和土壤因子为辅助变量的多元线性回归（MLR）、随机森林（RF）、回归克里格（RK）和地理加权回归克里格（GWRK）模型。通过模型评价，选择最优模型预测研究区华北落叶松人工林立地指数（SI）空间分布，采用相关和偏相关分析方法分析SI分布格局与主要环境因子的关系。结果: 海拔、最干月降水、土壤全氮和全磷是影响研究区华北落叶松人工林立地生产力的主要因子；地统计学的2种模型（RK和GWRK）拟合优度相近且均显著高于MLR和RF模型，与全局回归的RK模型相比，局部回归的GWRK模型具有较高的R²（0.780）及较低的AIC（160.533）、RMSE（1.593 m）和MAE（1.113 m），为最优预测模型；不同立地生产力等级地区内SI对地形和土壤因子的变化表现更为敏感，对气候因子的变化反应较弱。结论: 海拔、最干月降水较高以及土壤氮磷含量适中的研究区西北部，是华北落叶松人工林适生区，而海拔、最干月降水较低以及土壤氮磷含量不平衡的东、南部，华北落叶松人工林的适应性较差，可通过造林活动或适当添加氮磷肥削弱环境因子对树木的不利影响。

关键词: 立地指数, 环境因子, 地统计学, 随机森林, 华北落叶松

Abstract:

Objective: Exploring the spatial distribution of site productivity and analyzing its relationships with environmental factors for Larix principis-rupprechtii plantations are crucial for the effective forest management in the north area of Hebei Province. Method: Based on data from 1 179 forest inventory plots distributed across this area, we established multiple linear regression (MLR), random forest(RF), regression Kriging(RK) and geographical weighted regression Kriging(GWRK) models with topographic, climate and soil factors. Through model evaluation, the optimal model was selected to predict the spatial distribution patterns of site index(SI) for L. principis-rupprechtii plantations. The relationships between the distribution pattern of SI and environmental factors were analyzed by correlation and partial correlation analysis. Result: The major environmental factors influencing site productivity of L. principis-rupprechtii plantations in the study area included elevation(DEM), precipitation of the driest month(BIO14), total soil nitrogen and phosphorus(TN and TP). The values of goodness of fit of the two geostatistical models(RK and GWRK) were similar and both significantly higher than those of MLR and RF models. In contrast to RK model based on global regression, the GWRK model based on local regression had a higher fitting accuracy with R²=0.780, AIC=160.533, RMSE=1.593 m and MAE=1.113 m. The GWRK model based on local regression had a highly predictive power to predict SI of L. principis-rupprechtii plantations in the study area. SI was more sensitive to the variation of topographic and soil factors and less sensitive to the variation of the climate factors across regions with different site productivity classes. Conclusion: The site with a high growth suitability for L. principis-rupprechtii might tend to occur in the northwest area with a higher elevation, relatively high BIO14 and suitable soil TN and TP contents. However, the site with a low growth suitability for L. principis-rupprechtii were found in the eastern and southern areas with a lower elevation, relatively low BIO14, and unbalanced TN and TP contents. It might be feasible to mitigate the negative effects of environmental factors on growth and productivity of L. principis-rupprechtii plantations through silvicultural activities or appropriate nitrogen and phosphorus addition in the study area.

Key words: site index, environmental factors, geostatistics, random forest, Larix principis-rupprechtii

中图分类号:

S757

李文博,吕振刚,黄选瑞,张志东. 河北省北部华北落叶松人工林立地指数空间分布预测[J]. 林业科学, 2021, 57(3): 79-89.

Wenbo Li,Zhengang Lü,Xuanrui Huang,Zhidong Zhang. Predicting Spatial Distribution of Site Index for Larix principis-rupprechtii Plantations in the Northern Hebei Province[J]. Scientia Silvae Sinicae, 2021, 57(3): 79-89.

图/表 12

图1

图2

表1

图3

表2

图4

表3

表4

图5

表5

表6

图6

参考文献 0

	樊江文, 张良侠, 张文彦, 等. 中国草地样带植物根系N、P元素特征及其与地理气候因子的关系. 草业学报, 2014, 23 (5): 69- 76.
	Fan J W , Zhang L X , Zhang W Y , et al. Plant root N and P levels and their relationship to geographical and climate factors in a Chinese grassland transect. Acta Prataculturae Sinica, 2014, 23 (5): 69- 76.
	高敏, 马香丽, 杨晋宇, 等. 冀北山地华北落叶松人工林与白桦混交改造模式对土壤动物群落的影响. 林业科学, 2017, 53 (1): 70- 81.
	Gao M , Ma X L , Yang J Y , et al. Influence of the mixed modes of larch and birth on soil faunal community in mountain area of northern Hebei, China. Scientia Silvae Sinicae, 2017, 53 (1): 70- 81.
	雷相东, 符利勇, 李海奎, 等. 基于林分潜在生长量的立地质量评价方法与应用. 林业科学, 2018, 54 (12): 116- 126. doi: 10.11707/j.1001-7488.20181213
	Lei X D , Fu L Y , Li H K , et al. Methodology and applications of site quality assessment based on potential mean annual increment. Scientia Silvae Sinicae, 2018, 54 (12): 116- 126. doi: 10.11707/j.1001-7488.20181213
	卢同平, 张文翔, 牛洁, 等. 典型自然带土壤氮磷化学计量空间分异特征及其驱动因素研究. 土壤学报, 2017, 54 (3): 682- 692.
	Lu T P , Zhang W X , Niu J , et al. Study on spatial variability and driving factors of stoichiometry of nitrogen and phosphorus in soils of typical natural zones of China. Acta Pedologica Sinica, 2017, 54 (3): 682- 692.
	欧强新, 李海奎, 雷相东, 等. 基于清查数据的福建省马尾松生物量转换和扩展因子估算差异解析——3种集成学习决策树模型的比较. 应用生态学报, 2018, 29 (6): 2007- 2016.
	Ou Q X , Li H K , Lei X D , et al. Difference analysis in estimating biomass conversion and expansion factors of masson pine in Fujian Province, China based on national forest inventory data: a comparison of three decision tree models of ensemble learning. Chinese Journal of Applied Ecology, 2018, 29 (6): 2007- 2016.
	潘瑞炽. 植物生理学. 北京: 高等教育出版社, 2008.
	Pan R C . Plant physiology. Beijing: Higher Education Press, 2008.
	任丽娜, 王海燕, 丁国栋, 等. 密度调控对华北落叶松人工林土壤有机碳及养分特征的影响. 干旱区资源与环境, 2012, 26 (4): 138- 143.
	Ren L N , Wang H Y , Ding G D , et al. Effects of Larix principis-rupprechtii plantation density control on soil organic carbon and nutrients characteristics. Journal of Arid Land Resources and Environment, 2012, 26 (4): 138- 143.
	王冬至, 张冬燕, 蒋凤玲, 等. 塞罕坝华北落叶松人工林地位指数模型. 应用生态学报, 2015, 26 (11): 3413- 3420.
	Wang D Z , Zhang D Y , Jiang F L , et al. A site index model for Larix principis-rupprechtii plantation in Saihanba, north China. Chinese Journal of Applied Ecology, 2015, 26 (11): 3413- 3420.
	王云霓, 熊伟, 王彦辉, 等. 宁夏六盘山三种针叶林初级净生产力年际变化及其气象因子响应. 生态学报, 2013, 33 (13): 4002- 4010.
	Wang Y N , Xiong W , Wang Y H , et al. The interannual variation of net primary productivity of three coniferous forests in Liupan Mountains of Ningxia and its responses to climatic factors. Acta Ecologica Sinica, 2013, 33 (13): 4002- 4010.
	吴恒, 党坤良, 田相林, 等. 秦岭林区天然次生林与人工林立地质量评价. 林业科学, 2015, 51 (4): 78- 88.
	Wu H , Dang K L , Tian X L , et al. Evaluating site quality for secondary forests and plantation in Qinling Mountains. Scientia Silvae Sinicae, 2015, 51 (4): 78- 88.
	杨顺华, 张海涛, 郭龙, 等. 基于回归和地理加权回归Kriging的土壤有机质空间插值. 应用生态学报, 2015, 26 (6): 1649- 1656.
	Yang S H , Zhang H T , Guo L , et al. Spatial interpolation of soil organic matter using regression Kriging and geographically weighted regression Kriging. Chinese Journal of Applied Ecology, 2015, 26 (6): 1649- 1656.
	赵匡记, 王利东, 王立军, 等. 华北落叶松蓄积量及生产力研究. 北京林业大学学报, 2015, 37 (2): 24- 31.
	Zhao K J , Wang L D , Wang L J , et al. Stock volume and productivity of Larix principis-rupprechtii in northern and northwestern China. Journal of Beijing Forestry University, 2015, 37 (2): 24- 31.
	Benitez F L , Anderson L O , Formaggio A R . Evaluation of geostatistical techniques to estimate the spatial distribution of aboveground biomass in the Amazon rainforest using high-resolution remote sensing data. Acta Amazonica, 2016, 46 (2): 151- 160. doi: 10.1590/1809-4392201501254
	Blyth J F , Macleod D A . Sitka spruce (Picea sitchensis) in north-east Scotland I. Relationships between site factors and growth. Forestry, 1981, 54 (1): 41- 62.
	Bravo-Oviedo A , Roig S , Bravo F , et al. Environmental variability and its relationship to site index in Mediterranean maritine pine. Forest Systems, 2011, 20 (1): 50- 64. doi: 10.5424/fs/2011201-9106
	Bray E A . Plant responses to water deficit. Trends in Plant Science, 1997, 2 (2): 48- 54. doi: 10.1016/S1360-1385(97)82562-9
	Breiman L . Random forests. Machine Learning, 2001, 45 (1): 5- 32. doi: 10.1023/A:1010933404324
	Bueis T , Bravo F , Pando V , et al. Site factors as predictors for Pinus halepensis Mill. productivity in Spanish plantations. Annals of Forest Science, 2017, 74 (1): 6. doi: 10.1007/s13595-016-0609-7
	Deng M F , Liu L L , Sun Z Z , et al. Increased phosphate uptake but not resorption alleviates phosphorus deficiency induced by nitrogen deposition in temperate Larix principis-rupprechtii plantations. New Phytologist, 2016, 212 (4): 1019- 1029. doi: 10.1111/nph.14083
	Falkowski M J , Wulder M A , White J C , et al. Supporting large-area, sample-based forest inventories with very high spatial resolution satellite imagery. Progress in Physical Geography, 2009, 33 (3): 403- 423. doi: 10.1177/0309133309342643
	Farrelly N , Dhubhain Á N , Nieuwenhuis M . Site index of Sitka spruce (Picea sitchensis) in relation to different measures of site quality in Ireland. Canadian Journal of Forest Research, 2011, 41 (2): 265- 278. doi: 10.1139/X10-203
	Fayad I , Baghdadi N , Bailly J S , et al. Regional scale rain-forest height mapping using regression-kriging of Spaceborne and Airborne LiDAR data: Application on French Guiana. Remote Sensing, 2016, 8 (3): 240. doi: 10.3390/rs8030240
	Fierer N , McCain C M , Meir P , et al. Microbes do not follow the elevational diversity patterns of plants and animals. Ecology, 2011, 92 (4): 797- 804. doi: 10.1890/10-1170.1
	Fotheringham A S , Charlton M E , Brunsdon C . Geographically weighted regression: a natural evolution of the expansion method for spatial data analysis. Environment and Planning A, 1998, 30 (11): 1905- 1927. doi: 10.1068/a301905
	Georganos S , Abdi A M , Tenenbaum D E , et al. Examining the NDVI-rainfall relationship in the semi-arid Sahel using geographically weighted regression. Journal of Arid Environments, 2017, 146, 64- 74. doi: 10.1016/j.jaridenv.2017.06.004
	Hengl T , Heuvelink G B M , Rossiter D G . About regression-kriging: from equations to case studies. Computers & Geosciences, 2007, 33 (10): 1301- 1315.
	Hijmans R J , Cameron S E , Parra J L , et al. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 2005, 25 (15): 1965- 1978. doi: 10.1002/joc.1276
	Hlasny T , Trombik J , Bosela M , et al. Climatic drivers of forest productivity in central Europe. Agricultural and Forest Meteorology, 2017, 234-235, 258- 273. doi: 10.1016/j.agrformet.2016.12.024
	Huang S L , Ramirez C , Conway S , et al. Mapping site index and volume increment from forest inventory, Landsat, and ecological variables in Tahoe national forest, California, USA. Canadian Journal of Forest Research, 2017, 47 (1): 113- 124. doi: 10.1139/cjfr-2016-0209
	Jiang H Q , Radtke P J , Weiskittel A R , et al. Climate- and soil-based models of site productivity in eastern US tree species. Canadian Journal of Forest Research, 2015, 45 (3): 325- 342. doi: 10.1139/cjfr-2014-0054
	Jin Q T , Zhang J T , Shi M C , et al. Estimating Loess Plateau average annual precipitation with multiple linear regression Kriging and geographically weighted regression kriging. Water, 2016, 8 (6): 266. doi: 10.3390/w8060266
	Littke K M , Harrison R B , Zabowski D . Determining the effects of biogeoclimatic properties on different site index systems of Douglas-fir in the Coastal Pacific Northwest. Forest Science, 2016, 62 (5): 503- 512. doi: 10.5849/forsci.15-191
	McKenney D W , Pedlar J H . Spatial models of site index based on climate and soil properties for two boreal tree species in Ontario, Canada. Forest Ecology and Management, 2003, 175 (1/3): 497- 507.
	Nigh G D , Ying C C , Qian H , et al. Climate and productivity of major conifer species in the interior of British Columbia, Canada. Forest Science, 2004, 50 (5): 659- 671.
	Noordermeer L , Bollandsas O M , Gobakken T , et al. Direct and indirect site index determination for Norway spruce and Scots pine using bitemporal airborne laser scanner data. Forest Ecology and Management, 2018, 428, 104- 114. doi: 10.1016/j.foreco.2018.06.041
	Parresol B R , Scott D A , Zarnoch S J , et al. Modeling forest site productivity using mapped geospatial attributes within a South Carolina Landscape, USA. Forest Ecology and Management, 2017, 406, 196- 207. doi: 10.1016/j.foreco.2017.10.006
	Peng Y H , Lin W L , Cai W M , et al. Over expression of a Panax ginseng tonoplast aquaporin alters salt tolerance, drought tolerance and cold acclimation ability in transgenic Arabidopsis plants. Planta, 2007, 226 (3): 729- 740. doi: 10.1007/s00425-007-0520-4
	Philip J R . Plant water relations: some physical aspects. Annual Review of Plant Physiology, 1966, 17 (1): 245- 268. doi: 10.1146/annurev.pp.17.060166.001333
	Sabatia C O , Burkhart H E . Predicting site index of plantation loblolly pine from biophysical variables. Forest Ecology and Management, 2014, 326, 142- 156. doi: 10.1016/j.foreco.2014.04.019
	Skovsgaard J P , Vanclay J K . Forest site productivity: a review of the evolution of dendrometric concepts for even-aged stands. Forestry, 2008, 81 (1): 13- 31. doi: 10.1093/forestry/cpm041
	Steyerberg E W , Eijkemans M J C , Habbema J D F . Stepwise selection in small data sets: a simulation study of bias in logistic regression analysis. Journal of Clinical Epidemiology, 1999, 52 (10): 935- 942. doi: 10.1016/S0895-4356(99)00103-1
	Veronesi F , Schillaci C . Comparison between geostatistical and machine learning models as predictors of topsoil organic carbon with a focus on local uncertainty estimation. Ecological Indicators, 2019, 101, 1032- 1044. doi: 10.1016/j.ecolind.2019.02.026
	Wang S J , Ruan H H , Wang B . Effects of soil microarthropods on plant litter decomposition across an elevation gradient in the Wuyi Mountains. Soil Biology & Biochemistry, 2009, 41 (5): 891- 897.
	Wei S G , Dai Y J , Liu B Y , et al. A China data set of soil properties for land surface modeling. Journal of Advances in Modeling Earth Systems, 2013, 5 (2): 212- 224. doi: 10.1002/jame.20026
	Zhang H , Wu P B , Yin A J , et al. Prediction of soil organic carbon in an intensively managed reclamation zone of eastern China: a comparison of multiple linear regressions and the random forest model. Science of the Total Environment, 2017, 592, 704- 713. doi: 10.1016/j.scitotenv.2017.02.146

数据来源 Data source	项目 Items	样本量 Number	最小值 Min.	最大值 Max.	平均值 Mean	标准差 Std.	变异系数 CV(%)	K-S检验 K-S test
解析木 Analytical trees	平均木高 Average height/m	92	9.80	21.60	15.66	3.06	19.53	0.71
解析木 Analytical trees	优势木高 Dominant height/m	92	10.42	23.50	16.59	3.17	18.73	0.91
森林调查 Forest inventory	立地指数 Site index/m	1 179	6.00	20.90	13.40	3.02	22.56	0.06

因子类型 Factor type	因子 Factors	最小值 Min.	最大值 Max.	平均值 Mean	标准差 Std.	变异系数 CV(%)
气候 Climate	BIO1/℃	1.18	6.94	3.87	1.70	43.93
	BIO2/℃	9.88	14.42	12.50	0.83	6.64
	BIO3	22.65	29.72	27.03	1.42	5.25
	BIO5/℃	19.90	28.90	25.97	1.72	6.62
	BIO6/℃	-28.00	-16.50	-20.26	2.20	10.86
	BIO8/℃	13.52	21.43	18.54	1.44	7.77
	BIO9/℃	-17.88	-7.07	-11.91	2.17	18.22
	BIO11/mm	-17.89	-9.13	-12.03	2.03	16.87
	BIO14/mm	1.00	5.00	1.76	0.77	43.75^*
	BIO15	105.18	120.15	113.77	2.60	2.29
	BIO17/mm	5.00	17.00	8.28	2.14	25.85

土壤 Soil	SOM/(g·kg^-1)	0.24	10.66	3.63	2.06	56.75
	TN/(g·kg^-1)	0.01	0.50	0.20	0.09	45.00^*
	TP/(g·kg^-1)	0.03	0.11	0.06	0.02	33.33^***

地形 Terrain	DEM/m	663.00	2 042.00	1 272.36	221.17	17.38^***

预测模型 Prediction models	最小值 Min.	最大值 Max.	平均值 Mean	标准差 Std.	K-S检验∶P K-S test∶P-value	空间相关关系检验 Spatial correlation test
预测模型 Prediction models	最小值 Min.	最大值 Max.	平均值 Mean	标准差 Std.	K-S检验∶P K-S test∶P-value	Moran’I	Z	P
MLR	-6.096	7.886	-0.002	2.356	0.428	0.219	7.291	< 0.01
RF	-7.203	7.539	0.012	2.201	0.229	0.116	4.084	0.031
RK	-6.068	5.942	-0.014	1.665	0.274	0.074	2.306	0.085
GWRK	-6.167	6.040	-0.046	1.589	0.055	0.062	1.699	0.208

预测模型 Prediction models	评价指标平均值 Mean value of validation indicators
预测模型 Prediction models	R²	RMSE	MAE	AIC
MLR	0.376	2.364	1.830	416.702
RF	0.544	2.205	1.657	332.770
RK	0.751	1.669	1.232	190.450
GWRK	0.780	1.593	1.113	160.533

因子 Factors	平均值 Mean	中位数 Median	标准差 Std.	空间相关关系检验 Spatial correlation test
因子 Factors	平均值 Mean	中位数 Median	标准差 Std.	Moran’I	Z	P
截距 Intercept	8.710	11.194	7.219	0.842	79.268	< 0.01
DEM	0.003	0.002	0.004	0.649	61.088	< 0.01
BIO14	-0.273	-0.211	0.542	0.384	36.284	< 0.01
TP	26.045	9.710	51.983	0.653	61.550	< 0.01
TN	-3.528	-1.318	7.958	0.278	26.448	< 0.01