北京地区春季木本植物花粉起始期预报模型研究

doi:10.11924/j.issn.1000-6850.casb2021-1028

摘要/Abstract

摘要：

为解决城市花粉期预报的问题,以北京城区春季榆科、柏科、杨柳科、松科4类主要的木本植物为研究对象,利用2012—2019年北京市气象局的花粉浓度观测数据,分别基于Spring Warming、Alternating、Sequential和Parallel 4种积温物候模型进行植物花粉期建模。模型结果表明,4类木本植物的花粉期起始日期均明显受到春化作用影响,即冬季的低温刺激会促使春季植物花粉期的提前。对比不同模型的结果认为,榆科、柏科、松科植物的最适模型为Alternating模型,RMSE介于1~3天,判定系数R²介于91%~95%;杨柳科植物的最适模型为Sequential模型,RMSE小于2天,判定系数R²为92.8%。说明基于积温原理的物候模型,能够较为准确地模拟北京地区春季木本植物的花粉起始期,对于植物物候期的模拟研究具有参考意义。

关键词: 气传花粉, 花粉起始期, 木本植物, 北京, 物候模型

Abstract:

The paper aims to solve the problem of urban pollen period forecast. In this study, four major woody plants (i.e. Ulmaceae, Cupressaceae, Salicaceae and Pinaceae) in Beijing urban areas were selected as the research objects. Based on four accumulated temperature phenological models (i.e. Spring Warming, Alternating, Sequential and Parallel), using the pollen concentration data of 2012-2019 from Beijing Meteorological Bureau, the pollen period of each plant was modeled. The results show that the pollen start dates of these four woody plants are significantly affected by vernalization, that is, the stimulation of cold temperature in winter would lead to an earlier pollen season in spring. Comparing the results of these four different models, it is found that the optimum model for Ulmaceae, Cupressaceae and Pinaceae is the Alternating model, the RMSE of this model is between 1-3 days, and the determination coefficient R² is between 91%-95%. However, the best model for Salicaceae is Sequential model, RMSE is less than 2 days, and R² is 92.8%. This shows that the phenological model based on accumulated temperature can effectively simulate the pollen initiation period of woody plants in Beijing, which has certain significance for the simulation study of plant phenological stage.

Key words: airborne pollen, pollen initiation period, woody plants, Beijing, phenological model

中图分类号:

S165+.3

王春玲, 叶彩华, 姜江. 北京地区春季木本植物花粉起始期预报模型研究[J]. 中国农学通报, 2022, 38(28): 89-97.

WANG Chunling, YE Caihua, JIANG Jiang. Study on the Prediction Model of the Pollen Initiation Period of Woody Plants in Beijing[J]. Chinese Agricultural Science Bulletin, 2022, 38(28): 89-97.

图/表 8

参考文献 26

[1]	李俊, 王洪源, 张志刚. 等. 空气花粉浓度与人群呼吸系统疾病的关联研究[J]. 环境与健康杂志, 2018(6):43-46,98.
[2]	TRAIDL-HOFFMANN C, KASCHE A, MENZEL A, et al. Impact of pollen on human health: More than allergen carriers?[J]. International archives of allergy & immunology, 2003,131(1):1-13.
[3]	BRETON M C, GARNEAU M, FORTIER I, et al. Relationship between climate, pollen concentrations of Ambrosia and medical consultations for allergic rhinitis in Montreal, 1994-2002[J]. Science of the total environment, 2006,370(1):39-50. doi: 10.1016/j.scitotenv.2006.05.022 URL
[4]	欧阳志云, 嘉楠, 郑华. 等. 北京城区花粉致敏植物种类、分布及物候特征[J]. 应用生态学报, 2017,18(9):1953-1958.
[5]	孟龄, 王效科, 欧阳志云. 等. 北京城区气传花粉季节分布特征[J]. 生态学报, 2013,33(8):2381-2387.
[6]	徐景先, 李耀宁, 张德山. 等. 空气花粉变化规律和预测预报研究进展[J]. 生态学报, 2009,29(7):3854-3863.
[7]	CHUIN I, CAMBO G, COMTOIS P. Scaling phenology from the local to the regional level: advances from species-specific phenological models[J]. Global change biology, 2008,6(8):943-952. doi: 10.1046/j.1365-2486.2000.00368.x URL
[8]	CHUNG, MACK, YUN, et al. Predicting the timing of cherry blossoms in Washington, DC and mid-atlantic states in response to climate change[J]. Plos one, 2011,6(11):e27439. doi: 10.1371/journal.pone.0027439 URL
[9]	仲舒颖, 葛全胜, 戴君虎. 等. 中国典型观赏植物花期模型建立及过去花期变化模拟[J]. 资源科学, 2017(11):102-115.
[10]	陆佩玲. 中国木本植物物候对气候变化的响应研究[D]. 北京: 北京林业大学 ;2006.
[11]	GOLDBERG C, BUCH H, MOSEHOLM L, et al. Airborne pollen records in Denmark, 1977-1986[J]. Grana, 1988,27(3):209-217. doi: 10.1080/00173138809428928 URL
[12]	SMITH M, J€AGER S, BERGER U, et al. Geographic and temporal variations in pollen exposure across Europe[J]. Allergy, 2014,69(7):913-923. doi: 10.1111/all.12419 pmid: 24816084
[13]	CANNELL M G R, SMITH R I. Thermal time, chill days and prediction of budburst in Picea sitchensis[J]. Journal of applied ecology, 1983,20(3):951-963. doi: 10.2307/2403139 URL
[14]	HUNTER A F, LECHOWICZ M J. Predicting the timing of budburst in temperate trees[J]. Journal of applied ecology, 1992,29(3):597-604. doi: 10.2307/2404467 URL
[15]	种康, 雍伟东, 谭克辉. 高等植物春化作用研究进展[J]. 植物学通报, 1999,16(5):481-487.
[16]	赵大中, 陈民, 种康. 等. 高等植物的春化作用与低温适应[J]. 植物生理学通讯, 1998(2):155-159.
[17]	胡巍, 侯喜林, 史公军. 植物春化特性及春化作用机理[J]. 植物学通报, 2004(1):26-36.
[18]	MURRAY M B, CANNELL M G R, SMITH R I. Date of budburst of fifteen tree species in Britain following climate warming[J]. Journal of applied ecology, 1989,26(2):693-700. doi: 10.2307/2404093 URL
[19]	HANNINEN H. Effects of temperature on dormancy release in woody plants: implications of prevailing models[J]. Silva fennica, 1987,21:279-299.
[20]	LANDSBERG J J. Apple fruit bud development and growth: Analysis and an empirical model[J]. Annals of botany, 1974,38(5):1013-1023. doi: 10.1093/oxfordjournals.aob.a084891 URL
[21]	KRAMER K. Selecting a model to predict the onset of growth of Fagus sylvatica[J]. Journal of applied ecology, 1994,31(1):172-181. doi: 10.2307/2404609 URL
[22]	CHUINE I, COUR P, ROUSSEAU D. Fitting models predicting dates of flowering of temperate-zone trees using simulated annealing[J]. Plant cell and environment, 1998,21(5):455-466. doi: 10.1046/j.1365-3040.1998.00299.x URL
[23]	魏延, 谢开贵. 模拟退火算法[J]. 蒙自师范高等专科学校学报, 1999,1(4):5.
[24]	庞峰. 模拟退火算法的原理及算法在优化问题上的应用[D]. 长春: 吉林大学, 2006.
[25]	STEINBRUNN M, MOERKOTTE G, KEMPER A. Heuristic and randomized optimization for the join ordering problem[J]. Vldb journal, 1997:191-208.
[26]	CHUINE I, COUR P, ROUSSEAU D. Selecting models to predict the timing of flowering of temperate trees: implications for tree phenology modelling[J]. Plant cell and environment, 1999,22(1):1-13. doi: 10.1046/j.1365-3040.1999.00395.x URL

花粉观测站点	有效样本年数/年				气象站站号
花粉观测站点	榆科	柏科	杨柳科	松科	气象站站号
昌平	7	7	7	8	54499
海淀	6	7	7	8	54399
朝阳	6	6	7	8	54433
石景山	6	7	7	8	54513
丰台	7	6	7	8	54514
合计	32	33	35	40

花粉观测站点	有效样本年数/年				气象站站号
花粉观测站点	榆科	柏科	杨柳科	松科	气象站站号
昌平	7	7	7	8	54499
海淀	6	7	7	8	54399
朝阳	6	6	7	8	54433
石景山	6	7	7	8	54513
丰台	7	6	7	8	54514
合计	32	33	35	40

模型	参数	调参单位	参数区间	最优参数值
模型	参数	调参单位	参数区间	榆科	柏科	杨柳科	松科
Spring Warming	T_b/℃	0.1	-5.0~5.0	0.7	-3.9	-0.2	-2.2
Spring Warming	GDD_forc/(℃·d)	1	20~600	79	94	136	565
Alternating	T_b/℃	0.1	-5.0~5.0	-1.3	-0.9	-0.3	3.8
	GDD_chill/(℃·d)	1	0~60	17	4	37	5
	a	1	10~500	128	210	137	370
	b	-0.0001	-0.0001~-0.01	-0.0015	-0.0002	-0.0002	-0.0001
Sequential	T_b/℃	0.1	-5.0~5.0	0.5	-0.5	0.3	4.2
	GDD_forc/(℃·d)	1	20~600	42	103	65	330
	GDD_chill/(℃·d)	1	0~60	44	9	42	12
	T_o/℃	0.1	-3.4~10.4	-3.2	-1.6	-3.0	0.4
Parallel	GDD_forc/(℃·d)	1	20~600	59	84	81	373
	GDD_chill/(℃·d)	1	0~60	53	50	49	41
	T_o/℃	0.1	-3.4~10.4	-3.3	0.3	-3.2	-2.1
	K_m	-0.0001	-0.0001~-0.01	0.0009	0.0016	0.0001	0.0004

模型	参数	调参单位	参数区间	最优参数值
模型	参数	调参单位	参数区间	榆科	柏科	杨柳科	松科
Spring Warming	T_b/℃	0.1	-5.0~5.0	0.7	-3.9	-0.2	-2.2
Spring Warming	GDD_forc/(℃·d)	1	20~600	79	94	136	565
Alternating	T_b/℃	0.1	-5.0~5.0	-1.3	-0.9	-0.3	3.8
	GDD_chill/(℃·d)	1	0~60	17	4	37	5
	a	1	10~500	128	210	137	370
	b	-0.0001	-0.0001~-0.01	-0.0015	-0.0002	-0.0002	-0.0001
Sequential	T_b/℃	0.1	-5.0~5.0	0.5	-0.5	0.3	4.2
	GDD_forc/(℃·d)	1	20~600	42	103	65	330
	GDD_chill/(℃·d)	1	0~60	44	9	42	12
	T_o/℃	0.1	-3.4~10.4	-3.2	-1.6	-3.0	0.4
Parallel	GDD_forc/(℃·d)	1	20~600	59	84	81	373
	GDD_chill/(℃·d)	1	0~60	53	50	49	41
	T_o/℃	0.1	-3.4~10.4	-3.3	0.3	-3.2	-2.1
	K_m	-0.0001	-0.0001~-0.01	0.0009	0.0016	0.0001	0.0004

模型	内部检验（偶数年）								外部检验（奇数年）
	榆科（12个样本）		柏科（13个样本）		杨柳科（15个样本）		松科（20个样本）		榆科（20个样本）		柏科（20个样本）		杨柳科（20个样本）		松科（20个样本）
	RMSE/d	R²/%	RMSE/d	R²/%	RMSE/d	R²/%	RMSE/d	R²/%	RMSE/d	R²/%	RMSE/d	R²/%	RMSE/d	R²/%	RMSE/d	R²/%
Spring Warming	1.7	92.8	2.8	69.7	1.9	85.0	2.0	82.9	3.2	56.7	3.4	72.1	2.7	71.4	2	90.8
Alternating	1.8	90.5	2.3	84.2	1.9	86.1	1.3	90.6	1.6	83.1	2.4	89.4	2.1	72.0	1.3	96.1
Sequential	1.9	89.6	2.3	76.0	1.9	85.3	1.5	87.8	3.5	30.2	2.6	89.4	1.7	83.7	1.4	95.6
Parallel	1.7	92.8	2.2	76.5	1.8	86.8	1.5	89.2	4.7	49.6	2.9	86.3	3.1	76.6	1.4	95.9