This study aims to explore the spatial distribution characteristics of pine wilt disease in nine cities of eastern Liaoning Province (including Shenyang, Dalian, Anshan, Fushun, Benxi, Dandong, Yingkou, Liaoyang, and Tieling), construct a prediction model for pine wilt disease risk areas using the geographically weighted Logistic regression (GWLR) model, and provide a scientific basis for the prevention and control of pine wilt disease. Based on the compartments data with pine wilt disease information of Liaoning Province in 2023, kernel density analysis was used to characterize the spatial distribution of pine wilt disease, and thirty-two variables, including meteorological, topographic, vegetation index, and social factors, were considered. Key factors for the prediction model were screened using Spearman's correlation test, bidirectional stepwise regression, and variance inflation factor (VIF) analysis. Logistic regression and GWLR models were constructed to predict pine wilt disease, their prediction accuracies were compared and the spatial distribution of GWLR coefficients for different independent variables were analyzed. Pine wilt disease outbreaks in Liaoning Province were primarily concentrated around Fushun City. Five influential factors were key factors in the prediction model of pine wilt disease in nine cities of eastern Liaoning Province, including the distance from compartments to roads, distance from compartments to railways, annual average temperature, growing-season drought index, and soil moisture. The GWLR model significantly outperformed the traditional Logistic regression model, with an R² improvement of 0.42, a reduction of 0.14 in root mean square error (RMSE) and 0.10 in mean absolute error (MAE). The risk prediction showed high accuracy using GWLR: overall accuracy (OA) of 90.41%, producer's accuracy (PA) of 95.34% and user's accuracy (UA) of 86.79% for risk areas, and PA of 85.49% and UA of 94.83% for non-risk areas. The high-risk zones were concentrated in Fushun City, southeastern Tieling City, and northwestern Benxi City. Pine wilt disease in eastern Liaoning's nine cities was predominantly distributed around Fushun. The GWLR model effectively captured the spatial heterogeneity of the disease, with significant spatial variations in the coefficients of independent variables. The distances from compartments to roads and railways exhibited obvious spatial heterogeneity in affecting disease incidence, while annual average temperature promoted disease incidence in most areas. This study provides theoretical support for exploring the transmission patterns of pine wilt disease and formulating precise prevention strategies.
以松材线虫(Bursaphelenchus xylophilus)为病原的松材线虫病(pine wilt disease),又称松树萎蔫病,是松科(Pinaceae Spreng. ex F. Rudolphi)植物中松属(Pinus L.)和部分非松属针叶树种的毁灭性病害,被称为松树的“癌症”,严重危害了林业生态安全[1-3]。1969年,日本的森林学家Kiyohara和Tokushige从因感染松材线虫病死亡的松树中分离出一种线虫,并于1971年确认该线虫是落叶松萎蔫病的病原[4]。松材线虫病的传播机制主要依赖于媒介昆虫松墨天牛,受全球气候变暖的影响,松墨天牛的适生分布范围发生改变,致使松材线虫传播区域的边界产生了变化,具备了向更大范围传播的可能[5]。自1982年于江苏省南京市中山陵中山墓后面的黑松林中首次发现松材线虫病以来,松材线虫病的发生地区不断蔓延和扩大,其最西端已达四川省宜宾市,最北端已达吉林省通化市,共计18省(自治区、直辖市)622个县级行政区发生疫情,疫情发生面积达到171.65万hm2[6]。松材线虫病入侵我国40多年来已经累计致死松科植物达5 000万株,每年造成直接经济损失25亿元,间接经济损失更是无法估量[7]。为此探究松材线虫病发生风险在我国空间和时间上的分布特征,建立松材线虫病风险区时空分布预测模型,从而有效地预防松材线虫病的传播迫在眉睫。
在全球范围内,松材线虫病的分布和风险评估是研究的重点。通过结合气候、地形和土壤等数据,构建松材线虫病风险评估模型,系统评估当前气候条件和未来不同情景下潜在的松材线虫病分布风险,为松材线虫病监测和防控提供了重要的理论依据[8]。随着空间信息技术的发展,松材线虫病的监测与预测技术取得了显著进展,空间统计模型在松材线虫病预测中发挥着重要作用。Jiang等[9]探讨了逻辑回归(Logistic)在松材线虫病等森林病害预测中的应用,分析了传统统计模型与机器学习方法的优劣。Jung等[10]使用大规模时空数据集分析松材线虫病的传播距离,并发现其扩散受到昆虫媒介和人类活动的显著影响。在预测模型方面,陈伟华[5]基于松材线虫病的分布数据与气象数据,通过刀切法分析了中国松材线虫病近40 a时空分布变化及其驱动因素,并探究了其空间分布的质心变化。许格希等[11]利用地理信息系统(geographic information systems,GIS)技术和MaxEnt模型,预测川西林区松材线虫病入侵风险,为森林保护和质量改善提供关键参考。Yoon等[12]研究采用了一种半动态(semi-dynamic)方法,将物种分布建模(species distribution models,SDM)与空间预测相结合,以提高松材线虫病的监测和风险评估能力。应用空间统计模型,可以有效地识别和预测森林病虫害的发生,但是,对于病害发生的影响因子探究还存在不足,尤其是在存在空间异质性的大尺度研究中,而地理加权逻辑回归(geographic weighted logistic regression,GWLR)模型可以补充这方面的不足。王卫国[13]通过GWLR模型结合社会因子、遥感信息以及基础地形信息,对甘肃省火灾发生风险进行了有效地评价。Boussouf等[14]应用GWLR模型对西班牙东南部的滑坡风险进行了评价,获取了高危滑坡风险区的分布。梁慧玲等[15]研究证实GWLR模型在风险监测及对影响因子分析中有着较强的实用性。
应用的其他辅助数据包括地形数据、气象数据、社会因子以及植被指数。地形数据为航天飞机雷达地形测绘使命(shuttle radar topography mission,SRTM)获取的数字高程模型(digital elevation model,DEM)数据。本研究获取的气象数据包括2个数据源分别为全球陆地表面每月气候和气候水平衡的高分辨率数据集(monthly climate and climatic water balance for global terrestrial surfaces,TerraClimate)和第二代全球成像仪(second generation global imager,SGLI)数据集。社会因子主要获取了铁路、公路和河网数据,该数据来源于2020年的开放街道地图(open street map,OSM)数据,包括城市主干道、高速公路、国道、省道、县道、乡道、铁路、轻轨、窄轨、地铁和有轨电车等。植被指数包括叶面积指数(leaf area index,LAI)、归一化植被指数(normalized difference vegetation index,NDVI)和增强型植被指数(enhanced vegetation index,EVI),其中,LAI数据来源于SGLI的GCOM-C/SGLI L3 Leaf Area Index(V3)数据产品,时间分辨率为3~4 d,空间分辨率约4 km(即1/24°)。NDVI与EVI则是来源于中分辨率成像光谱仪(moderate resolution imaging spectroradiometer,MODIS)影像的MOD13A1.061产品,时间分辨率16 d,空间分辨率500 m,包含了2000年2月18日至2024年10月31日全球范围的NDVI与EVI,2个植被指数产品均通过谷歌地球引擎(Google Earth Engine,GEE)平台下载获得(https://earthengine.google.com/)。
本研究使用决定系数(coefficient of determination,R2)、均方根误差(root mean squared error,RMSE)、平均绝对误差(mean absolute error,MAE)、贝叶斯信息准则(bayesian information criterion,BIC)以及赤池信息准则(akaike information criterion,AIC),对模型的拟合效果进行评价。为评价风险区预测的准确性,本研究使用2023年松材线虫病疫情数据作为验证数据,与松材线虫病风险区预测结果构建混淆矩阵,应用用户者精度(user's accuracy,UA)、生产者精度(producer's accuracy,PA)以及总体精度(overall accuracy,OA),验证模型对松材线虫病风险区预测的准确性。
WANGF, LID L, YUS F.Database of pest risk assessment for Bursaphelenchus xylophilus [J].Journal of Laiyang Agricultural College,2004(2):162-163.
[3]
郝昕.松材线虫响应强力节丛孢胁迫的分子机制研究[D].哈尔滨:东北林业大学,2023.
[4]
HAOX.Molecular response mechanisms of Bursaphelenchus xylophilus exposed to Arthrobotrys robusta[D].Harbin:Northeast Forestry University,2023.
[5]
张瑞芝.表儿茶素延缓红松松树萎蔫病显症[D].哈尔滨:东北林业大学,2023.
[6]
ZHANGR Z.Epicatechin delays progression of pine wilt disease in Pinus koraiensis [D].Harbin:Northeast Forestry University,2023.
[7]
叶建仁,吴小芹.松材线虫病研究进展[J].中国森林病虫,2022,41(3):1-10.
[8]
YEJ R, WUX Q.Research progress of pine wilt disease[J].Forest Pest and Disease,2022,41(3):1-10.
[9]
陈伟华.不同气候情景下松材线虫病地理分布时空变化与预测研究[D].南京:南京林业大学,2023.
[10]
CHENW H.Spatial and temporal variation in the geographical distribution and prediction of pine wood nematode disease under different climate[D].Nanjing:Nanjing Forestry University,2023.
LIS, SUNH, ZHOUY T,et al.Occurrence of major forest pests in China in 2021 and prediction of occurrence trends in 2022[J].Forest Pest and Disease,2022,41(2):44-47.
[13]
张星耀,吕全,冯益明,等.中国松材线虫病危险性评估及对策[M].北京:科学出版社,2011.
[14]
ZHANGX Y,LYU Q, FENGY M,et al.Risk assessment and countermeasures of pine wood nematode disease in China[M].Beijing:Science Press,2011.
[15]
WANGG, AIERKENN, CHAIG Q,et al.A novel BH3DNet method for identifying pine wilt disease in Masson pine fusing UAS hyperspectral imagery and LiDAR data[J].International Journal of Applied Earth Observation and Geoinformation,2024,134:104177.
[16]
JIANGX S, RONGZ F, HUANGL F,et al.Research progress on monitoring and early warning technology of forestry pests and diseases[J].Journal of Agricultural Science & Technology,2025,27(1):1-16.
[17]
JUNGJ M, YOONS, HWANGJ,et al.Analysis of the spread distance of pine wilt disease based on a high volume of spatiotemporal data recording of infected trees[J].Forest Ecology and Management,2024,553:121612.
XUG X, YUR B, YANGC X,et al.Prediction of invasion risk of pine wilt disease based on GIS spatial technology and Max Ent model in western Sichuan Province of Southwestern China[J].Journal of Beijing Forestry University,2023,45(9):104-115.
[20]
YOONS, JUNGJ M, KIMD,et al.Risk monitoring of pine wilt disease based on semi-dynamic spatial prediction in South Korea[J].Agricultural Systems,2025,224:104253.
[21]
王卫国.甘肃省火灾风险评价与区划研究[D].兰州:西北师范大学,2017.
[22]
WANGW G.Study on the assessment of fire risk and zoning in Gansu Province[D].Lanzhou:Northwest Normal University,2017.
[23]
BOUSSOUFS, FERNÁNDEZT, HARTA B.Landslide susceptibility mapping using maximum entropy (MaxEnt) and geographically weighted logistic regression (GWLR) models in the Río Aguas catchment (Almería,SE Spain)[J].Natural Hazards,2023,117(1):207-235.
LIANGH L, WANGW H, GUOF T,et al.Comparing the application of logistic and geographically weighted logistic regression models for Fujian forest fire forecasting[J].Acta Ecologica Sinica,2017,37(12):4128-4141.
CAOY Q, ZHOUS H, YANGX T.Vegetation dynamics and its response to climate change in Liaoning Province in last 20 years[J].Acta Ecologica Sinica,2022,42(14):5966-5979.
[28]
李文翰.辽宁省中低产田障碍分析与改良优先序研究[D].沈阳:沈阳农业大学,2023.
[29]
LIW H.Analysis of obstacles and improvement priority of middle and low yield fields in Liaoning Province[D].Shenyang:Shenyang Agricultural University,2023.
JIAK, YANGZ, WEIC,et al.Pilot protection based on Spearman rank correlation coefficient for transmission line connected to renewable energy source[J].Automation of Electric Power Systems,2020,44(15):103-111.
YUL Y, ZHANJ H.A research on probability of default prediction based on logistic regression analysis[J].Journal of Finance and Economics,2004,30(9):15-23.
[34]
开璇.多尺度地理加权逻辑回归模型理论及其应用[D].呼和浩特:内蒙古工业大学,2024.
[35]
KAI X.Study on the multiscale geographically weighted logistic regression model and its application[D].Hohhot:Inner Mongolia University of Technology,2024.
CHENS Y.Prediction of grassland fire occurrence probability based on geographically weighted logistic regression model[D].Hohhot:Inner Mongolia University of Technology,2022.
ZHANGZ Y, LIUY F, ZHANGY,et al.Spatial non-stationary response of the ecosystem services synergy and tradeoff to influencing factors:A case study of ecological function area in Fujian Province[J].Geomatics and Information Science of Wuhan University,2022,47(1):111-125.
WANGJ N, JIANGS W, ZHANGR Z,et al.Prolonged incubation period of Bursaphelenchus xylophilus results in the over-year death of Pinus koraiensis [J].Scientia Silvae Sinicae,2024,60(10):67-75.
[42]
李洋.辽宁抚顺松材线虫病发生区姬蜂种类及其丰富度研究[D].沈阳:沈阳农业大学,2023.
[43]
LIY.Study on species and richness of ichneumonidae in occurrence area of pine wilt disease in Fushun City,Liaoning Province[D].Shenyang:Shenyang Agricultural University,2023.