当前位置: X-MOL 学术Forest Ecol. Manag. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Using climatic and biophysical attributes to model the presence and severity of root disease across the U.S. Northern Rocky Mountains
Forest Ecology and Management ( IF 3.7 ) Pub Date : 2020-10-01 , DOI: 10.1016/j.foreco.2020.118355
Zachary A. Holden , Alan Swanson , Blakey Lockman , Joel Egan , Christy M. Cleaver , Kathleen McKeever , Renate Bush , Erin L. Landguth

Abstract Root disease fungi are native, soil-borne pathogens that impact forest stand dynamics and productivity through direct tree mortality or reduced yield. Despite being a major contributor of mortality in forests of western North America, particularly in the northern Rocky Mountains in the United States, little is known about the underlying factors that influence root disease occurrence and severity at landscape and broader scales. We used a database with >15,000 geographically-distributed, spatially-referenced root disease assessments from United States Department of Agriculture (USDA) Forest Service Forest Inventory and Analysis (FIA) plots from across the US Northern Rocky mountains and high resolution (30 m) climate and soil-water balance grids to examine climatic and biophysical factors associated with presence and severity of root disease. We used a two-stage modeling approach that combined boosted regression-tree models and residual kriging to predict the probability of occurrence and severity of root disease across the area of analysis. A best-fit model explained root disease occurrence moderately well (AUC = 0.77) and included average annual dewpoint temperature, evapotranspiration and climatic water deficit. Root disease occurrence was associated with moist, humid sites subject to low or moderate climatic water deficits. Root disease severity was not well explained by climatic and biophysical factors alone (R2 = 0.19). Residual kriging only marginally improved model fit (R2 = 0.20). We used the fitted models to produce 30 m resolution gridded root disease occurrence and severity maps across the US Forest Service Northern Region for use in broad-scale planning and analysis. Weak spatial autocorrelation at distances beyond 5 km and relatively low accuracy of the severity model suggests that the FIA plot distribution is not suited to predict spatial patterns of root disease severity. Additional factors related to root disease severity, such as vegetation type and site-specific disturbance histories, are likely needed for fine-scale severity predictions.

中文翻译:

使用气候和生物物理属性来模拟美国北落基山脉根部疾病的存在和严重程度

摘要 根病真菌是一种土传病原体,通过直接的树木死亡或产量降低影响林分动态和生产力。尽管是北美西部森林,特别是美国落基山脉北部森林死亡率的主要贡献者,但人们对在景观和更广泛范围内影响根病发生和严重程度的潜在因素知之甚少。我们使用了一个在地理上分布超过 15,000 个的数据库,来自美国农业部 (USDA) 林务局森林清查和分析 (FIA) 地块的空间参考根病评估来自美国北部落基山脉和高分辨率 (30 m) 气候和水土平衡网格,以检查气候和与根病的存在和严重程度相关的生物物理因素。我们使用了一种两阶段建模方法,该方法结合了增强回归树模型和残差克里金法来预测整个分析区域中根病发生的概率和严重程度。最佳拟合模型可以很好地解释根病的发生 (AUC = 0.77),并包括年平均露点温度、蒸散量和气候水分亏缺。根病的发生与受低或中度气候缺水的潮湿、潮湿场所有关。仅用气候和生物物理因素并不能很好地解释根病严重程度(R2 = 0.19)。残差克里金法仅略微改善模型拟合(R2 = 0.20)。我们使用拟合模型生成了美国林务局北部地区的 30 m 分辨率网格根病发生率和严重程度图,用于大规模规划和分析。距离超过 5 公里时空间自相关较弱,严重性模型的准确性相对较低,这表明 FIA 图分布不适合预测根病严重程度的空间模式。精细尺度的严重程度预测可能需要与根病严重程度相关的其他因素,例如植被类型和特定地点的干扰历史。残差克里金法仅略微改善模型拟合(R2 = 0.20)。我们使用拟合模型生成了美国林务局北部地区的 30 m 分辨率网格根病发生率和严重程度图,用于大规模规划和分析。距离超过 5 公里时空间自相关较弱,严重性模型的准确性相对较低,这表明 FIA 图分布不适合预测根病严重程度的空间模式。精细尺度的严重程度预测可能需要与根病严重程度相关的其他因素,例如植被类型和特定地点的干扰历史。残差克里金法仅略微改善模型拟合(R2 = 0.20)。我们使用拟合模型生成了美国林务局北部地区的 30 m 分辨率网格根病发生率和严重程度图,用于大规模规划和分析。距离超过 5 公里时空间自相关较弱,严重性模型的准确性相对较低,这表明 FIA 图分布不适合预测根病严重程度的空间模式。精细尺度的严重程度预测可能需要与根病严重程度相关的其他因素,例如植被类型和特定地点的干扰历史。我们使用拟合模型生成了美国林务局北部地区的 30 m 分辨率网格根病发生率和严重程度图,用于大规模规划和分析。距离超过 5 公里时空间自相关较弱,严重性模型的准确性相对较低,这表明 FIA 图分布不适合预测根病严重程度的空间模式。精细尺度的严重程度预测可能需要与根病严重程度相关的其他因素,例如植被类型和特定地点的干扰历史。我们使用拟合模型生成了美国林务局北部地区的 30 m 分辨率网格根病发生率和严重程度图,用于大规模规划和分析。距离超过 5 公里时空间自相关较弱,严重性模型的准确性相对较低,这表明 FIA 图分布不适合预测根病严重程度的空间模式。精细尺度的严重程度预测可能需要与根病严重程度相关的其他因素,例如植被类型和特定地点的干扰历史。
更新日期:2020-10-01
down
wechat
bug