基于气象因子深度学习的森林火灾预测方法

森林火灾一旦发生将对生态系统造成严重的破坏,间接导致气候的变化和极端天气频发。对森林火灾的发生进行准确预测可提前采取有效的防控措施,具有重要意义。传统林火预测模型多为数学方法和浅层神经网络,当数据量增大时易出现建模困难以及预测精度降低等问题。深度学习模型在处理大量非线性数据上具有一定的优势,其模型具有多层网络结构,通过训练大量数据可提取出具有代表性的特征值,发现数据间的隐含关系,达到准确分类预测的目的。因此,本研究提出一种基于深度学习的林火预测方法,将深度信念网络(deep belief network,DBN)作为预测模型,气象因子作为输入数据,以解决传统林火预测模型在面对大量数据时预测效果不佳的问题;同时结合过采样SMOTE(synthetic minority oversampling technique)算法,平衡林火数据集和增加训练数据量,提升了森林火灾的预测准确度。结果表明,在面对更大的数据量时,该模型预测精度明显优于其他传统林火预测模型,证明了将深度学习应用在林火预测的优越性。该研究可为深度学习在林业领域的应用提供参考。     

Assessing forest vulnerability and the potential distribution of pine beetles under current and future climate scenarios in the Interior West of the US

The aim of our study was to estimate forest vulnerability and potential distribution of three bark beetles (Curculionidae: Scolytinae) under current and projected climate conditions for 2020 and 2050. Our study focused on the mountain pine beetle (Dendroctonus ponderosae), western pine beetle (Dendroctonus brevicomis), and pine engraver (Ipspini). This study was conducted across eight states in the Interior West of the US covering approximately 2.2 million km² and encompassing about 95% of the Rocky Mountains in the contiguous US. Our analyses relied on aerial surveys of bark beetle outbreaks that occurred between 1991 and 2008. Occurrence points for each species were generated within polygons created from the aerial surveys. Current and projected climate scenarios were acquired from the WorldClim database and represented by 19 bioclimatic variables. We used Maxent modeling technique fit with occurrence points and current climate data to model potential beetle distributions and forest vulnerability. Three available climate models, each having two emission scenarios, were modeled independently and results averaged to produce two predictions for 2020 and two predictions for 2050 for each analysis. Environmental parameters defined by current climate models were then used to predict conditions under future climate scenarios, and changes in different species’ ranges were calculated. Our results suggested that the potential distribution for bark beetles under current climate conditions is extensive, which coincides with infestation trends observed in the last decade. Our results predicted that suitable habitats for the mountain pine beetle and pine engraver beetle will stabilize or decrease under future climate conditions, while habitat for the western pine beetle will continue to increase over time. The greatest increase in habitat area was for the western pine beetle, where one climate model predicted a 27% increase by 2050. In contrast, the predicted habitat of the mountain pine beetle from another climate model suggested a decrease in habitat areas as great as 46% by 2050. Generally, 2020 and 2050 models that tested the three climate scenarios independently had similar trends, though one climate scenario for the western pine beetle produced contrasting results. Ranges for all three species of bark beetles shifted considerably geographically suggesting that some host species may become more vulnerable to beetle attack in the future, while others may have a reduced risk over time.     

Estimating forest net primary production under changing climate: adding pests into the equation

The current approach to modelling pest impacts on forest net primary production (NPP) is to apply a constant modifier. This does not capture the large spatial and temporal variability in pest abundance and activity that can occur, meaning that overestimates or underestimates of pest impacts on forest NPP are likely. Taking a more mechanistic approach that incorporates an understanding of how physiology is influenced by pest attack, enables us to better capture system feedbacks and dynamics, thereby improving the capacity to predict into novel situations such as changing climate, and to account for both changes in pest activity and host responses to the growing environment now and into the future. We reviewed the effects of pests on forest NPP and found a range of responses and physiological mechanisms underlying those responses. Pest outbreaks can clearly be a major perturbation to forest NPP, and it seems likely that the frequency and intensity of pest outbreaks, and the ways in which host species respond to pest damage, will change in the future. We summarized these impacts in the form of a conceptual model at leaf, tree and stand scales, and compared the physiological processes embedded within that framework with the capacity of a representative range of NPP models to capture those processes. We found that some models can encapsulate some of the processes, but no model can comprehensively account for the range of physiological responses to pest attack experienced by trees. This is not surprising, given the paucity of empirical data for most of the world's forests, and that the models were developed primarily for other purposes. We conclude with a list of the key physiological processes and pathways that need to be included in forest growth models in order to adequately capture pest impacts on forest NPP under current and future climate scenarios, the equations that might enable this and the empirical data required to support them.     

新西兰森林火险等级系统及预测应用

新西兰是森林火灾多发的国家,由于近年来气候变化导致新西兰火灾季节延长且火灾频发。探究新西兰火险等级体系的火天气指数和火行为分析模型,可以用于模拟历史、当前和未来的火险。文中对新西兰火险等级、火行为分析模型以及用于监测当前火险的火天气系统进行了探讨,认为开展相关火行为模拟研究是上述3种管理工具的研发基础,并且可以确保火险信息的有效性和准确性。不同植被的火行为是林火管理系统的重要输入因子,当前的天气、火行为和火险由新西兰国家乡村消防局进行每日更新并通过网络发布给全国的防火机构和公众,用于中、短期防火管理计划的实施。这些火行为模型已经成为新西兰很多火管理决策支持工具的基础。新西兰气象局把火天气指标系统和他们的预测模型结合起来,产生每小时的预测图用于短期计划和应急管理。借鉴新西兰的森林火险等级系统可以为我国构建完备的国家火险等级预报系统提供技术参考,而且探讨世界各国火险等级体系和火险模型可以促进我国与世界各国森林火险信息平台共享,提高预测和应对重特大森林火灾的能力,减少森林资源的损失,助力我国“双碳”目标的实现。     

气候变化情景下河北省3个优势树种适宜分布区预测

【目的】探究河北省3个优势树种分布与气候因子的关系,并进行适宜分布区预测,以期为评估气候变化的影响及制定适宜未来气候变化的森林经营策略提供理论依据。【方法】依据河北省森林资源调查数据,选取华北落叶松、蒙古栎和油松这3个主要树种,采用ClimateAP气候模型生成当前及未来(2040—2069年和2070—2099年)与降水和温度相关的10个气候因子,利用MaxEnt生态位模型和基于3个气候变化情景(温室气体最低排放,RCP2.6;中度稳定排放,RCP4.5;高度排放,RCP8.5)的一致性预测,模拟3个树种当前和未来的潜在适宜分布区,并采取响应曲线分析主要气候因子对3个树种适宜分布区的影响。【结果】3个树种MaxEnt模型的受试者工作特征曲线下面积(AUC值)都大于0.85,具有较好的预测能力;当前3个树种主要适宜分布在燕山和太行山地区;影响3个树种分布的主导气候因子存在差异,华北落叶松主要受小于0℃年积温和湿季降水量的影响,蒙古栎则主要受最热月平均气温、Hargreaves水分亏缺和湿季降水量的影响,而最热月平均气温、湿季降水量、大于5℃年积温和年均气温是影响油松分布的主要气候因子;一致性预测表明,在2040—2069年,河北省华北落叶松分布面积明显扩大,蒙古栎分布面积变化较小,而油松分布面积显著缩小;在2070—2099年,3个树种的适宜分布面积都显著缩小,幅度均超过3%。【结论】随着气候变化,3个树种均有向高海拔地区迁移的趋势,但在经纬度方向上的分布变化不大。在未来3个树种的适宜分布区,采取人工手段(如造林)辅助树种扩散以适应气候变化,有利于提高森林生产力,构建健康稳定的森林生态系统。     

Climate change and nature conservation in Central European forests: A review of consequences, concepts and challenges

With a predicted rise in average global surface temperature at an unprecedented rate, as well as changes in precipitation and disturbance regimes, climate change will bring forth new challenges for nature conservation in forest ecosystems. Species and habitats to be protected will be affected as well as related concepts and area specific objectives. Climate change impacts are likely to be aggravated by other anthropogenic stresses such as fragmentation, deposition or habitat destruction. To be reliable and effective, current objectives and guidelines of forest conservation need to be reassessed and improved. Our study analyses possible impacts of climate change on forests and identifies key future challenges for nature conservation in forests and ecosystem research. We reviewed 130 papers on climate change impacts on forest ecosystems and species published between 1995 and 2010. The geographical focus of the study is Central Europe. Papers were analysed accounting for direct and indirect impacts of gradual changes as well as stochastic disturbance events in forest ecosystems and their possible consequences for nature conservation.Even though broader aspects of nature conservation (protected areas, biodiversity) are frequently mentioned, little attention is given to forest-specific nature conservation. Particular aspects are insufficiently represented, such as the influence of climate change on different forest succession stages, the development of dead wood volume and quality, responses of secondary broadleaved species, azonal or extrazonal forests as well as ancient woodlands or remnants of historical silvicultural systems. Challenges arise in the context of great uncertainties about future developments. Nature conservation concepts and objectives in forests need to be adapted either within a permanent evaluation process or through the inclusion of further changes a priori, even if they are to some extent unpredictable. In some cases adaptation measures within nature conservation (e.g. adjusting protected areas) may conflict with interests of other stakeholders. Further research, particularly on interrelations between different impacts and the adaptive capacity of current forest ecosystems, associated species and existing genotypes is urgently needed. The scale and complexity of the task at hand calls for the establishment and further strengthening of international research networks.     

Change in current and future geographic distributions of Ulmus lamellosa in China

Prediction of potential geographic distributions is important for species protection and habitat restoration.Ulmus lamellosa is an endangered and endemic species in China for which conservation efforts are required.The maximum entropy(MaxEnt) model was used to predict the current and future geographic distribution(from 2030 to 2070) of U.lamellosa in China and discuss the reasons for changes in climatic suitability.The MaxEnt model provided a good fit to our data as confirmed by an AUC value of 0.948.The suitable areas for U.lamellosa were primarily projected in the northern part of China from 2030 to2070, especially in Liaoning province.The variables"temperature seasonality", "precipitation of wettest month" and "precipitation of warmest quarter" were the most influential climatic variables in limiting the distribution of U.lamellosa.Our results clearly predict the future impacts of climate change on the geographic distribution of U.lamellosa and this can help prioritize design of localized conservation strategies in China.     

Vulnerability of Pinus cembra L. in the Alps and the Carpathian mountains under present and future climates

Proactive management should be applied within a forest conservation context to prevent extinction or degradation of those forest ecosystems that we suspect will be affected by global warming in the next century. The aim of this study is to estimate the vulnerability under climate change of a localized and endemic tree species Pinus cembra that occurs in the alpine timberline. We used the Random Forest ensemble classifier and available bioclimatic and ecological data to model present and future suitable areas for P. cembra and estimate its current and future vulnerability. Future projections for years 2020, 2050 and 2080 were simulated using two IPCC Special Report on Emission Scenarios run under four global climate models.     

林火与气候变化研究进展

火是全球大多森林生态系统中的一个重要干扰因子, 它对大气中的温室气体和气溶胶的增加有显著影响。林火与气候变化是当前林火研究领域的热点问题。文中综述了气候变化对林火的影响和林火排放物对气候变化的影响。大量研究表明, 气候变化将导致森林火险期延长, 出现潜在极端火行为的天数增多, 森林火灾更加严重, 特别是北方森林火灾增加显著。未来的研究趋势是, 采用卫星遥感数据在大尺度上研究气候变化对林火的影响, 把林火模型与气候模式和全球植被动力学模型耦合, 构建更为复杂的林火排放模型, 以深入揭示林火与气候变化的关系。     

Uncertainty in detecting climate change impact on the projected yield of black spruce (Picea mariana)

The empirical growth and yield (G&Y) curves used in most timber supply models assume that growth conditions are invariable over time, which may not be correct given the projected future climate. However, errors in G&Y models can be quite large, and we therefore wanted to know the probability of detecting climate change-induced growth anomalies as a departure from current G&Y predictions. The work was carried out in a boreal forest study area in Eastern Canada, where Picea mariana Mill. BSP (black spruce) is the dominant species. Climate change effects were incorporated into G&Y predictions as correction factors obtained from process-based simulations of growth using current or projected climate. Uncertainty in G&Y projections was quantified through the inclusion of sampling and model errors in a bootstrap re-sampling scheme that yielded a percentile distribution of possible differences between G&Y curves that included or did not include climate change effects. Results yielded an average projected increase in productivity of 29% under future climate conditions for the black spruce strata in the study area. The probability of the climate change-modified G&Y predictions being significantly different from the current G&Y projections is 75% when considering the sampling error alone, and of 67% when both the sampling and modelling errors are included. Although incomplete in terms of what sources of error have been included, this study demonstrates the type of information that can be generated on the propagation of uncertainty in G&Y projections and, ultimately, on timber supply.     

Modeling the long-term natural regeneration potential of woodlands in semi-arid regions to guide restoration efforts

Understanding forest regeneration at sites previously used for agriculture underlies the establishment of science-based woodlands management policies. This is especially relevant in semi-arid areas, where the tree cover is critical in ameliorating the effects of aridity and in preventing desertification and land degradation. Natural regeneration in semi-arid areas occurs very slowly, which in part explains why it has hardly been studied. In the present work, we sought to devise a method to predict the natural regeneration potential of woodlands in semi-arid areas, to be used in guiding restoration efforts. Specifically, we evaluated holm oak coverage at a long-term ecological research site and then designed and validated a model to predict the natural regeneration of holm oak based on a few environmental variables. Unlike available studies, we obtained long-term information on tree regeneration (making use of >60 years of aerial photography) and climate (using long-term climate and microclimate data). We found that microclimate, measured using the potential solar radiation as a proxy, was a key driver of natural regeneration: after 60 years of agricultural abandonment, less sun-exposed areas attained a tree cover >90 %, whereas in more sun-exposed areas it remained below 20 %. We then used the model to map the natural regeneration potential, first in the study area and then in an area where holm oak plantations had been unsuccessfully introduced. In the latter case, the model successfully predicted the failure of this reforestation effort. Our results support the use of this model by decision makers to optimize management practices, as it will encourage the concentration of efforts in areas more prone to successful reforestation and allow the identification of areas more likely to benefit from natural regeneration processes.     

Climate change and the probability of wind damage in two Swedish forests

We simulated how possible changes in wind and ground-frost climate and state of the forest due to changes in the future climate may affect the probability of exceeding critical wind speeds expected to cause wind damage within one northern and one southern study area in Sweden, respectively. The topography of the study areas was relatively gentle and the forests were dominated by Norway spruce (Picea abies (L.) Karst.) and Scots pine (Pinus sylvestris L.). Using estimated changes in the net primary production (NPP) due to climate change and assuming a relative change in the site productivity equal to a relative change in NPP, we simulated possible future states of the forest under gradual adjustment of the site index in response to climate change using the model The Forest Time Machine. Global climate change scenarios based on two emission scenarios and one general circulation model were downscaled to the regional level. The modified WINDA model was used to calculate the sensitivity of the forest to wind and the probability of wind damage for individual forest stands for the periods 2011–2041 and 2071–2100 and for a control period 1961–1990. This was done while taking into account effects on stability of the forest from expected changes in the occurrence of ground frost. Increasing sensitivity of the forest to wind was indicated for both study areas when adhering to recommended management rules of today. Adding also a changed wind climate further increased the probability of wind damage. Calculated probabilities of wind damage were generally higher in the southern study area than in the northern one and were explained by differences in wind climate and the state of the forests, for example with respect to tree species composition. The indicated increase in sensitivity of the forest to wind under the current management regime, and possibly increasing windiness, motivate further analysis of the effects of different management options on the probability of wind damage and what modifications of Swedish forest management are possibly warranted.     

基于Landsat时间序列数据的森林地上生物量估测研究进展

利用时间序列遥感数据可以提高森林生物量估测精度和估测生物量动态变化,为森林碳储量和气候变化研究提供更为精确的反演基础。文中基于Landsat时间序列数据研究森林地上生物量的遥感估测方法,从数据预处理、数据重构、遥感因子选取以及模型构建与精度评价4个方面进行回顾和评述,并提出研究建议与展望。     

Weather and human impacts on forest fires: 100 years of fire history in two climatic regions of Switzerland

Understanding the factors driving past fire regimes is crucial in the context of global change as a basis for predicting future changes. In this study, we aimed to identify the impact of climate and human activities on fire occurrence in the most fire-prone regions of Switzerland. We considered forest fires, land use and meteorological data over the period 1904-2008 in the neighboring mountain cantons (states) of Valais and Ticino, which are characterized by distinct climatic regimes.The presence/absence of fire ignitions was analyzed using the Nesterov ignition index (as a proxy for fire weather), road density (for ignition sources), livestock density (for biomass removal), and change in forest area (for fire-prone abandoned agricultural areas).We found that fire weather played a key role in fire occurrence in both regions. Road and livestock densities had similar influences in the two cantons. However, while the increase in forest area was well correlated with fire occurrence in Ticino, no such correlation was evident in Valais, probably because land abandonment and forest cover change have been less extensive there. Our findings emphasize the non-linear nature of the relationships between fire occurrence and anthropogenic drivers, as we found thresholds above which road density was no longer correlated with fire occurrence. This implies that the projected future increase and spatial concentration of the human population may not result in a further increase in fire risk in intermediately to densely populated areas in both cantons.The driving factors behind fire activity differ slightly in the two cantons, in particular with increasing forest area enhancing fire occurrence in Ticino but not in Valais. These differences should be taken into account when assessing future fire risk, especially in Valais where the potential for an increase in the fire-prone area is still high. Fires are likely to become more frequent in a warmer climate, but future fire activity may develop differently in the two cantons. This should be taken into account when planning optimized fire prevention measures. This case study should help to better understand fire activity in highly populated regions where fire activity is still moderate but might markedly increase under a projected more fire-prone climate.     

Forest decline: modeling the effect of climate in tree rings

Tree rings provide an historical record of forest growth that reflects changes with time in site factors including, competition, tree and stand age, fire and other disturbances, and climate. Statistical methods can be used to factor out climatic influences on radial growth to yield a climate response model that can indicate whether declines in forest productivity are related to the modeled climatic variables or to other influences such as atmospheric pollutants. A general method, based on ordinary least squares, is presented for creating climatic response models for forest decline studies. The crux of the method is model verification, whereby the time-stability of the model is tested before it is used to forecast tree-ring variations during a period of decline. Three studies are described that employ monthly mean temperatures to predict tree-ring indices in declining red spruce (Picea rubens Sarg.) stands in the Appalachian Mountains of North America. The results indicate that, since 1960, red spruce trees at most of the sites sampled have undergone a period of growth decline that is unrelated to changes in mean monthly temperature. However, an association between annual ring width and unusual departures from the mean summer and winter temperatures during both present and past periods of decline suggests that climatic effects are implicated to some degree in the current decline.     

Should we respect the historical reference as basis for the objective of forest restoration? A case study from Northeastern China

Under climate change, the adoption of historical reference as the objective of forest restoration is being questioned. In this study, the spatially explicit forest landscape model LANDIS was utilized to analyze how the forest landscape in the upper Hun River area of Liaoning province in northeast China would be affected under current climate trends and future climate change; and to explore whether the historical reference should be the objective of restoration efforts. The results showed that (1) the area percentage (AP) of Quercusmongolica under climate change is always higher than that under the current climate regime, while the AP of Pinuskoraiensis is lower than that under current climate; and (2) the competitive ability of Q. mongolica and Populus davidiana increases, while that of other species decreases under climate change. As interspecies competition shifts under climate change, the historical reference appears in appropriate to serve as the objective of forest restoration. In addition, although Q. mongolica would likely benefit from a warmer and drier climate, use of this species for forest restoration under climate change still requires further research.     

基于全球数据库的森林土壤呼吸模型研究

森林土壤呼吸在陆地生态系统的碳平衡中发挥了重要作用,准确估算森林土壤呼吸量对于了 解陆地碳平衡的变化至关重要。这项研究以全球气候数据、全球森林土壤呼吸数据库为基础数据,通过 开发人工神经网络（ANN）模型建立由年平均气温（MAT）、年平均降水（MAP）、森林类型驱动的土壤 呼吸模型,预测全球森林土壤呼吸变化。模型估算的结果表明,从 1960 年到 2017 年,全球森林平均年 土壤呼吸量为 40.10±0.48 Pg C yr-1,全球森林土壤对全球土壤呼吸的贡献在 40.9% - 49.8% 之间。人工神 经网络模型预测的准确度达到 0.63,进一步改善了全球森林土壤呼吸模型预测的精度。     

Multiple-use forest management in consideration of climate change and the interests of stakeholder groups

In this study, the overall utility of forest management alternatives at the forest management unit level is evaluated with regard to multi-purpose and multi-user settings by a multi-criteria analysis (MCA) method. The MCA is based on an additive utility model. The relative importance of partial objectives of forest management (carbon sequestration, ground water recharge, biodiversity, and timber production) is defined in cooperation with stakeholders. The forest growth model 4C (Forest Ecosystems in a Changing Environment) is used to simulate the impact of six forest management strategies and climate on forest functions. Two climate change scenarios represent uncertainties with regard to future climatic conditions. The study is based on actual forest conditions in the Kleinsee management unit in east Germany, which is dominated by Scots pine (Pinus sylvestris L.) and oak (Quercus robur L. and Quercus petraea Liebl.) stands. First, there is an analysis of the impact of climate and forest management on forest functions. Climate change increases carbon sequestration and income from timber production due to increased stand productivity. Secondly, the overall utility of the management strategies is compared under the priority settings of different stakeholder groups. From an ecological perspective, a conservation strategy would be preferable under all climate scenarios, but the business as usual management would also fit the expectations under the current climate due to high biodiversity and carbon sequestration in the forest ecosystem. In contrast, a forest manager in public-owned forests or a private forest owner would prefer a management strategy with an intermediate thinning intensity and a high share of pine stands to enhance income from timber production while maintaining the other forest functions.     

Restoring forests: regeneration and ecosystem function for the future

Conventions and policies for biodiversity conservation and climate change mitigation state the need for increased protection, restoration and climate change adaptation of forests. Much degraded land may be targeted for large-scale forest restoration, yet challenges include costs, a shortage of regeneration material and the need for restored forests to serve as a resource for communities. To ensure ecosystem function for the future, forest restoration programs must: (1) learn from the past; (2) integrate ecological knowledge; (3) advance regeneration techniques and systems; (4) overcome biotic and abiotic disturbances and (5) adapt for future forest landscapes. Historical forest conditions, while site-specific, may help to identify the processes that leave long-term legacies in current forests and to understand tree migration biology/population dynamics and their relationship with climate change. Ecological theory around plant–plant interactions has shown the importance of negative (competition) and positive (facilitation) interactions for restoration, which will become more relevant with increasing drought due to climate change. Selective animal browsing influences plant–plant interactions and challenges restoration efforts to establish species-rich forests; an integrated approach is needed to simultaneously manage ungulate populations, landscape carrying capacity and browse-tolerant regeneration. A deeper understanding of limiting factors that affect plant establishment will facilitate nursery and site preparation systems to overcome inherent restoration challenges. Severe anthropogenic disturbances connected to global change have created unprecedented pressure on forests, necessitating novel ecological engineering, genetic conservation of tree species and landscape-level approaches that focus on creating functional ecosystems in a cost-effective manner.

     

Evaluation of the growth function of an empirical forest scenario model

Forests have an important role in the global carbon cycle as carbon pools, sinks and sources, and their quantification has become a relevant task. Empirical models based on national forest inventories are widely used for the assessment of carbon sequestration. However, these models do not treat explicitly all the processes occurring in the ecosystem, as they are mainly based on statistical relations to estimate forest development. Therefore, there is a need for validation of these models to increase confidence in the predictions of future forest development. This study evaluates an empirical single-tree model that was developed in Switzerland (MASSIMO). The accuracy and precision of the growth function of the model is evaluated with data from the National Forest Inventory (NFI) of Liechtenstein. MASSIMO was found to predict the basal area per hectare of the Liechtenstein data very precisely (underestimation of 0.65%). The main differences between observed and predicted diameter increment occur mostly for larger DBH classes, where the increment is underestimated by the model. However, these differences may be related to the precision of the input variables. For example, the explanatory variable stand age is determined with relatively low precision; therefore it shows a high variability. For future model development, either the variable stand age should be estimated more reliably, or stand age should not be an explanatory variable of the growth function.