基于随机森林模型的城市不透水面提取研究 ——以呼和浩特市为例

城市不透水面信息对于城市生态环境动态演化过程研究具有重要意义。以Landsat 8遥感影像为数据源,以呼和浩特市为实证区域,进行了随机森林模型应用于城市不透水面的提取研究,并与目前应用广泛的支持向量机模型进行了对比分析。研究表明：在不同的抽样比例训练样本条件下,随机森林模型对于城市不透水面的提取精度均优于支持向量机的提取精度;对于随机森林模型和支持向量机模型,70%的训练样本比例均为最佳训练样本抽样比例。在该抽样比例下,随机森林模型提取城市不透水面的总体分类精度为93.29%,Kappa系数为0.9051,支持向量机模型的总体分类精度为91.26%,Kappa系数为0.8757;随机森林模型对于城市裸土的识别度较高,能更好地将城市裸土和不透水面进行区分,而支持向量机模型对于城市裸土、不透水面和绿地的区分能力均弱于随机森林模型。综合而言,随机森林模型对城市不透水面的提取精度优于支持向量机模型,随机森林模型可以有效应用于城市不透水面提取领域,进一步丰富了城市不透水面提取方法体系构成。     

Landslide susceptibility mapping using random forest and boosted tree models in Pyeong-Chang,Korea

Landslides susceptibility maps were constructed in the Pyeong-Chang area, Korea, using the Random Forest and Boosted Tree models. Landslide locations were randomly selected in a 50/50 ratio for training and validation of the models. Seventeen landslide-related factors were extracted and constructed in a spatial database. The relationships between the observed landslide locations and these factors were identified by using the two models. The models were used to generate a landslide susceptibility map and the importance of the factors was calculated. Finally, the landslide susceptibility maps were validated. Finally, landslide susceptibility maps were generated. For the Random Forest model, the validation accuracy in regression and classification algorithms showed 79.34 and 79.18%, respectively, and for the Boosted Tree model, these were 84.87 and 85.98%, respectively. The two models showed satisfactory accuracies, and the Boosted Tree model showed better results than the Random Forest model.     

城市生态用地需求测算研究进展与展望

作为城市地域自然生态系统服务供给的基本载体,生态用地基于土地利用/土地覆被的内在关联将人类社会发展与自然生态演替耦合关联在一起,已成为统筹解决城市建设扩展与自然生态保护矛盾的重要综合途径;城市生态用地的需求测算直接决定供需均衡分析结果,是城乡统筹生态用地规划与管理的关键环节。本文在系统梳理生态用地概念内涵的基础上,对比分析了生态需求与生态用地需求的概念异同;提出基于空间与功能的城市生态用地需求测算概念框架,并据此将已有测算方法归并为经验判定法、生态系统服务法、空间格局法3大类,从综合性、代表性、阈值与规模、位置等方面对比分析了法定定额法、历史趋势预判法、生态足迹法、碳氧平衡法、生态安全格局法等具体方法的基本特征及其优缺点;最后将下一步研究的重点趋向概括为基于土地生态功能的城市生态用地分类、基于土地生态格局的城市生态用地效益综合测算、基于多目标情景的城市生态用地功能综合权衡、基于生态用地供需平衡分析的城市生态腹地识别4个方面。     

水分对米槠天然林土壤有机碳矿化和微生物群落的影响

通过室内设置30%~80%WHC(Water holding capacity,土壤持水量,WHC)土壤含水量水平,研究土壤含水量变化对亚热带米槠天然林土壤有机碳矿化和微生物群落的影响。试验包括3种处理,处理1:干湿交替处理(30%~80%WHC交替);处理2:恒干处理(30%WHC);处理3:恒湿处理(80%WHC)。研究结果表明:1)整个培养过程,CO2排放速率均呈现波动式下降。湿润期CO2累积量显著高于干旱期CO2累积量,并且随着交替次数的增加,湿润期与干旱期CO2累积值之差逐渐增大。2)采用因子分析法得出干湿交替并不影响土壤真菌和革兰氏阳性菌群落结构,但对土壤放线菌和革兰氏阴性菌的群落有较大影响。     

广州白云山风景区森林自然度评价

森林自然度是描述森林状态的指标,是指森林生态系统保持原生状态或受人类干扰的程度大小,对其进行评价和等级划分是森林近自然经营研究中的关键环节。研究基于空间尺度,构建白云山森林自然度评价体系,通过格网形式对研究区域进行细化,并对白云山森林自然度进行评价。结果表明:白云山森林自然度指标值大部分处于0.31~0.60和0.61~0.80之间,就其等级而言,属于第三、第四等级,即较低等级;而整个研究区范围内自然度最高的是荷依岭景区、明珠楼景区与摩星岭景区相交的天然次生林群落,自然度最低的是白云山群落的外围度假区。     

不同林龄人促更新林枯枝落叶层碳及养分贮量

对福建南平顺昌县不同林龄米槠(Castanopsis carlesii)人促更新林枯枝落叶层和主要营养元素现存量进行测定,探究其随林龄的变化趋势。结果表明:不同林龄米槠人促更新林枯枝落叶层现存量为3.94~4.95 t·hm-2,其中49 a生人促林枯枝落叶层现存量最小,31 a生的最大。不同林龄枯枝落叶层C、N、P、K的浓度分别为44.09%~47.46%,11.9~13.8 g·kg-1,0.5~0.9 g·kg-1,1.4~5.1 g·kg-1。21 a生和49 a生人促更新林枯枝落叶层C浓度明显高于31 a生林分。49 a生人促更新林枯枝落叶层N和P浓度均高于21 a生和31 a生林分,但该林分枯枝落叶层K浓度却最低。21 a生米槠人促更新林枯枝落叶层C贮量分别是31 a生和49 a生林分的1.03倍和1.23倍,49 a生米槠人促更新林K贮量远低于21 a生和31 a生。不同林龄枯枝落叶层碳和养分贮量差异与林分特征、各林分枯枝落叶层现存量及养分浓度等有关。     

Total and pyrogenic carbon stocks in black spruce forest floors from eastern Canada

Pyrogenic carbon (PyC), a by-product of recurrent boreal wildfires, is an important component of the global soil C pool, although precise assessment of boreal PyC stock is scarce. The overall objective of this study was to estimate total C stock and PyC stock in forest floors of Eastern Canada boreal forests. We also investigated the environmental conditions controlling the stocks and characterized the composition of the forest floor layers. Forest floor samples were collected from mesic black spruce sites recently affected by fire (3–5 yr) and analyzed using elemental analysis and solid state ¹³C nuclear magnetic resonance (NMR) spectroscopy. PyC content was further estimated using a molecular mixing model. Total C stock in forest floors averaged 5.7 ± 2.9 kg C/m² and PyC stock 0.6 ± 0.3 kg C/m². Total stock varied with position in the landscape, with a greater accumulation of organic material on northern aspects and lower slope positions. In addition, total stock was significantly higher in spruce-dominated forest floors than stands where jack pine was present. The PyC stock was significantly related to the atomic H/C ratio (R² 0.84) of the different organic layers. ¹³C NMR spectroscopy revealed a large increase in aromatic carbon in the deepest forest floor layer (humified H horizon) at the organic-mineral soil interface. The majority of the PyC stock was located in this horizon and had been formed during past high severity fires rather than during the most recent fire event.     

水稻冠层氮素含量光谱反演的随机森林算法及区域应用

利用地面实测数据构建高精度的水稻冠层氮素含量光谱反演点模型并将其进行尺度转换,实现了水稻冠层氮素含量准实时、大区域监测。以氮素光谱敏感指数作为输入变量,冠层氮素含量数据为输出变量,利用随机森林算法构建水稻冠层氮素含量高光谱反演模型,并用苏州市水稻农田验证区数据,检验模型的普适性和有效性;利用准同步的Hyperion数据,采用对输入、输出变量进行线性变换的简单尺度转换方法实现了点模型的区域应用。结果表明:基于随机森林算法的水稻冠层氮素含量高光谱反演模型可解释、所需样本少、不会过拟合、精度高(模型在实验区的预测精度为R2=0.82,验证区检验精度为R2=0.73)且具有普适性;点模型基于高光谱遥感卫星影像和尺度转换进行区域应用,精度较高(R2=0.81)。     

Video monitoring of in-channel wood: From flux characterization and prediction to recommendations to equip stations

Wood flux (piece number per time interval) is a key parameter for understanding wood budgeting, determining the controlling factors, and managing flood risk in a river basin. Quantitative wood flux data is critically needed to improve the understanding of wood dynamics and estimate wood discharge in rivers. In this study, the streamside videography technique was applied to detect wood passage and measure instantaneous rates of wood transport. The goal was to better understand how wood flux responds to flood and wind events and then predict wood flux. In total, one exceptional wind and seven flood events were monitored on the Ain River, France, and around 24,000 wood pieces were detected visually. It is confirmed that, in general, there is a threshold of wood motion in the river equal to 60% of bankfull discharge. However, in a flood following a windy day, no obvious threshold for wood motion was observed, which confirms that wind is important for the preparation of wood for transport between floods. In two multi-peak floods, around two-thirds of the total amount of wood was delivered on the first peak, which confirms the importance of the time between floods for predicting wood fluxes. Moreover, we found an empirical relation between wood frequency and wood discharge, which is used to estimate the total wood amount produced by each of the floods. The data set is then used to develop a random forest regression model to predict wood frequency as a function of three input variables that are derived from the flow hydrograph. The model calculates the total wood volume either during day or night based on the video monitoring technique for the first time, which expands its utility for wood budgeting in a watershed. A one-to-one link is then established between the fraction of detected pieces of wood and the dimensionless parameter “passing time × frame rate ”, which provides a general guideline for the design of monitoring stations.     

Scientists' warning on extreme wildfire risks to water supply

2020 is the year of wildfire records. California experienced its three largest fires early in its fire season. The Pantanal, the largest wetland on the planet, burned over 20% of its surface. More than 18 million hectares of forest and bushland burned during the 2019–2020 fire season in Australia, killing 33 people, destroying nearly 2500 homes, and endangering many endemic species. The direct cost of damages is being counted in dozens of billion dollars, but the indirect costs on water-related ecosystem services and benefits could be equally expensive, with impacts lasting for decades. In Australia, the extreme precipitation (“200 mm day −1 in several location”) that interrupted the catastrophic wildfire season triggered a series of watershed effects from headwaters to areas downstream. The increased runoff and erosion from burned areas disrupted water supplies in several locations. These post-fire watershed hazards via source water contamination, flash floods, and mudslides can represent substantial, systemic long-term risks to drinking water production, aquatic life, and socio-economic activity. Scenarios similar to the recent event in Australia are now predicted to unfold in the Western USA. This is a new reality that societies will have to live with as uncharted fire activity, water crises, and widespread human footprint collide all-around of the world. Therefore, we advocate for a more proactive approach to wildfire-watershed risk governance in an effort to advance and protect water security. We also argue that there is no easy solution to reducing this risk and that investments in both green (i.e., natural) and grey (i.e., built) infrastructure will be necessary. Further, we propose strategies to combine modern data analytics with existing tools for use by water and land managers worldwide to leverage several decades worth of data and knowledge on post-fire hydrology.