基于高分辨率遥感影像的北亚热带森林生物量反演

以北亚热带湖北省太子山林场为研究对象,基于高空间分辨率GF-2与SPOT-6卫星影像,提取不同窗口大小下的纹理信息与光谱信息,利用随机森林回归算法,并结合野外实测106块样地的生物量数据,建立不同影像下的太子山林场森林生物量反演模型。结果显示：（1） GF-2和SPOT-6虽然空间分辨率有差异,但是从其不同波段反射率的相关系数（0.75、0.78、0.73、0.61）发现,两种影像的波段反射率具有较高的相关性,说明两者的辐射性能相近;（2）通过分析不同纹理特征对生物量模型的影响,发现均值和对比度纹理参数对生物量反演具有很好的效果。（3）高分辨率的遥感数据在生物量反演中具有较好的表现,且GF-2生物量模型精度（R²=0.88,RMSE=27.11 Mg/hm²）与SPOT-6生物量模型的精度（R²=0.89,RMSE=23.93 Mg/hm²）相近。（4）两种影像对不同森林类型的生物量预测值不存在显著差异,都适合对不同林分类型的生物量进行预测。     

河西走廊植被净初级生产力时空变化及其影响因子研究

干旱半干旱区植被NPP变化对全球碳循环有重要影响,该区域NPP对气候变化的响应表现出较大的时空异质性,其驱动机制并不十分清楚。选择中国河西走廊,利用随机森林算法估算了2002-2018年的NPP,基于偏导数法计算了气候与人类活动对NPP的影响。结果表明：（1）随机森林算法能较好的适用于干旱半干旱区NPP估算。（2）2002-2018年间河西走廊年NPP的平均值为153.32 gC m^-2 a^-1,总量为37.468 Tg C/a,呈东南向西北递减的分布特征,研究期间NPP呈2.37 gC m^-2 a^-1（P=0.09）增长趋势。（3）河西走廊NPP变化52.51%由气候因子贡献,47.49%由人类活动贡献。（4）在气候变化对NPP的影响中,降水主导了该区72.21%的区域,温度对NPP变化量的贡献占73.71%,前者影响着NPP变化格局,后者主导NPP变化数量。升温和增湿均有利于该区NPP增加,随着西北地区气候暖湿化,河西走廊植被会持续改善,该研究有助于理解干旱半干旱区NPP对气候变化的响应机制,为适应气候变化政策制定提供理论依据。     

Repeated measures random forests (RMRF): Identifying factors associated with nocturnal hypoglycemia

Nocturnal hypoglycemia is a common phenomenon among patients with diabetes and can lead to a broad range of adverse events and complications. Identifying factors associated with hypoglycemia can improve glucose control and patient care. We propose a repeated measures random forest (RMRF) algorithm that can handle nonlinear relationships and interactions and the correlated responses from patients evaluated over several nights. Simulation results show that our proposed algorithm captures the informative variable more often than naïvely assuming independence. RMRF also outperforms standard random forest and extremely randomized trees algorithms. We demonstrate scenarios where RMRF attains greater prediction accuracy than generalized linear models. We apply the RMRF algorithm to analyze a diabetes study with 2524 nights from 127 patients with type 1 diabetes. We find that nocturnal hypoglycemia is associated with HbA1c, bedtime blood glucose (BG), insulin on board, time system activated, exercise intensity, and daytime hypoglycemia. The RMRF can accurately classify nights at high risk of nocturnal hypoglycemia.     

Evolutionary heritage shapes tree distributions along an Amazon‐to‐Andes elevation gradient

Understanding how evolutionary constraints shape the elevational distributions of tree lineages provides valuable insight into the future of tropical montane forests under global change. With narrow elevational ranges, high taxonomic turnover, frequent habitat specialization, and exceptional levels of endemism, tropical montane forests and trees are predicted to be highly sensitive to environmental change. Using plot census data from a gradient traversing > 3,000 m in elevation on the Amazonian flank of the Peruvian Andes, we employ phylogenetic approaches to assess the influence of evolutionary heritage on distribution trends of trees at the genus‐level. We find that closely related lineages tend to occur at similar mean elevations, with sister genera pairs occurring a mean 254 m in elevation closer to each other than the mean elevational difference between non‐sister genera pairs. We also demonstrate phylogenetic clustering both above and below 1,750 m a.s.l, corresponding roughly to the cloud‐base ecotone. Belying these general trends, some lineages occur across many different elevations. However, these highly plastic lineages are not phylogenetically clustered. Overall, our findings suggest that tropical montane forests are home to unique tree lineage diversity, constrained by their evolutionary heritage and vulnerable to substantial losses under environmental changes, such as rising temperatures or an upward shift of the cloud‐base.     

Application of machine learning to identify predators of stocked fish in Lake Ontario: using acoustic telemetry predation tags to inform management

Understanding predator–prey interactions and food web dynamics is important for ecosystem-based management in aquatic environments, as they experience increasing rates of human-induced changes, such as the addition and removal of fishes. To quantify the post-stocking survival and predation of a prey fish in Lake Ontario, 48 bloater Coregonus hoyi were tagged with acoustic telemetry predation tags and were tracked on an array of 105 acoustic receivers from November 2018 to June 2019. Putative predators of tagged bloater were identified by comparing movement patterns of six species of salmonids (i.e., predators) in Lake Ontario with the post-predated movements of bloater (i.e., prey) using a random forests algorithm, a type of supervised machine learning. A total of 25 bloater (53% of all detected) were consumed by predators on average (± S.D. ) 3.1 ± 2.1 days after release. Post-predation detections of predators occurred for an average (± S.D. ) of 78.9 ± 76.9 days, providing sufficient detection data to classify movement patterns. Tagged lake trout Salvelinus namaycush provided the most reliable classification from behavioural predictor variables (89% success rate) and was identified as the main consumer of bloater (consumed 50%). Movement networks between predicted and tagged lake trout were significantly correlated over a 6 month period, supporting the classification of lake trout as a common bloater predator. This study demonstrated the ability of supervised learning techniques to provide greater insight into the fate of stocked fishes and predator–prey dynamics, and this technique is widely applicable to inform future stocking and other management efforts.     

高血压性心脏病生物标志物的筛选：基于中国重庆市7家医疗机构的数据分析研究*

目的: 筛选高血压性心脏病（HHD）的影响因素,建立HHD的预测模型,为HHD的发生提供预警。方法: 选取中国重庆市某医科院校数据研究院平台2016年1月1日至2019年12月31日主要诊断为高血压性心脏病和高血压患者。通过单因素分析、多因素分析筛选HHD的影响因素,采用R语言分别构建Logistics模型、随机森林（RF）模型和极限梯度上升（XGBoost）模型。结果: 单因素分析筛选出60项差异指标,多因素分析筛选出18项差异指标（P<0.05）。Logistics模型、RF模型、XGBoost模型曲线下面积（AUC）分别为0.979、0.983和0.990。结论: 本文建立的3种HHD预测模型结果稳定,其中XGBoost模型对于HHD的发生具有良好的诊断效应。     

High-altitude tree growth responses to climate change across the Hindu Kush Himalaya

兴都库什喜马拉雅地区高海拔树木生长对气候变化的响应 高海拔地区快速升温可能导致树木对温度响应更为敏感,而限制高海拔地区树木生长的关键气候因子以及气候变化对树木生长产生多大程度的影响尚不清楚。本研究在兴都库什喜马拉雅地区收集了73 个样点的树轮数据,包括3个优势属的树种(Abies属、Juniperus属和Picea属),样点海拔均在3000 m以上。 将时间动态规整(dynamic time warping)的方法用于建立亚区域年表,以考虑不同站点年表之间变化的同步 性。同时,定量分析了气候因子对树木生长的贡献以及树木生长与气候因子关系的时空动态。研究结果发现,73个站点年表可以聚为3类,且与其所处的生物气候区相对应,即西喜马拉雅地区,中东喜马拉雅地区和藏东南地区。在干旱的西喜马拉雅地区,树木生长与冬、春两季的降水呈正相关关系,而在湿润的藏东南地区,树木生长与冬季温度和春季降水呈正相关关系。树木生长受最低温度的影响最大,特别是冬季温度,其重要性从西到东呈现递增趋势。滑动窗口相关分析表明,在中西喜马拉雅地区,影响树木生长的冬季温度信号在减弱,然而在藏东南地区该信号随着1980年以来的快速升温而增强。本研究结果表明,若该地区升温持续,在西喜马拉雅地区可能会因变暖引起的水分亏缺而造成森林衰退,而在藏东南地区因树木生长得益于变暖而使得森林扩张。     

黄河首曲玛曲县高寒湿地景观格局演变

高寒湿地是青藏高原地区最重要的生态水源涵养区之一, 也是局部气候的有效调节者, 其动态变化与成因亟待深入研究。该研究基于遥感图像分析、地理信息系统空间分析和景观生态指数分析结合的方法, 以黄河首曲玛曲县高寒湿地为研究对象, 对1995-2018年6期湿地的动态变化进行研究。结果表明, 研究区湿地在1995-2010年间不断退化, 1995-2010年湿地面积总共减少了18 680.31 hm²。在2010-2018年间黄河首曲高寒湿地面积有所增加, 但与20世纪90年代相比, 21世纪初开始湿地的面积普遍呈现下降趋势; 1995-2010年湿地斑块数不断增加, 斑块密度不断增大, 平均斑块面积下降, 景观的破碎度升高; 2010-2015年湿地斑块数和斑块密度减少, 2015-2018年湿地斑块数和斑块密度增加, 平均斑块面积先增大后减小, 景观的破碎度先降低后升高。1995-2010年研究区高寒湿地景观Shannon多样性指数和Shannon均匀度指数均呈现下降的趋势, 湿地的景观结构趋于简单, 景观类型分布更加集中。2010-2018年湿地景观Shannon多样性指数和Shannon均匀度指数均呈现上升趋势, 湿地的景观结构趋于复杂, 景观类型增加且分布更加分散。进一步的驱动力分析表明, 引起黄河首曲高寒湿地景观格局演变的主要因素是蒸发量和降水量, 其次是人口数量和大牲畜数量等人类活动影响。气候因子是影响黄河首曲高寒湿地面积变化的主要原因, 过度的人类经济活动在一定程度上加剧了湿地的变化。     

Resprouting Ability of Dry Forest Tree Species after Disturbance Does Not Relate to Propagation Possibility by Stem and Root Cuttings

Tropical dry forest tree species are recognized for their high resprouting ability after disturbance. We tested whether species that commonly produce root and stem suckers can be propagated by large stem and root cuttings, a useful method for landscape restoration programs. We performed four experiments: (1) In a greenhouse, we tested the propagation of six species using large stem cuttings collected from early successional sites. We used the following treatments: (i) dry season collection and planting; (ii) dry season collection, storage in humid soil, and wet season planting; (iii) wet season collection and planting; and (iv) wet season collection and planting after treatment with commercial NAA auxin. (2) Stem cuttings of Myracrodruon urundeuva were planted in a pasture during the rainy season after either NAA, IBA, or no auxin treatment. (3) As a control experiment, we also planted cuttings of Spondias mombin, a species known for successfully regenerating from cuttings. (4) Root cuttings of six species were collected in recently plowed pastures and planted in the greenhouse with and without treatment with NAA auxin. No root cuttings rooted. Only M. urundeuva and Astronium fraxinifolium stem cuttings rooted. Maximum success was obtained for stem cuttings collected and planted in the dry season (23%). Only 13% of M. urundeuva had sprouted by the 15th month of the field experiment. As a result, large cuttings are not recommended for propagation of the studied species. Future studies should include development of suitable methods of root harvesting and prospection of traditional knowledge for species selection.