湛江市红树林资源及其可持续利用

广东省湛江市现有红树林面积7242.0hm²,人工造林未成林25.6hm²,天然更新红树林509.4hm²,另有红树林宜林地9688.6hm²。红树林及其伴生植物主要有木榄(Bruguiera gymnorrhiza)、秋茄(Kandelia candel)、无瓣海桑(Sonneratia apetala)等15科24种。由于毁林养虾、养鱼等导致1980～2001年占用红树林地面积6363.6hm²。从总体上看,湛江红树林种类多,群落类型丰富;红树林面积大幅减少,影响区域社会经济发展;除乡土种外,无瓣海桑是一种较成功的外来引树种;宜林地多,但人工造林保存率不高;破坏红树林,挖塘养殖依然较普遍。为了实现湛江红树林的可持续发展,需要加强下列工作:建立自然保护区加强对现有红树林湿地的保护;开展红树林湿地生态系统资源监测和鸟类环志;科学规划后加快红树林的恢复和重建;积极推动退塘还林并合理利用红树林;依法保护管理好红树林;多途径筹集保护资金;加强红树林、湿地的保护和宣传。     

淇澳岛两种红树林区大型底栖动物群落比较

对珠海淇澳岛天然秋茄(Kandelia candel)与人工恢复无瓣海桑(Sonneratia apetala)两种红树林林地间大型底栖动物群落结构进行了比较研究.结果显示,天然秋茄林与人工无瓣海桑林林地大型底栖动物种类组成及数量分布都存在差异,秋茄林大型底栖动物平均生物量、栖息密度分别为668.64g·m^-2和8370个·m^-2,无瓣海桑林分别为511.64g·m^-2和4748个·m^-2.两个红树林剖面不同潮位带大型底栖动物结构存在差异,秋茄林生物量和栖息密度的最高值都在中潮带,主要是由节肢动物和软体动物所主导:无瓣海桑林栖息密度的最高值也是在中潮带,生物量变化比较不规则.两个红树林剖面大型底栖动物群落特征的季节变化较明显,生物量都是秋季处于最高水平:栖息密度都是冬季处于最高水平,秋季最低:生物多样性都是夏季处于最低,但最高值秋茄林剖面是在春季,而无瓣海桑林剖面则是秋季.研究结果表明,林地底质和盐度等生境条件以及红树林群落的高度、郁闭度和林下通风透光等生态特征的差异是造成红树林底栖动物群落差异的重要原因.     

深圳福田红树林植被碳储量和净初级生产力

红树林是滨海湿地“蓝碳”的主要类型之一.准确和定位评估不同植物群落的固碳能力,对于红树林保育管理和恢复造林具有指导作用.本研究对深圳福田红树林4种代表性群落(白骨壤群落、秋茄群落、海桑群落、无瓣海桑群落)的各个植被碳库组分(乔木植物生物量碳库、林下灌丛碳库、呼吸根碳库、枯立木碳库、枯倒木碳库和枯枝落叶层碳库等)进行调查,计算各群落的植被碳储量,并通过生长增量-凋落物产量法计算得到各群落的净初级生产力.结果表明: 白骨壤群落、秋茄群落、海桑群落和无瓣海桑群落的植被碳储量分别为28.7、127.6、100.1、73.6 t C·hm^-2,各群落的净初级生产力分别为8.75、7.67、9.60、11.87 t C·hm^-2·a^-1.位于深圳市中心的福田红树林,每年固定大气CO₂高达4000 t.本研究结果将为红树林“蓝碳”碳汇功能的评估提供理论指导,并为我国红树林碳汇林建设提供依据.     

珠海淇澳岛无瓣海桑种苗天然更新与扩散分析

无瓣海桑(Sonneratia apetala Buxh-Ham Buxh-Ham.)在珠海淇澳岛抑制互花米草(Spartina alterniflora Loisel.)蔓延恢复红树林方面发挥了积极的作用。然而,由于它是外来引进种,在当地能否繁殖更新或造成种苗大量扩散成灾?为此,针对1999年以来在淇澳岛种植的大片无瓣海桑林及其周边湿地进行了种子产量与种苗天然扩散性的调查。结果表明:淇澳岛4年生无瓣海桑果实年产量为30.51kg·棵^-1,果粒数为2706粒·棵^-1,每果粒内含30～50粒种子。无瓣海桑幼苗在水沟两边分布最多,为40株·16m^-2,其次是其林分边缘的光滩,为8株·16m^-2,互花米草内为3.5株·16m^-2,最少是无瓣海桑林内,1株·16m^-2,种子多但幼苗少,且这些扩散苗绝大多数为当年生幼苗,很少发现多年生小苗,极难更新成林。     

我国亚热带次生林乔木地上生物量估算的适宜样地面积初探

我国亚热带森林生物量估算研究常基于400~900 m²的小面积样地,但到底多大面积样地才较为适宜却鲜有探究。该文以浙江九龙山国家级自然保护区内三个1 hm²样地亚热带次生林为研究对象,利用生物量回归方程估算木本植物(胸径≥1 cm)的地上生物量,分析地上生物量的空间分布格局,并利用移动窗口法探讨三个次生林地上生物量估算的适宜样地面积。结果表明:(1)三个次生林木本植物的地上生物量分别为63.75 Mg·hm^-2(大岩前)、84.70 Mg·hm^-2(八通岭)和128.20 Mg·hm^-2(屁股窟),地上生物量集中分配在个体数量较少的大径级个体;屁股窟次生林的地上生物量空间变异程度高于大岩前和八通岭次生林。(2)利用移动窗口法确定的三个次生林木本植物地上生物量估算的适宜样地面积分别为2025 m²(大岩前)、2500 m²(八通岭)和3600 m²(屁股窟),森林地上生物量越高且空间变异程度越高,所需调查的样地面积越大。该研究结果可为我国亚热带森林地上生物量估算的样地面积设置提供证据,并为该区域森林生物量与碳储量的估算提供基础数据。     

苗期无瓣海桑与互花米草的营养竞争关系研究

采用间植分隔法开展了无瓣海桑和互花米草营养竞争关系研究,结果表明竞争作用刺激了无瓣海桑高、径、冠幅、分枝及叶片数的增长,却抑制了叶面积的增加。地上竞争可增加无瓣海桑的地上生物量,但对根冠比影响不大。地下竞争增加了互花米草的地上生物量,降低了其根生物量,此时根冠比最小。地下竞争促进了无瓣海桑的主根生长,抑制了侧根生长,而地上竞争则正相反;随着竞争程度的加深,根活力逐渐降低。地下竞争时互花米草根活力显著下降。无瓣海桑和互花米草在有地上竞争时叶绿素a和b的含量均降低。各竞争处理未对P、K、Ca和Mg元素的含量产生显著影响,但地上竞争显著降低了N元素在无瓣海桑和互花米草根茎叶中的含量。互花米草的竞争强度略高于无瓣海桑,但是地上竞争强度无瓣海桑远远高于互花米草,而二者地下竞争强度相当。     

湿地植物生长模型在红树林湿地人工恢复调控中的应用--以珠江口淇澳岛红树林湿地恢复为例

选取珠江口淇澳岛红树林人上恢复为例,借助湿地大型植物生长控制模型,通过3种不同预案的设置,定量分析外来种无瓣海桑与生态侵入种互花米草之间的竞争关系。研究表明：在有互花米草的滩涂上人工种植的无瓣海桑．在其生物量小于互花米草生物量时,初始生物量大的超过互花米草所需时问短于初始生物量小的;当无瓣海桑生物量超过互花米草生物量后．无瓣海桑生长基本小受互花米草的影响,同时互花米草的生物量开始下降。在人工付出劳动相同的情况下（即人人工割除互花米草的频度）．初始生物量大的无瓣海桑超过互花米草生物量所需的时间短于初始生物最小的无瓣海桑。     

三亚河外来红树植物无瓣海桑种群动态与扩散特征

无瓣海桑具有耐贫瘠、生长快等特点,已成为我国东南沿海滩涂主要造林先锋种。然而,作为典型外来物种,无瓣海桑是否具有入侵性也一度引起各界广泛关注。三亚河是我国无瓣海桑引种的最南区域,但目前尚未有对该区无瓣海桑入侵性的报道。本研究利用全面踏查与样地调查相结合的方法,分析了三亚河外来红树植物无瓣海桑的种群动态及扩散特征。结果表明:无瓣海桑在样地群落上层平均密度为4 ind·100 m^-2,重要值为41.2%,具有较明显优势,而在群落中下层无其小树和幼苗分布,无瓣海桑种群更新难以维持;群落中、下层红树植物的Shannon指数分别为1.68和1.09,Simpson指数分别为0.78和0.57,物种多样性较高,本地红树植物长势良好,生存空间未受到明显挤占;三亚河亦未见无瓣海桑种群的自然扩散,较高的盐度可能是抑制无瓣海桑种群扩散的关键。鉴于无瓣海桑优良的生长适应能力,可在无淡水汇入的滨海湿地红树林生态工程中作为先锋种进行合理利用,同时对其做好长期监测与定位研究。     

高原鼠兔对高寒草甸植物群落生物量的影响

2010 年和2011 年8 月在青海省果洛州矮嵩草草甸,采用样方法测定了不同密度高原鼠兔栖息地内杂草斑和秃斑植物群落特征及地上生物量的变化。结果表明:2010 年与2011 年相比,相同高原鼠兔密度的栖息地内杂草斑和秃斑植物总盖度、平均高度和总地上生物量差异不显著,但同一年份不同密度栖息地间差异显著。高原鼠兔低、中、高密度栖息地植物总盖度、平均高度和总地上生物量显著低于无鼠兔对照组(P < 0.05),不同密度间无明显变化规律。低、中、高密度栖息地的莎草、禾草和豆科植物盖度、平均高度和地上生物量显著低于无鼠兔对照组(P <0.05),秃斑上豆科植物消失;杂类草平均高度在低密度栖息地和无鼠兔对照组显著高于中、高密度栖息地(P < 0. 05),盖度和地上生物量对照组最小,随高原鼠兔密度增加呈增加趋势。2010 年和2011 年,高原鼠兔低、中、高密度栖息地内杂草斑面积分别占样地面积的4. 0% 、4.3% 、13. 3% 和3.8% 、4. 3% 、11. 0% ,秃斑面积分别占样地面积的0 2% 、2 6% 、4 0% 和0 2% 、2 2% 、3 4% ;植物损失地上生物量分别为110.84 kg/ hm² 、203. 18 kg/ hm² 、431.58 kg / hm² 和107.67 kg / hm² 、189. 46 kg / hm² 、365. 72 kg/ hm² 。高原鼠兔
的密度(x)与植物损失地上生物量(y)之间存在对数曲线关系,说明高原鼠兔的活动显著降低了植物总盖度、平均高度和总地上生物量,同时也改变了植物功能群组成。随高原鼠兔密度增加,栖息地内杂草斑、秃斑的面积和植物损失地上生物量呈增加趋势。     

无瓣海桑与三种乡土红树植物混交的生态效应

为了解人工种植红树林的生态效应,对2001年种植于广州市南沙区的三种人工混交红树林进行了群落结构和土壤理化性状的调查分析。结果表明:由于红海榄的不耐荫导致样地3无法形成复层结构,但也不排除是红海榄化感作用的影响;无瓣海桑+木榄群落、无瓣海桑+秋茄群落的群落结构和对土壤中重金属的吸收效果优于无瓣海桑+红海榄群落,因此在该实验的立地条件下营造红树林混交林时可以考虑无瓣海桑与秋茄、木榄混交,不应将无瓣海桑与红海榄混交;随着演替的进行,该实验地在无重大干扰的情况下将向着近天然红树林生态系统演化。     

基于机器学习算法的草地地上生物量估测——以祁连山草地为例

草地地上生物量（Aboveground Biomass,AGB）是指导畜牧业生产管理的重要指标,是草畜平衡综合分析的基础。目前,有关祁连山草地AGB反演的研究较少,且多源数据间的尺度差异问题并未得到很好的解决。为了解祁连山草地AGB的空间分布状况,利用Sentinel-2多光谱数据、无人机（Unmanned Aerial Vehicle,UAV）数据以及2021年植被生长期实测草地AGB数据实现了空天地一体化监测,通过决策树回归（Decision Tree Regression,DTR）、随机森林回归（Random Forest Regression,RFR）、梯度提升决策回归树（Gradient Boosting Regression Tree,GBRT）以及极致梯度提升（eXtreme Gradient Boosting,XGBoost）共4种算法反演草地AGB的适用性分析,利用最优模型反演了祁连山草地的AGB空间分布状况。结果表明：研究区内多种植被指数所表现出的特性有所差异。祁连山地区AGB在空间分布上呈现出由西北向东南递增的趋势,平均AGB为925.43kg/hm²。6种植被指数与实测AGB之间均表现为显著正相关,适合作为祁连山草地AGB遥感反演的指标;XGBoost模型较其它模型具有最高的R²值（0.78）和精度（74.75%）、最低的均方根误差（RMSE,99.74 kg/hm²）和平均绝对误差（MAE,71.60 kg/hm²）,模型反演效果最好;UAV数据能够提供更加详细的空间细节特征,减小Sentinel-2数据和实地采样数据间的尺度差异;因此,基于6种植被指数与祁连山草地AGB间的相关性,构建XGBoost模型反演研究区草地AGB空间分布状况是具有实践意义的。研究结果将为指导祁连山草地畜牧业的发展和维护草地生态系统的平衡提供一定的参考价值与数据支撑。     

Biomass accumulation and carbon storage of four different aged Sonneratia apetala plantations in Southern China

The objectives of this study were to examine plant biomass accumulation and carbon (C) storage in four different aged Sonneratia apetala plantations in the Leizhou Bay in South China. The allometric equations using diameter at breast height (DBH) and height (H) were developed to quantify plant biomass. The total forest biomass (TFB) of S. apetala plantation at 4, 5, 8, and 10 years old was 47.9, 71.7, 95.9, and 108.1 Mg ha^?1, respectively. The forest biomass C storage in aboveground (AGB) and roots at 4, 5, 8, and 10-year plantation was 19.9, 32.6, 42.0, 49.0 Mg ha^?1, respectively. Soil organic C (SOC) on the top 20 cm of sediments increased by 0.3, 6.8, 27.4, and 35.0 Mg ha^?1after 4, 5, 8, and 10 years of reforestation, respectively. The average annual rate of total carbon storage (TCS) accumulation at 4, 5, 8, and 10-year S. apetala plantation was 5.0, 7.9, 8.7, and 8.4 Mg ha^?1 yr^?1, respectively. The TCS values in this study were underestimated because we only estimated SOC storage on the top 20-cm sediments in these plantations. This study suggests these young S. apetala plantations have the characteristics of fast growth, high biomass accumulation, and high C storage capacity, especially in sediments. They sequestrated C at a high but varying rate over time. The large-scale reforestation of S. apetala plantations in the open coastal mudflats in southern China has great potential to sequestrate more C as well as restore the degraded coastal land. The potential ecological issues associated with the increasing monoculture plantations were discussed. More long-term monitoring and research are needed to further evaluate biomass and C accumulation of S. apetala plantations over time as well as how the increasing distribution of this monoculture plantation will influence the few native mangrove remnants.     

联合GF-6和Sentinel-2红边波段的森林地上生物量反演

光谱反射率能反映地物差异,是森林地上生物量（Aboveground Biomass,AGB）遥感反演的理论基础。红边波段处于近红外与红光波段交界处快速变化的区域,能对植被冠层结构和叶绿素含量的微小变化做出快速反应,对植被生长状况较敏感。研究以GF-6和Sentinel-2多光谱影像作为数据源,结合野外调查AGB数据,构建落叶松和樟子松AGB线性和非线性估测模型,通过比较模型精度选择最优模型进行森林AGB反演和空间分布制图。结果表明：GF-6和Sentinel-2影像红边波段反射率与落叶松、樟子松AGB均呈显著相关（P<0.05）,红边波段对AGB估测较敏感。多变量估测模型整体估测效果优于单变量模型,所有模型中多元线性回归模型取得了最优的决定系数（落叶松R²=0.66,樟子松R²=0.65）和最低的均方根误差（落叶松RMSE=31.45 t/hm²,樟子松RMSE=54.77 t/hm²）。相比单个数据源,联合GF-6和Sentinel-2影像构建的多元线性回归模型估测效果得到了显著提升,模型RMSE对于落叶松和樟子松AGB估测分别最大降低了22.9%和11.2%。增加红边波段进行AGB估测能显著提高模型估测精度,三组数据源分别加入红边波段信息后进行建模,模型RMSE得到了显著降低。GF-6拥有800 km观测幅宽和高效的重访周期,可以快速地提供大尺度时间序列数据,在森林地上生物量反演和动态监测方面有着很大潜力。     

Allometric growth and carbon storage in the mangrove Sonneratia apetala

Allometric growth reflects different allocation patterns and relationships of different components or traits of a plant and is closely related to ecosystem carbon storage. As an introduced species, the growth and carbon storage of Sonneratia apetala are still unclear. To derive allometric relationships of the mangrove S. apetala and to estimate carbon storage in mangrove ecosystems, we harvested 12 individual Sonneratia apetala trees from four different diameter classes in the Futian National Nature Reserve, Guangdong, China. Allometric growth models were fitted. The results showed that diameter at breast height (DBH) and wood density were better variables for predicting plant biomass (including above- and below-ground biomass) than plant height. There were significant power function relationships between biomass and DBH, with a mean allometric exponent of 2.22, and stem biomass accounted for 97% of the variation in S. apetala total biomass. Nearly isometric scaling relationships were developed between stem biomass and other biomass components. To better understand the carbon stocks of the S. apetala ecosystem, we categorized all trees into five age classes and quantified vegetation carbon storage. The S. apetala vegetation carbon storage ranged from 96.48 to 215.35 Mg C ha^?1, and the carbon storage significantly increased with stand age. The allometric equations developed in this study are useful to estimate biomass and carbon storage of S. apetala ecosystems.

     

森林生物量遥感降尺度研究

森林生物量是评价全球碳氧平衡、气候变化的重要指标。目前已有基于星载激光雷达数据的全球森林生物量产品,但空间分辨率较低,不能很好地满足小区域森林调查和动态监测的需要。针对这一现状,以美国马里兰州两个森林分布状况不同的区域为研究区,基于CMS(Carbon Monitoring System)30 m分辨率和GEOCARBON 1 km分辨率森林地上生物量产品以及TM等数据源,通过升尺度模拟低分辨率生物量数据和直接使用低分辨率产品两种方式,分别尝试建立了多光谱地表参数和低分辨率森林地上生物量之间的统计关系,以此作为降尺度模型实现了森林地上生物量空间分辨率从1 km到30 m的转换,并对降尺度结果进行精度评价和误差分析。结果表明:模拟数据降尺度后的30 m分辨率森林地上生物量空间分布和CMS森林地上生物量分布状况大致相同,RMSE=59.2—65.5 Mg/hm~2,相关系数约为0.7;其降尺度结果优于GEOCARBON产品直接降尺度结果RMSE=75.3—79.9 Mg/hm~2;相较于线性模型,非线性模型能更好地呈现森林地上生物量和地表参数间的关系;总体上,降尺度生物量呈现高值区低估,低值区高估的现象。     

Forest aboveground biomass estimation using machine learning regression algorithm in Yok Don National Park,Vietnam

Forest biomass is one of the key measurement for carbon budget accounting, carbon flux monitoring, and climate change studies. Hence, it is essential to develop a credible approach to estimate forest biomass and carbon stocks. Our study applied Sentinel-2 satellite imagery combined with field-measured biomass using Random Forest (RF), a machine learning regression algorithm, to estimate forest aboveground biomass (AGB) in Yok Don National Park, Vietnam. A total of 132 spectral and texture variables were extracted from Sentinel-2 imagery (February 7, 2017) to predict AGB of the National Park using RF algorithm. It was found that a combination of 132 spectral and texture variables could predict AGB with an R² value of 0.94, RMSE of 34.5 Mgha⁻¹ and %RMSE of 18.3%. RF regression algorithm was further used to reduce the number of variables in such a way that a minimum number of selected variables can be able to estimate AGB at a satisfactory level. A combination of 11 spectral and texture variables was identified based on out-of-bag (OOB) estimation to develop an easy-to-use model for estimating AGB. On validation, the model developed with 11 variables was able to predict AGB with R² = 0.81, RMSE = 36.67 Mg ha⁻¹ and %RMSE of 19.55%. The results found in the present study demonstrated that Sentinel-2 imagery in conjunction with RF-based regression algorithm has the potential to effectively predict the spatial distribution of forest AGB with adequate accuracy.     

Competitive Control of an Exotic Mangrove Species: Restoration of Native Mangrove Forests by Altering Light Availability

Recently, many studies have focused on the possibility of restoring mangrove ecosystems by introducing fast‐growing mangroves. However, methods for managing an exotic fast‐growing species to restore mangrove ecosystems and at the same time preventing invasion by introduced species remains unclear. Sonneratia apetala Buch‐Ham is one example of an exotic mangrove with both high ecological value and potential risk for invasion after introduction. To investigate the possibility of reducing the potential for invasion by altering light availability, we simulated different irradiances of S. apetala understory in the greenhouse. For each irradiance treatment, three levels of competition between S. apetala and native mangroves Aegiceras corniculatum (L.) were used: no competition, intraspecific competition and interspecific competition. Compared with A. corniculatum, S. apetala showed a significantly higher growth rate for both height and biomass accumulation under full irradiation. Compared to the full irradiation treatment, the shading treatment significantly reduced the height, total biomass and biomass allocation to leaves of S. apetala by 61.31, 71.0, and 76.2%, respectively, whereas the growth of A. corniculatum was not affected. The results suggested that lowering light availability could inhibit the growth of S. apetala and increase the competitiveness of A. corniculatum. Planting introduced fast‐growing mangroves at a density of approximately 2,000 plants/hm² is an effective strategy for preventing potential invasion and restoring wetland habitats. By taking advantage of the differences in shade tolerance between fast‐growing exotic mangroves and native mangroves, introduction of fast‐growing mangroves in coastal areas could have huge potential for reforesting mangrove ecosystems.     

寒害后珠海淇澳岛红树林群落水土生要素的CCA分析

选取2008年寒害后和恢复重建5年后红树林群落水土生要素进行纵向对比,选取水域pH、化学需氧量(COD)、溶解氧(DO)、总悬浮物(TSS)、总无机氮(TIN)、正磷酸盐(PO_4-P)等水环境指标和(S)pH、氧化还原电位(Eh)、有机质(OM)等沉积物指标进行不同类型红树林群落间的横向对比。自变量指标中水环境指标所占比例较大,约为2—3,弥补了红树林湿地植被、沉积物和水文等3个自然因素中水文条件研究较为缺乏的现状。结合各类红树林群落水、土、生三要素进行典范对应分析(CCA)的结果表明:1)典范对应分析方法用于分析淇澳岛大围湾各类红树林典型群落寒害后及寒害5年后的群落生物过程特征与各种生境因子间的相关关系,效果良好,前4组自变量的典型变量可以解释标准变量中绝大部分(约80%以上)的信息。2)恢复重建5年后,淇澳岛大围湾各类红树林群落中中潮位的无瓣海桑(Sonneratia apetala)群落和高潮位的桐花树(Aegiceras comiculatum)群落拓展最快,面积分别增加26.58hm~2和23.1 hm~2,而互花米草(Spartina alterniflora)群落因无瓣海桑群落的扩张而受到抑制,面积下降31.51hm~2。3)水域pH、沉积物(S)pH、OM、DO、COD等水土指标与冠幅、胸径、基径、盖度和高度等生物过程指标依次正相关,而PO_4-P、TSS和TIN等水域指标与生物过程指标依次负相关。4)珠海淇澳岛大围湾红树林湿地生态系统对水域悬浮物和营养元素的固定效果明显,但其植被生物过程受水土因素,尤其是水文因素,影响较大,其中影响最大的生境因素是pH值(包括水域pH值和沉积物(S)pH值)。低中潮位、中性至弱碱性的表层沉积物生境条件较为适合无瓣海桑和老鼠簕(Acanthus ilicifolius)等红树植物群落的初期恢复。     

Restoration of mangrove plantations and colonisation by native species in Leizhou bay,South China

To examine the natural colonisation of native mangrove species into remediated exotic mangrove stands in Leizhou Bay, South China, we compared soil physical–chemical properties, community structure and recruitments of barren mangrove areas, native mangrove species plantations, and exotic mangrove species—Sonneratia apetala Buch.Ham—between plantations and natural forest. We found that severely degraded mangrove stands could not regenerate naturally without human intervention due to severely altered local environments, whereas some native species had been recruited into the 4–10 year S. apetala plantations. In the first 10 years, the exotic species S. apetala grew better than native species such as Rhizophora stylosa Griff and Kandelia candel (Linn.) Druce. The mangrove plantation gradually affected soil physical and chemical properties during its recovery. The exotic S. apetala was more competitive than native species and its plantation was able to restore soil organic matter in about 14 years. Thus, S. apetala can be considered as a pioneer species to improve degraded habitats to facilitate recolonisation by native mangrove species. However, removal to control proliferation may be needed at late stages to facilitate growth of native species. To ensure sustainability of mangroves in South China, the existing mangrove wetlands must be managed as an ecosystem, with long-term scientific monitoring program in place.     

A new synergistic approach for Sentinel-1 and PALSAR-2 in a machine learning framework to predict aboveground biomass of a dense mangrove forest

Mangroves are well-recognized for their very high carbon sequestration potential. However, studies on their role in global carbon cycling and climate change are hindered due to lack of operational methodologies to model and map their biophysical properties. This study establishes a robust methodological protocol for aboveground biomass (AGB) estimation using i) field measurements, ii) a generic (in the absence of species-specific) allometric equation, iii) multi-sensor/polarization SAR data and derived variables thereof, and iv) machine learning models; that demonstrated high prediction accuracy (R² = 0.93) and low uncertainty (mean ≤ 3% and median ≤ 1.5%) for a mangrove forest. Following stratified random sampling and on-field accessibility criteria, we laid out 314 elementary sampling points of 0.04 ha each at Bhitarkanika wildlife sanctuary (BWS), India and measured circumference at breast height (CBH) and canopy tree height for 18 species. The estimated AGB range of a generic allometric equation was 9–474 Mg/ha for BWS, with a major representation of 9–347 Mg/ha. We utilized Sentinel-1 and ALOS-2/PALSAR-2 and derived their variables for AGB prediction. Compared to single sensor-based model, we observed higher prediction accuracy for combined sensor data (R² = 0.63, 0.87, 0.93; RMSE = 66.75, 39.95, 28.35 Mg/ha; MAE = 52.63, 24.21, 19.55 Mg/ha; and Bias = 3.42, 0.22, 2.15 Mg/ha for C, L and C & L bands respectively using a Generalized Additive Model (GAM) over Random Forest (RF), Gradient Boosting Machines (GBM) and Support Vector Regression (SVR). The higher uncertainty pixels represented seasonal grassland and scrubs in the swampy areas and along the fringes of the creeks that experience diurnal tidal fluctuations. This robust methodology can be replicated for AGB estimates in other mangrove ecosystems to meet the operational carbon accounting requirements of various entities.