融合景观连通性的城镇规划与生物多样性生态服务效能优化

著：(澳)陈思清 汪洁琼

译：王南

Authors: (Australia) CHEN Si-qing, WANG Jie-qiong

Traslator: WANG Nan

融合景观连通性的城镇规划与生物多样性生态服务效能优化

著：(澳)陈思清 汪洁琼

译：王南

Authors: (Australia) CHEN Si-qing, WANG Jie-qiong

Traslator: WANG Nan

随着世界范围内濒危物种灭绝速率的日益增长，生物多样性已成为生态系统服务中最重要的组成部分之一。无论地区、区域还是全球尺度范畴，生物多样性的保护都是保持生态系统活力的关键。景观连通性包含两层含义，既是对景观自然结构的描述，也包含特定物种对该结构的响应，它为景观连通性原理应用于景观规划实践提供了理论基础，尤其是对那些通过生态系统服务的整合来改善环境影响的项目，这种理论基础的作用更为突出。基于美国东南部生态城镇开发项目的案例分析，本文试图探索出能够促进景观连通性并推进生物多样性保护的设计思路。基于地理信息系统(GIS)与空间数据分析(FRAGSTATS)，本研究旨在量化美国东南部以林地和林带为特色的景观连通性。在美国东南部，由于房地产业持续扩张、新增住宅不断开发，大量拥有自然景观的区域正经历着城镇化进程。研究结果表明，通过增加栖息地斑块面积以及它们之间的连通性，在基地内已建绿色基础设施中增植新的林带可显著增加景观连通性，从而有效地提供生态系统服务。结论部分，本研究聚焦在如何通过更高的景观连通性来实现新城发展需求与生态系统服务供给两者间的平衡，也因此提出了设计干预需将宜居性与可持续性并重的观点。

风景园林；景观连通性；生态系统服务；生物多样性保护；景观规划

1 引言

“景观连通性”这一概念强调的是物种特性与景观结构在决定栖息地斑块内部有机体运动时的互动[1]。人类活动，如农业发展、商业造林、基础设施建设和城镇化等，已导致栖息地破碎化，即原栖息地的丧失、栖息地斑块面积减少或隔离，以及景观连通性的降低[2]。然而，这些关键性的因素并未受到科学研究的重视，许多实际运行的土地开发项目则是号称已为景观连通性的提高做出了努力。实际上，由于缺乏对景观连通性含义的正确理解，这些研究和项目反而可能会减小景观连通性对土地管理和生物多样性保护的潜在效用。根据 Taylor等人(1993)最初的定义，景观连通性是“景观推动或阻碍资源斑块间运动的程度”[3]，这一定义强调了景观内部栖息地或土地利用的种类、数量及布局对种群动态及群落结构产生变化的影响。景观连通性应将上述两层含义结合，既包含对景观自然结构的描述(结构连通性)，也包含特定物种对该结构的响应(功能连通性)，这正是景观连通性原理应用于景观规划及设计实践的理论基础。基于地理信息系统(GIS)，本研究对景观连通性进行了量化。通过对美国南部某住宅开发项目的案例研究，本文进一步探索了景观连通性的含义及其在城市规划中的应用。

生态系统服务已被定义为人们从不同生态系统中能直接或间接获取的利益，如大自然提供的生物多样性为人类提供的各种益处，包括食物、纤维、气候及温度的调节、授粉以及能够加强不同人群福祉的审美价值观[4]。然而，人口增长、城镇化及相关的自然资源开发已导致全世界约60%的生态系统服务普遍退化[4]。要逆转这一趋势就意味着必须尽快改变人们不可持续的现代生活方式以及对生态系统服务认知的缺乏。因此，如何使人们正确理解生态系统提供服务的方式就变得至关重要，人们对于生态系统服务和产品价值的认知亟需提升。景观和城市规划的从业者也必须意识到人类，无论城市居民或乡村居民，都只是地球上栖息的物种之一，他们必须摒弃将生态系统服务视为理所应当的态度，参与到保护生态系统的活动中来，才能更好地保持生态系统服务的可持续供给。

生物多样性具有多种定义及多重考量标准。它可以被定义为“地球上生命的多样性”[4]，也常被认为是“支撑生态系统过程的调节者，最终实现生态系统服务或提供产品”[5]。可以说，生物多样性是最重要的生态系统服务，而生物多样性的保护能够在不同的空间尺度上实现。在全球或区域范围内，2011年通过的《生物多样性公约》确立了爱知生物多样性目标，即为保护和促进全球生物多样性而制定的一系列目标。其中，目标11呼吁要加强17%的陆地面积(不包括南极)的保存及保护，尤其是生物多样性受到威胁的区域更应努力达到这一目标。该面积相当于2 294万km2，约等于加拿大、中国和印度3个国家的面积总和[6]。从大的空间尺度及长期视角谈保护生物多样性固然有其重要意义，但保存及保护措施的具体落实却只能在地区或场地范围内实现。就场地层面来看，生物多样性会受到土地利用方式(如农业发展)的巨大影响[7]。它与4个影响栖息地质量的因素相关：每个威胁的相对影响、每个栖息地对每个威胁的相对敏感度、威胁源与栖息地间的距离、以及该土地利用方式受到法律保护的程度。此外，威胁有时正是人类主导的景观本身，如农田或城市地区[8]。本研究将从栖息地连通性和生物多样性保护入手，对场地层面的设计干预进行调查研究。

1 黑色地带的位置及区域环境，阿拉巴马州黑色地带的传统县及哈德逊农场的位置Location and regional context of the black belt, traditional counties in the Alabama black belt, and location of Hudson farm

2 案例研究的项目场地

2.1 林带

美国南部广袤的乡村景观以林带乔木或树丛作为显著特征。除了限定空间与定义边界，林带在乡村景观中也提供了许多其他的生态系统服务，例如为人类提供食物、衣物、庇护，提升视觉质量，维持乡村景观的真实性等。这些林带的生态系统服务可就其彼此间的关系进行评估[9-13]。然而，林带在生物保护中最重要的功能在于它们为诸如鸟、鼠、蝴蝶等野生动物提供了重要的栖息地。与此同时，具备生态廊道功能的林带可以保持并增大景观连通性，保持生态可变性，从而保护并提升景观的生物多样性[14-18]。

由高耸的林带围合起来的土地所形成的斑块景观是美国乡村为人熟知的传统特征。林带实际上是带状的树木、树丛或林地，它们通常是野生动物的重要栖息地，对景观的视觉质量也具有十分重要的作用。本研究试图通过对哈德逊农场项目进行案例分析，阐明景观连通性如何得以量化。哈德逊农场位于美国东南地区的“黑色地带”。“黑色地带”最初是指阿拉巴马州中部及密西西比州东北部的大草原和黑色土壤，但现已被长期用来指称美国南部的一个非裔美国人口众多的广阔区域。通常认为，“黑色地带”覆盖了佐治亚州中部、佛罗里达州北部、密西西比州西部、阿拉巴马州中部及南部、路易安娜州中部及东部、卡罗莱纳州东部及北部、弗吉尼亚州东南部的广大区域。哈德逊农场正位于这一“黑色地带”之上，是阿拉巴马州首府蒙哥马利县东南部的一个郊区(图1)。哈德逊农场的最显著的特色正是这种林带乔木或树丛所形成的景观。林带构建了一系列斑块所形成的网络，创造出高耸的乔木林带和灌木形成的廊道环绕着较低的田野的景观。

林带和林地，对于鸟类、蜜蜂、鼠类、蝴蝶等野生动物而言，是重要栖息地(图2)。同时，具备生态斑块功能的林地和具备生态廊道功能的林带可在保持栖息地类型的多样性和景观连通性方面相辅相成。在这一案例中，林带和林地形成的斑块—廊道—基质对于保护及促进场地的生物多样性至关重要。林带廊道和林地斑块作为项目场地内的绿色基础设施，不仅使乡村景观具备了强烈的场所感，而且能唤起人们对美国乡村的情感共鸣[19-20]。因此，必须对林带廊道和林地斑块进行深入探究，通过生态景观规划和创新性城市设计策略将其融入土地开发项目。

2 哈德逊农场上的灌木树篱、草地、林地栖息地及生物多样性Hedgerow, grassland, and woodland habitat and biodiversity on Hudson Farm

3 场地的高程、坡度、朝向分析Elevation, slope and slope aspect analysis of the site

2.2 地形特征

项目基地的显著特征是一条山脊线将整个场地分成了若干个次一级汇水区(图3)。基地的最高点310英尺(译者注：约94m)坐落于场地的南侧，而基地的最低点227英尺(译者注：约69m)则坐落于场地的北侧。哈德逊农场以地形稍有起伏的草原和疏林草地景观为特征。

GIS的坡度分析显示整个基地较为平坦，大部分的土地的坡度都在7%以下(图3)。坡度朝向分析则显示：山脊线南侧的大部分地块都是西向或西南向的，而基地北侧的大部分地块朝向都是北向或者东北向的(图3)。基于坡度朝向的分析，可以进一步地考虑日照、利用太阳能作为可替换能源方式的使用潜能等。

3 景观连通性

3.1 结构连通性

通常情况下，使用“连通性”这一术语来强调结构层面，景观连通性就简单地等同于景观促进传播的线性特征，如自然连接的线性廊道。这种意义上的连通性意味着，从A出发沿一定路线可到达B。而在网络系统中(图4)，如果A到B有更多可选路线，就认为该网络系统联系更紧密，即连通性更高。

4 点A至点B的路线数量，体现环密度及自然连通性Number of travelling routes from point A to point B, showing loop density and physical connectivity

5 林带廊道及连通性(a)飞地；(b)飞地间距离；(c)飞地缺失；(d)网络连通性；(e)环及可选路线；(f)交叉影响Hedgerow corridor and connectivity: (a) stepping stone; (b) distance between stepping stone; (c) loss of stepping stone; (d) network connectivity; (e) loops and alternatives; (f) intersection effect

6 哈德逊农场上作为景观基础设施的林带廊道(a)航拍林带廊道与场地边界的叠加；(b)场地内作为生态基础设施的林带廊道及林地斑块Hedgerow corridors as landscape infrastructure on Hudson Farm (a) aerial photo overlaid with site boundary, and (b) hedgerow corridors and woodland patches as ecological infrastructure on site

3.2 功能连通性

自然连通性通常以网状系统中环的数量来衡量。但在景观生态学中，通常采用的连通性衡量方法不仅注重自然连接的线性廊道，也关注斑块区域以及斑块间距离对斑块间运动的影响，即非自然连接的廊道，如某些物种用作连接廊道的飞地(图5)。这种连接被称作功能连通性[21]。

宽度和连通性是廊道五大主要功能—栖息地、通道、过滤、源、汇—的主要控制因素[9,22-23]。廊道内缺口对某一物种运动的影响取决于该缺口相对物种运动幅度的长度以及廊道和缺口间的差别。然而，一排飞地(小斑块)在廊道和非廊道之间只具备中等连通性，因此它只能为斑块间的内部物种运动起到中等作用(图5a)。对于高度视觉导向的物种来说，其在飞地之间运动的有效距离则取决于其看到每个连续飞地的视觉能力(图5b)。一个用作其他斑块间运动飞地的小斑块的缺失，通常会抑制运动进而加大斑块隔离(图5c)。而对大斑块间的飞地群进行最优空间布局可以生成备选或冗余路线，同时在大斑块间保持整体的线性导向阵列[30]。许多研究均已证明，这种结构可以促进野生动物的运动[21-22, 24-32]。

3.3 栖息地连通性

哈德逊农场是一个家庭农场，它位于阿拉巴马州蒙哥马利县郊区，占地2 200英亩(译者注：约合8.9km2)。该农场土地曾主要用于牛群放牧和干草收割。场地内的景观要素，如树、洞穴、林带、谷仓和篱笆，构成了蒙哥马利县城乡结合部的独特地标。深入了解和理解场地是做好规划设计的基本要求，而对诸如树、洞穴、林带、谷仓和篱笆这些与众不同的地标进行保护和加强，则能保持该场地独一无二的特点。地产开发的目标是打造一个步行为主、混合使用的紧凑型新社区，并不采取用途单一的传统郊区发展模式。在可持续发展的道路上，应注重打造完整的社区和小镇，而非零星式的郊区发展[33]。

在哈德逊农场，景观中的林带以廊道的形式存在(图6)，是支撑生物多样性的重要栖息地。农场区域的航拍照片展现了该场地的景观结构(图6a)。农场周围是大面积的河流廊道和湿地，为候鸟和其他野生动物提供了重要的栖息地。哈德逊农场之所以独特，就在于它拥有大面积的森林斑块、开阔的田野、林带廊道以及高林带围起的大片“景观空间”等景观要素。高大的林带乔木和树丛，或点缀、或包围着开阔的田野，放眼望去，农场上的林带和乔木呈现出深远的视觉感受，增强了农场的场所感(图6b)。

4 景观尺度的连通性量化

4.1 基于GIS的景观指标

景观连通性的量化有多种方法。例如，基于福尔曼(Forman)的城市生态角度[9]，利用联结与节点的数量，可通过一个简易公式对连通性(con)进行计算。

con= L / 3 (V-2) (1)

其中，

L = 联结的数量；

V = 节点的数量。

这种方法在景观联结与节点均容易识别的精密尺度下可能方便使用，但在景观尺度上，场地情况高度多样化、复杂化(比如有不同形式的连通性)，就需要使用基于GIS的新方法对连通性进行量化。

很多基于GIS的景观指标都是由景观生态学家和科学家们提出并供公众使用的[34]。FRAGSTATS是美国麻省大学景观生态实验室开发的程序，用来计算不同地图模式下多种景观指标(包括景观连通性)，其初始软件(版本2)于1995年向公众发布，与之联合发布的是美国农业部的《森林服务通用技术报告》[5]。本文运用了其版本3.3，该版本可从该实验室网站下载。FRAGSTATS对连通性的计算均基于连通性指标(表1)。

例如，连通性指标(con)可以计算为：

其中，

Cijk=基于用户指定的距离阈值，属同一斑块类型的斑块j与斑块k间的接合点(0=未接合，1=接合)。

ni=每一斑块类型i的景观内部的斑块数量。

在这个方程式中，连通性等于所有同类型斑块间功能性接合点的数量(Cijk总和，如斑块j和k不在彼此的指定距离内，那么Cijk=0；如斑块j和k在指定距离内，那么Cijk=1)，除以所有同类型斑块间的所有可能存在的接合点总数，再乘以100以转化为百分比。因此连通性数值在0～100之间。当景观仅由一个斑块构成，或所有类别只有一个斑块，或景观内部所有斑块均不互相连接(即在用户指定的另一个同类斑块的距离阈值内)时，连通性=0。当景观内部每个斑块均有连接时，连通性=100[34]。

4.2 数据输入与模型参数化

使用FRAGSTATS空间格局分析程序，需要准备好可识别格式的文件作为GIS数据输入程序中，以计算景观指标。在运行并输出统计数据前，该程序必须进行正确的参数化(图7)。关于FRAGSTATS程序、输入数据准备、模型参数化及输出数据格式的细节请见该软件网站。

表1 FRAGSTATS中使用的连通性指标Tab.1 Connectivity metrics used in FRAGSTATS

7 FRAGSTATS程序参数化界面Parameterization interface of the FRAGSTATS program

4.3 实证研究：哈德逊农场

为保障景观内部野生动物活动并增强生物多样性，维持廊道网络是设计功能完整且健康的景观时需要遵循的基本原则。在哈德逊项目的总体规划制定过程中，林带之间相互连接，且有连贯的乔木覆盖。这一理念为诸多景观生态学家所倡导[35]。该网络叠加于排水系统和已有林带之上。其空间布局也考虑了如农场主与工人关系等历史因素[36]。基于以上观点，对已有林带的研究就此展开(图8)。

为了进一步增加连通性，提出新的林带与已有林带连接以形成一个林带网络。该网络将作为绿色基础设施网络或生态基础设施网络，兼具设计与生态功能。基本目的是为了展示它们为什么要连接起来，并如何连接起来。左侧图(图8a)显示了已有林带，中间图(图8b)的红色区域表示新增种植的林带，右侧图(图8c)则是已有林带与新增种植林带的叠加效果。

8 哈德逊农场上已有林带与计划种植林带的连接Existing and proposed hedgerow connection on Hudson Farm

9 用于包含生物多样性及相关生态系统服务的景观规划的尺度思考框架Scale thinking framework for landscape planning considering biodiversity and related ecosystem services

5 结果

根据FRASTATS的空间格局分析程序提供的数据结果可以发现，通过整合已建绿色基础设施和新增的林带廊道，该地的景观连通性已提升40%(表2)。新增的林带种植在现有的农田边、废弃的农场设备原址且遍布牧场，它们连接起现有的残存树木及树丛，形成了互相联结的林带网络。因此，在未牺牲现有作为核心栖息地的大型景观斑块的情况下，就可以实现40%的连通性增长。新计划的林带重新连接起破碎的廊道，大幅增加了整体景观连通性。这表明保持景观完整性以及自然植被廊道至关重要。

分析表明，可将FRAGSTATS同GIS结合使用，能计算出景观尺度的连通性及其他参数。但表2中的数据结果必须谨慎解读。比如，连通性通过“连接性”和“内聚力”来衡量(表1、2)；以公式(2)测量的“连接性”提高了40.7%，而景观“内聚力”却未以同等幅度增加，反而略有下降[34]。该例表明，设计干预可以改变场地生态，因此必须在设计实施后对结果进行恰当评估以实现设计干预，而数字技术的进步已使这一过程变得越发容易。

6 讨论

对生物多样性及相关生态系统服务的保护已成为全球性的挑战，需以跨越时间尺度与空间尺度的系统方法来应对。生物多样性规划需要设计能力能强调对这种复杂性的考虑。跨尺度的系统思考方法[36]可在全球及区域尺度下对栖息地连通性及生物多样性保护进行深入解读。生态系统科学研究已显示出多尺度分析的必要性，并将特定尺度下输入的分辨率，以及评估过程考虑在内。在景观规划中，跨尺度的思考能够检验土地系统在场地、地区、区域、国家甚至全球尺度下相互之间的联系和影响(图9)。跨尺度的思考对于从整体上把握场地及其大背景尤为重要，也可以帮助理解不同尺度下的同步作用力，并将这些跨尺度的功能联系在一起，从而更好的、更整体性的把握相互联系的景观过程，从而为设计策略寻找思路。

表2 哈德逊农场新计划林带种植前后的景观连通性指数对比Tab.2 Comparison of landscape connectivity index before and after proposed hedgerow on Hudson Farm

生物多样性及相关生态系统服务十分重要，它们对于人类在这个星球上的可持续发展有着决定性的影响。我们必须始终将此牢记在心，即使是在场地尺度的设计中也应予以贯彻。多环节设计工作的第一步是在更大尺度范围下进行背景分析。从全球尺度来看，哈德逊农场正位于密西西比—美洲(鸟类)飞行迁徙路线上(图10)。哈德逊农场内的森林斑块、林带和湿地应予以悉心照料，场地开发不应使其栖息地发生损毁或退化，而应尽量保持、改善并维护其在全球迁徙路线中的生态功能[37]。图10表明了该场地在国家尺度乃至全球尺度下的生态敏感性。从这一宏观角度出发，林带廊道、林地斑块及湿地栖息地的保护与整合，在场地分析及总体规划阶段的重要性不言而喻。场地层面的景观连通性能反作用于区域或全球尺度下的连通性。

对于阿拉巴马州层面及地区尺度的更深层次分析也揭示了该场地的生态重要性，这意味着设计团队在规划制定过程中进行整合性决策。只有当捍卫各自利益的不同人群达成共识，同时又关注到生物多样性保护带来的共同长期利益时，可持续发展才有可能实现。

本研究以30m的一般距离(缺口小于30m仍视作连接状态)计算连通性。然而，景观连通性评估更需要以物种为中心的研究方法进行[38]。连接结构对某一物种而言为廊道，但对另一物种而言则可能是屏障。同时，栖息地的功能连通性未必需要结构连通性来保障。某些生物体天生具备越隙能力，能够跨越完全不宜栖居或局部不宜栖居的基质以连接资源；而另外一些物种则无法越隙，因而需要更高的结构连通性。基于此，连通性研究需要收集物种运动对景观结构的响应信息(如穿越不同景观要素的运动速度、传播范围、传播期间死亡率和边界互动等)，以及在较大尺度影响作用下，这些响应的变化情况。然而，因以物种为基础的研究极为有限，这类信息往往很难获取。因此，景观尺度下的总体连通性评估只能是根据现有的不同景观要素连通性而展开的宏观概论。

尽管关于生态系统服务及生物多样性保护已有大量研究，但真正的挑战来源于将二者整合于景观规划之中[39-40]。我们需要创新性设计工具对设计干预的影响进行实时评估，从而将生态科学及设计实践的优势更好地结合并应用于可持续社区的设计中。例如，强调生物多样性生态系统服务融于生态重建给人类生计带来的好处，从而获得更多的公众支持。进行生态重建，也可对已处于衰落阶段的社区进行再生性设计。将人类生计同环境管理职责综合考虑后，才有可能通过公众参与，实现和谐的设计成果。新兴的地理设计方法[41-42]及生态智慧理论[43]也为这些问题的解决提供了新的视角。

10 全球迁徙路线。哈德逊农场位于密西西比—美洲迁徙路线上Global migrating flyway. Hudson Farm is located on the Mississippi-Americas flyway

7 结语

连通性是景观生态学和景观设计中的一个重要概念。景观连通性可用多种方式进行量化。FRAGSTATS空间分析模型以GIS数据作为输入层，计算出连通性及许多其他景观参数。在可以获取GIS数据的情况下，这种方法效率较高。通过比较已有场地和计划开发后场地的景观连通性，能够得出二者的显著差异，而这一差异可用于表征项目开发后景观营造的生态影响，从而避免在实际城市开发过程中，对连通性造成负面冲击。事实上，在总体规划制定过程中也可采取一定措施保持并提升景观连通性。鉴于众多大型项目均旨在打造新镇或新城，上述理念可谓影响深远。通过在设计实施前对生态系统服务进行评估，可选择最佳或最优化设计方案以便实施设计干预。

除此之外，对于既定地区的整体生态系统服务，需要仔细评估，包括根据不同景观要素或生态系统内不同物种对这些服务加以区分。要完成这项任务，我们迫切需要更为先进的工具。城市、城市的某些组成部分、几个城市作为一个整体，均可视作“新的生态系统”，而在这些生态系统中，生物多样性及其他生态系统服务的价值不应根据其原先情况来判断[44-45]，而应根据其在不断进行的生态系统演替过程中生成生态系统服务的潜力来评定。因此，对于景观要素或特定物种对生物多样性、人类健康及生态系统服务产生的利弊，必须理清其优先考虑顺序，进而指导设计干预。无论是设计和生态系统间跨尺度的影响－响应环，还是设计塑造景观以及景观改变设计，相互之间均是不断作用、不断增加且共同进化的。正如罗伯特·帕克(Robert E. Park)在80多年前所表述的：“城市和城市环境，是人类遵循自己心意改造其居住的世界所做出过的最持久、也可说是最成功的尝试。但如果说城市是人类创造的世界，那么它也是人类注定要生活的世界。因此，在未能清晰认识这一创造任务本质的情况下，人类在创造城市的同时其实也间接地改造了自己。[46]”

注释：

①图1、3、4、6、8、9为作者自绘；图2为作者拍摄；图5来自： Dramstad 等, 1996；图7来自：FRAGSTATS程序界面；图10来自：联合国环境项目，2006。

②表1来自：FRAGSTATS程序界面；表2为作者自绘。

致谢：

感谢哈德森项目团队Chad Adams，Franklin Collins，Carol Collins，Nick Murray，Nick Koncinja，Frost Rollins，Fitz Hudson，Nan Hudson，Jeff Speck等，没有他们的热情帮助本研究就无法实现。特别感谢Jack Williams，Michael Robinson和Charlene LeBleu教授，他们自项目伊始的鼓励与关切一直陪伴着我完成了这项研究。最特别需要感谢的是我们优秀的合作伙伴Joao Xavier。同时还要感谢内布拉斯加大学林肯分校的Richard Sutton教授，无私地分享了他关于林带的研究。本文的早期版本发表于2010年在巴黎召开的国际地理信息系统大会，在此也向当时对此研究提出问题及给出建议的与会学者们表示感谢。

[1] Merriam G. Connectivity: a fundamental ecological characteristic of landscape pattern[M] // Brandt J, Agger P. (Eds.), Methodology in landscape ecological research and planning: proceedings, 1st seminar, International Association of Landscape Ecology. Denmark: Roskilde University, 1984: 5-15.

[2] Andrén H. Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat: a review[J]. Oikos,1994 (71):355-366.

[3] Taylor P D, Fahrig L, Henein K, Merriam G. Connectivity is a vital element of landscape structure[J]. Oikos, 1993 (3): 571-572.

[4] Millennium Ecosystem Assessment. Ecosystems and human well-being our human planet: Summary for decision-makers[R]. Washington DC: Island Press, 2005.

[5] Mace G M, Norris K, Fitter A H. Biodiversity and ecosystem services: a multilayered relationship[J]. Trends in ecology & evolution, 2012, 27(1): 19-26.

[6] Eitelberg D A, van Vliet J, Doelman J C, et al. Demand for biodiversity protection and carbon storage as drivers of global land change scenarios[J]. Global Environmental Change, 2016, 40: 101-111.

[7] Swift M J, Izac A M N, van Noordwijk M. Biodiversity and ecosystem services in agricultural landscapes—are we asking the right questions?[J]. Agriculture, Ecosystems & Environment, 2004, 104(1): 113-134.

[8] Bai Y, Zhuang C, Ouyang Z, et al. Spatial characteristics between biodiversity and ecosystem services in a humandominated watershed[J]. Ecological Complexity, 2011, 8(2): 177-183.

[9] Forman R T T. Land mosaic[M]. Cambridge: Cambridge University presses, 1995.

[10] Burel F. Hedgerows and their role in agricultural landscapes [J]. Critical reviews in plant sciences, 1996, 15(2): 169-190.

[11] Burel F, Baudry J. Control of biodiversity in hedgerow network landscapes in Western France[J]. Hedgerow management and nature conservation. Ashford: Wye College Press, 1994: 47-57.

[12] Cooper A, McCann T. Northern Ireland Countryside Survey 2000: Field Handbook[M]. Coleraine: University of Ulster, 2000.

[13] Battershill M R J, Gilg A W. Socio-economic constraints and environmentally friendly farming in the Southwest of England[J]. Journal of Rural Studies, 1997, 13(2): 213-228.

[14] Beedell J, Rehman T. Using social-psychology models to understand farmers’ conservation behavior[J]. Journal of rural studies, 2000, 16(1): 117-127.

[15] Barnes G, Williamson T. Hedgerow history: ecology, history and landscape character[M]. Windgather Press, 2008.

[16] Birks P, Kaland E, Moe D. (Eds). The cultural landscape: past, present and future[M]. Cambridge: Cambridge University Press, 1988: 53-77.

[17] Braekevelt A. Evolution of the spatial structure of the hedgerows in the Houtland (NW-Belgium)[C] //Connectivity in landscape ecology. Proceedings of the 2nd International Seminar of the International Association for Landscape Ecology. Münster: Münstersche Geographische Arbeiten, 1988: 153-155.

[18] Watt T A, Buckley G P. Hedgerow management and nature conservation: British Ecological Society Conservation Ecology Group[M]. Wye College, University of London, Wye College Press, 1994: 47-57.

[19] Cronon W. Changes in the land: Indians, colonists, and the ecology of New England[M]. Macmillan, 2011.

[20] Rackham O, Moody J. The making of the Cretan landscape[M]. Manchester University Press, 1996.

[21] Dramstad W, Olson J D, Forman R T T. Landscape ecology principles in landscape architecture and land-use planning[M]. Island press, 1996.

[22] Forman R T T, Godron M. Landscape Ecology[M]. New York: John Wiley & Sons, 1986: 619.

[23] Ryan R L, Walker J T H. Protecting and managing private farmland and public greenways in the urban fringe[J]. Landscape and urban planning, 2004, 68(2): 183-198.

[24] Erickson D L, Ryan R L, De Young R. Woodlots in the rural landscape: landowner motivations and management attitudes in a Michigan (USA) case study[J]. Landscape and Urban Planning, 2002, 58(2): 101-112.

[25] Fabos J G. From parks to greenways into the 21st century[C]// Proceedings from Selected Educational Sessions of the 1991 ASLA Annual Meeting, Kansas City, MO, 1991.

[26] Forman R T T, Baudry J. Hedgerows and hedgerow networks in landscape ecology[J]. Environmental management, 1984, 8(6): 495-510.

[27] Lindhult M S, Fabos J, Brown P, et al. Using geographic information systems to assess conflicts between agriculture and development[J]. Landscape and Urban Planning, 1988, 16(4): 333-343.

[28] Antrop M. Why landscapes of the past are important for the future[J]. Landscape and urban planning, 2005, 70(1): 21-34.

[29] Marshall E J P, Arnold G M. Factors affecting field weed and field margin flora on a farm in Essex, UK[J]. Landscape and Urban Planning, 1995, 31(1): 205-216.

[30] Merriam G, Lanoue A. Corridor use by small mammals: field measurement for three experimental types of Peromyscus leucopus[J]. Landscape ecology, 1990, 4(2): 123-131.

[31] Nabhan G P, Sheridan T E. Living fencerows of the Rio San Miguel, Sonora, Mexico: traditional technology for floodplain management[J]. Human Ecology, 1977, 5(2): 97-111.

[32] Nassauer J. Westmacott R. Progressiveness among farmers as a factor in heterogeneity of farmed landscapes[M]. In: M.G. Turner (eds.), Landscape Heterogeneity and Disturbance. New York: Springer, 1987: 199-210.

[33] Opdam P, van Apeldoom R, Schotman A. and Kalkhoven J. Population responses to landscape fragmentation [M] // Vos C C, Opdam P. (Eds.), Landscape Ecology of a Stressed Environment. Netherlands: Springer, 1993: 147-171.

[34] McGarigal K, Cushman S A, Neel M C, et al. FRAGSTATS: spatial pattern analysis program for categorical maps[Z]. Computer software program produced by the authors at the University of Massachusetts, 2002.

[35] Saunders D A, Hobbs R J. (Eds). The role of Corridors [M]. Surrey Beatty, Chipping Norton. Australia, 1991: 442.

[36] Chen S, Lee V. From metropolis to allotment: Scaled system thinking in advancing landscape studio knowledge[M]//Buhmann E, et al.(Eds.) Systems Thinking in Landscape Planning and Design: Landscape Architecture, 2015, 3: 344-353.

[37] Smith J A. Space Based Ornithology: On the Wings of Migration and Biophysics[R]. Greenbelt, MD, NASA Goddard Space Flight Center. 2005.

[38] Hansen A J, Urban D L. Avian response to landscape pattern: the role of species’ life histories[J]. Landscape Ecology, 1992, 7(3): 163-180.

[39] Daily G C, Polasky S, Goldstein J, et al. Ecosystem services in decision making: time to deliver[J]. Frontiers in Ecology and the Environment, 2009, 7(1): 21-28.

[40] De Groot R S, Alkemade R, Braat L, et al. Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision making[J]. Ecological complexity, 2010, 7(3): 260-272.

[41] Steinitz C. A framework for geodesign: Changing geography by design[M]. Redlands, CA: esri, 2012.

[42] You H. Practice of geodesign approach in macroscale land use planning: scenario-based simulation and evaluation of land use change in Beijing[J]. Landscape Architecture Frontiers, 2013, 1(6): 122-132.

[43] Steiner F. Frontiers in urban ecological design and planning research[J]. Landscape and urban planning, 2014, 125: 304-311.

[44] Oreszczyn S, Lane A. The meaning of hedgerows in the English landscape: different stakeholder perspectives and the implications for future hedge management[J]. Journal of Environmental Management, 2000, 60(1): 101-118.

[45] Standish R J, Hobbs R J, Miller J R. Improving city life: options for ecological restoration in urban landscapes and how these might influence interactions between people and nature[J]. Landscape Ecology, 2013, 28(6): 1213-1221.

[46] Park R E. (1929). The city as a social laboratory[M]// Smith T V, White L D. (Eds.), Chicago: An experiment in social science research. Chicago: University of Chicago Press, 1929: 1-19.

(编辑/张希)

1 Introduction

The concept of 'landscape connectivity' was introduced to emphasize the interaction between species' attributes and landscape structure in determining movement of organisms among habitat patches[1]. Human activities such as agricultural development, commercial afforestation, infrastructure construction and urbanization have led to habitat fragmentation, namely, loss of the original habitat, reduction in habitat patch size and isolation of habitat patches, and decreasing landscape connectivity[2]. Numerous scientific studies continue to ignore key elements of the original concept while many practical land development projects claim efforts are taken to enhance landscape connectivity. However, without understanding the meaning of landscape connectivity, these studies/projects might actually diminish its potential utility for land management and the conservation of biodiversity. As originally defined by[3], landscape connectivity is 'the degree to which the landscape facilitates or impedes movement among resource patches'. This definition emphasizes that the types, amounts and arrangements of habitat or land use on the landscape influence movement and population dynamics and community structure. However, landscape connectivity should combine twofold of meaning: the description of the physical structure of the landscape (structural connectivity) with special species' response to that structure (functional connectivity), which forms the theoretical background of applying landscape connectivity principles in the practices of landscape planning and design. In this study, a GIS-based approach is used to quantify landscape connectivity. Furthermore, a residential development project in the southern United States was used to explore the meaning of landscape connectivity and its application in town planning.

Ecosystem services have been defined as the benefits that people obtain, either directly or indirectly, from various ecosystems, such as benefits provided by the biodiversity of the nature for humans comprising, for example, food, fiber, climate and regulation, pollination and aesthetic values enhancing well-being of various groups of people[4]. Population growth, urbanization and associated exploitation of natural resources have led to the widespread degradation approximately 60% of the world's ecosystem services[4]. The unsustainable nature of the modern life style and the public perception of ecosystem services must be drastically changed in order to reverse this trend. Thus it is crucial to raise public awareness of the value of ecosystem services and goods, and make them understand how ecosystems function to provide these services. Landscape and urban planning practitioners must also recognise that humans, urban or rural, are only one of the species inhabiting the earth, must abandon their attitude that ecosystem services are taken for granted, and engage in activities to preserve ecosystems in order for them to maintain sustainable provision of ecosystem services.

Biodiversity has many definitions and multiple measures. It can be defined as 'the diversity of life on Earth'[4], and it is always regarded as 'a regulator of underpinning ecosystem processes, as a final ecosystem service and as a good'[5]. Biodiversity is arguably the most important ecosystem services. The conservation of biodiversity can be achieved at different spatial scales. At the global or regional level, the 2011 Convention on Biological Diversity established the Aichi biodiversity targets— a set of goals and targets put in place to protect and promote global biodiversity. Target 11 calls for 17% of terrestrial area (not including Antarctica) to be conserved and protected, specifically those areas where biodiversity is threatened. This equals 22.94 million km2, or an area roughly equal to the size of Canada, China, and India combined[6]. The large scale long term vision for biodiversity conservation is important, but the actions to implement biodiversity conservation or protection can only be achieved at local or site scale. At the site level, biodiversity is significantly influenced by land use, such as agricultural development[7]. It is related to four factors that influence habitat quality: the relative impact of each threat, the relative sensitivity of each habitat to each threat, the distance between the sources of threats and the habitats, and the degree to which the land use is legally protected, and threats are sometimes human-dominated landscapes, such as cropland and urban areas[8].In this study, site level design interventions are investigated in relation to habitat connectivity and biodiversity conservation.

2 Case Study Project Site

2.1 Hedgerow

The vast rural landscape in the southern United States is conspicuously characterized by the hedgerow trees or groves. Hedgerow's primary function in the landscape is to serve as limits, marking boundaries and borders. But hedgerows can also provide products for human in his pursue of food, clothes and shelters, among many other ecosystem services it provides in the rural landscape. The improvement of the visual quality, authenticity of the rural landscape is another important function of hedgerow. Many of the functions of hedgerow can be assessed in the relationship one another[9-13]. However, the most important function of hedgerows in biological conservation is that they are important habitat for wildlife such as bird, mouse, butterfly, etc. Meanwhile, hedgerows functioning as ecological corridors maintain and increase the connectivity of the landscape, maintaining the ecological variability, thus protect and improve the biodiversity of the landscape[14-18].

The patchwork landscape of fields surrounded by high hedgerows is a traditional and familiar feature of the American countryside. Hedgerows are in effect linear strips of trees, groves, or woodlands, which are often critical habitats for wildlife and important for the visual quality of the landscape. The Hudson Farm project was used as the case study to demonstrate how landscape connectivity can be quantified. Hudson Farm is located right on the Black Belt, which is a region of the southeastern U.S. Originally the term describes the prairies and dark soil of central Alabama and northeast Mississippi; however, it has long been used to describe a broad region in the American South characterized by a high percentage of African Americans. It is regarded that the Black Belt covers large areas of Central Georgia, North Florida, Western Mississippi, South Central Alabama, East Central Louisiana, Eastern North Carolina and Southeastern Virginia. Hudson Farm is right located on the black belt, a suburb to the southeast of Montgomery, the capital city of the state of Alabama (Fig. 1). The landscape of Hudson Farm is remarkably characterized by the hedgerow trees or groves. The hedgerows form a series of network of patchwork, creating a landscape of low fields surrounded by high corridors in the form of hedgerow trees and groves.

Hedgerows and woodland are important habitat for wildlife such as bird, bee, mouse, butterfly, etc. (Fig. 2)Meanwhile, the woodland functioning as ecological patch and hedgerows as ecological corridors complement each other in maintaining the diversity of habitat type and connectivity in the landscape. The patch-corridor-matrix formed by hedgerows and woodlands are important to protect and promote biodiversity at the site level in this case. As green infrastructure on site, hedgerow corridors and woodland patches not only give a strong sense of place in the rural landscape but also invite an intimate emotional association with the American countryside[19-20]. Therefore, they must be carefully considered and thoughtfully integrated into the land development project– through ecological landscape planning and innovative urban design strategies.

2.2 Topography

The site was remarkably characterized by a ridgeline that divides the property into several sub-watersheds (Fig. 3). The highest point on the site is located at the lower part of the property, which is 310 feet, and the lowest point of the site is located in the upper part of the property, which is 227 feet. The Hudson farm is a slightly rolling pasture and savannah landscape.

The GIS slope analysis indicates that the site is generally flat, large portion of the land has a maximum slope of 7% (Fig. 3). The slope aspect analysis shows that to below the ridgeline, the site is mostly west- or southwest- facing, while on the upper side of the site, most of the land are north- or northeast- facing (Fig. 3). The slope aspect analysis is for considering the sun pattern, the potential of using solar energy as an alternative energy source.

3 Landscape Connectivity

3.1 Structural Connectivity

Common usage of the term 'connectivity' generally emphasizes the structural aspect, where landscape connectivity is simply equated with linear features of the landscape that promote dispersal, such as physically connected linear corridors. This connectivity allows route from A to B. In a network system (Fig. 4), if there are more alternative routes to travel from A to B, then the network is considered more connected, or it has higher connectivity.

3.2 Functional Connectivity

Physical connectivity is measured by the numbers of loops present in the network. However, in landscape ecology, commonly employed measures of connectivity focus not only on physically connected linear corridor, but also on patch area and how inter-patch distances affect movement in between; i.e. corridors not physically connected, for instance, the stepping stones that can be used by certain species as connected corridor (Fig. 5). This can be called as functional connectivity[21].

Width and connectivity are the primary controls on the five major functions of corridors, i.e., habitat, conduit, filter, source, and sink[9,22-23]. The effect of a gap in corridor on movement of a species depends on length of the gap relative to the scale of species movement, and contrast between the corridor and the gap. However, a row of stepping stones (small patches) is intermediate in connectivity between a corridor and no corridor, and hence intermediate in providing for movement of interior species between patches (Fig. 5a). For highly visually-oriented species, the effective distance for movement between stepping stones is determined by the ability to see each successive stepping stone (Fig. 5b). Loss of one small patch, which functions as a stepping stone for movement between other patches, normally inhibits movement and thereby increases patch isolation (Fig. 5c). The optimal spatial arrangement of a cluster of stepping stones between large patches provides alternate or redundant routes, while maintaining an overall linearly-oriented array between the large patches[30]. This structure facilitates wildlife movements as evidenced by many studies[21-22,24-32].

3.3 Habitat Connectivity

Hudson Farm is a family-owned 2,200 acre property at the suburb of Montgomery, Alabama. The land was used primarily for cattle grazing or hay harvesting. Landscape elements such as trees, hollows, hedgerows, barns, and fences serve as unique landmarks in the Montgomery urban-rural interface. A deep knowledge and understanding of the site will serve as the foundation for planning and design. The preservation and enhancement of distinctive landmarks such as trees, hollows, hedgerows, barns, and fences will maintain the site's unique character. The development of the property is to create a new community with pedestrian-oriented, compact, and mixed-use neighborhoods, in contrast to the single-use conventional suburban development. Creating whole neighborhoods and towns, rather than pockets of suburban development, is a vital step towards creating a sustainable development footprint[33].

On Hudson Farm, hedgerows exist in the landscape in the form of corridor (Fig. 6) and are considered important habitat for biodiversity. Before development, Hudson Farm is used for cattle grazing and hay harvesting. The aerial photography of Hudson Farm area shows the landscape structure of the site (Fig. 6a). Hudson farm is surrounded by large stream corridors and wetlands, which provide critical wildlife habitats for migrating birds and other wildlife. The landscape elements that make Hudson Farm unique are big patches of forests, open fields, corridors in the form of hedgerows, large 'landscape rooms' enclosed by high hedgerows. The hedgerows and tress on the farm enhance the sense of place by providing refreshing long views across the open field dotted or enclosed by high hedgerow trees and groves (Fig. 6b).

4 Quantifiying Connectivity at Landscape Scale

There are various approaches to quantifying landscape connectivity. For example, connectivity (con) is calculated through the a simple equation using numbers of linkages and nodes based on Forman’s urban ecological perspective[9]

con = L / 3 (V-2) (1)

Where

L = number of linkages;

V = number of nodes

However, this approach may be easy to use at a very fine scale where landscape linkages nodes are easily identified. At landscape scale, where the site condition is highly diversified and complicated (e.g. with different forms of connectivity), a new method based on GIS is needed to quantify connectivity.

4.1 GIS-based Landscape Metrics

Many GIS-based landscape metrics are developed by landscape ecologist and scientists and provided for public use[34]. FRAGSTATS is one of these applications designed to compute a wide variety of landscape metrics (including landscape connectivity) for categorical map patterns. This program is developed by the Landscape Ecology Lab at the University of Massachusetts. The original software (version 2) was released in the public domain during 1995 in association with the publication of a USDA Forest Service General Technical Report[5]. This study used the version 3.3 in calculation, which is available for download at the lab's website. FRAGSTATS calculate connectivity based on connectivity metrics (Tab. 1).

For example, the Connectivity Index (con) is calculated as

Where

Cijk= joining between patch j and k (0 = unjoined, 1 = joined) of the same patch type, based on a user-specified threshold distance.

ni= number of patches in the landscape of each patch type i.

In this matrix, connectivity equals the number of functional joining between all patches of the same patch type (sum of Cijkwhere Cijk= 0 if patch j and k are not within the specified distance of each other and Cijk= 1 if patch j and k are within the specified distance), divided by the total number of possible joining between all patches of the same type, multiplied by 100 to convert to a percentage. Therefore the connectivity ranges between 0 and 100. Connectivity = 0 when either the landscape consists of a single patch, or all classes consist of a single patch, or none of the patches in the landscape are connected (i.e., within the userspecified threshold distance of another patch of the same type). Connectivity = 100 when every patch in the landscape is connected[34].

4.2 Input Data and Model Parameterization

The FRAGSTATS spatial pattern analysis program requires GIS data to be prepared in a recognizable file as input to the program to calculate the landscape metrics using FRAGSTATS. The program has to be properly parameterized before it can be run to produce the output statistics (Fig. 7). Details about the FRAGSTATS program, input data preparation, model parameterization, and output data format are available at the program's website.

4.3 Case Study: Hudson Farm

Maintaining networks of corridors is a principle in the design of functional and healthy landscape so as to allow wildlife movement through the landscape and enhance biodiversity. During the master plan making process of the Hudson project, hedgerows are connected, with a continuous tree cover. This concept is generally advocated by many landscape ecologists[35]. This network is superimposed on the ditch network and based on the existing hedgerows. Its spatial arrangement is related to historical factors, such as landlord-worker relationship[36]. Based on this idea, the existing hedgerows are studied (Fig. 8).

To increase connectivity, new hedgerows are proposed to connect the existing hedgerow to create a hedgerow network. The network is meant to be green infrastructure network or ecological infrastructure network which will have both designed functions and ecological functions. Basic ideas are to show why they should be connected and how should they be connected. The left diagram (Fig. 8a) shows the existing hedgerow, the diagram in the middle (Fig. 8b), the red color area, shows the proposed hedgerows. The diagram on the right (Fig. 8c) shows the overlay of existing hedgerow and the proposed hedgerow.

5 Results

Through integrating existing green infrastructure with newly proposed hedgerow corridors, the landscape connectivity has been improved 40% (Tab. 2), according to results from the FRASTATS spatial patter analysis program. The proposed hedgerows are located in existing farm land boundaries, abandoned farm facility sites, and across grazing field to chain up existing remnant trees and groves to form connected hedgerow networks. Therefore the 40% increase of connectivity is achieved without losing much of the core habitat in the form of current large landscape patches. The proposed hedgerows reconnect the broken corridors and increase the overall landscape connectivity dramatically. This indicates the importance of maintaining the intactness of the landscape and its natural vegetation corridor.

The analysis demonstrated that FRAGSTATS can be combined with GIS to calculate connectivity and other parameters at landscape scale. However, the results in Table 2 must be interpreted with caution. For instance, connectivity is measure in 'CONNECT' and 'COHENSION' (Tab. 1-2); even the overall 'CONNCET' as measured use Equation (2) is increased by 40.7%, however, the landscape 'COHENSION’ does not increase as the same magnitude. On the contrary, it decreases slightly[34]. This example illustrates that design interventions can change the ecology of the site; therefore, design intervention must be informed by appropriate evaluation of the consequences following the implementation of the design. This is becoming increasingly easier thanks to the advancement of digital technologies.

6 Discussion

The conservation of biodiversity and related ecosystem services are a global challenge. It needs systematic approaches that work across time and spatial scales. Biodiversity planning requires design capability that addresses these complexities. A scaled system thinking approach is used to further interpret habitat connectivity and biodiversity conservation at global and regional scales[36]. Ecosystem science research has demonstrated the need for multi-scale analysis that takes into consideration the scale and resolution of inputs, and of the processes being evaluated. In landscape planning, scale thinking is to examine interactions in the land systems across site level, local level, regional level, national level, and even global level (Fig. 9). Scale thinking is essential to gain a holistic perspective of the site and its broader context, understand of these very forces that are simultaneously functioning at other scales, and relate these cross-scale functions to gain a holistic view of boarder interrelated landscape processes and their potential influences on the design solution.

Biodiversity and related ecosystem services are important issues that have critical influence on the human sustainability on this planet. This notion must always be kept in mind even when designing at the site scale. The first step in these chained design efforts is to conduct context analysis and broader scales. Hudson Farm is right located on the Mississippi Americas Flyway (Fig.10). The forest patches, hedgerows and wetlands within Hudson Farm should receive careful consideration where development should not eliminate or degrade these habitats but maintain or improve them in order to keep its ecological function in the global flyway[37]. The image serves as strong arguments that the site is ecologically sensitive, not only at the national scale, but also at the global level. With these bigpicture images in mind, it is easily understood why the conservation and integration of hedgerow corridors, woodland patches, and wetland habitats are so important in the site analysis and master planning phases. Landscape connectivity at site level forms augmented connectivity at regional or global scale.

Further analysis at the Alabama state level and local scale also reveals the ecological significance of the site, which requires the design team to exercise integrated decision making in the plan-making process. Sustainable development is only possible when consensus is reached among different groups defending their own interests without neglecting the commonlong-term benefits of biodiversity conservation and preservation.

In this study, a general distance of 30m (a gap below this distance is still considered connected) is used to calculate the connectivity. However, assessing landscape connectivity requires a species-centered approach[38]. A connected structure may serve as a corridor for one species, but a barrier for another. Meanwhile, habitat does not necessarily need to be structurally connected to be functionally connected. Some organisms, by virtue of their gap-crossing abilities, are capable of linking resources across an uninhabitable or partially inhabitable matrix, while other species cannot cross gaps therefore requires higher structure connectivity. Therefore, the study of connectivity requires information on species' movement responses to landscape structure (e.g., movement rates through different landscape elements, dispersal range, mortality during dispersal, boundary interactions, etc.) and how those responses differ as a function of broader-scale influences. Such information is typically quite difficult to obtain, as very limited study is carried out on a species to species basis. Therefore, the assessment of the overall connectivity at landscape scale is but a big-picture overview of the connectivity of different landscape elements present.

Despite numerous studies on ecosystem services and biodiversity conservation, challenges still remain in integrating ecosystem services and biodiversity conservation into landscape planning[39-40]. New innovative design tools that enable real-time assessment of the impact of design intervention are needed so that the strength of ecological sciences and design practices can be combined for designing sustainable community. For example, the integration of biodiversity ecosystem services into ecological restoration offers an opportunity to enhance public support by emphasizing its benefits to human livelihood. The approach of ecological restoration may take form of the regenerative design of a community that had been undergoing a decay process. A harmonious design outcome may be achieved through engaging the public considering both human livelihood and environmental stewardship. The emerging geodesign approach[41-42]and ecowisdom theories[43]provide new insights to these questions.

7 Conclusion Remarks

Connectivity is an important concept in landscape ecology and landscape architecture. Landscape connectivity can be measured in different ways. FRAGSTATS spatial analysis model uses use GIS data as input layers to calculate connectivity along with many other landscape parameters. This is efficient when GIS data are available. Significant difference when comparing the landscape connectivity of the existing site with that of proposed development can be easily used to assess the impact of the modified landscape after proposed development, thus negative impacts on connectivity can be avoid in real urban development projects. Instead, measures can be taken to maintain and improve landscape connectivity during the master plan making process. The impact of this notion is profound considering the nature of the large projects alike aiming at creating new towns or cities. Pre-implementation ecosystem services evaluation can be done to inform design intervention through optimization or selecting the most capable design solution.

In addition, the overall ecosystem services for a given region should be carefully evaluated, including differentiating the services by different landscape components, or different species in the ecosystems. More advanced tools capable of completing this task are highly demanded. The sustainability of city of some components of cities or even some cities as a whole may be viewed as "novel ecosystems" in which the value of biodiversity and other ecosystem services should not be judged by its origins[44-45]but by its potential to produce ecosystem services in the ongoing process of ecosystem succession. Therefore, priorities around whether alandscape component or certain species are producing benefits or harm to biodiversity, human health, and ecosystem services must be organised and used to guide design intervention. The influence-response loop between design and ecosystem across scales, or the interplay between shaping landscape through design and design being shaped by the landscape, is mutual, incremental, and coevolutionary. This is similar to what Robert E. Park said more than 80 years ago: "For the city and the urban environment represent man's most consistent and, on the whole, his most successful attempt to remake the world he lives in more after his heart's desire. But if the city is the world which man created, it is the world in which he is henceforth condemned to live. Thus, indirectly, and without any clear sense of the nature of his task, in making the city man has remade himself"[46].

Acknowledgment

The author wish to thank the Hudson project team for their kind help that makes this study possible: Chad Adams, Franklin and Carol Collins, Nick Murray, Nick Koncinja, Frost Rollins, Fitz Hudson, Nan Hudson, Jeff Speck, etc. Special thanks go to Professors Jack Williams, Michael Robinson and Charlene LeBleu whose encouraging words from the very beginning of the project have longlasting effects on this work. The most special thanks go to Joao Xavier, thanks very much for being a nice working partner. The author also wishes to thank Professor Richard Sutton in the University of Nebraska-Lincoln for sharing with the author his research and publications on hedgerow. An early version of this paper has been presented at the 2012 International Conference in GIS in Paris and the author wish to thank the conference participants who asked questions or offered comments on this study.

Integrating Landscape Connectivity into Town Planning for Biodiversity Ecosystem Service Provision

Biodiversity is one of the most important ecosystem services in the wake of increasing extinction rate of endangered species worldwide. Biodiversity conservation is critical in maintaining the viability of ecosystems at local, regional, and global scales. Landscape connectivity combines a description of the physical structure of the landscape with special species' response to that structure, which forms the theoretical background of applying landscape connectivity principles in landscape planning practices, particularly for these environmentally progressive projects that integrate provision for ecosystem services. In this paper, an eco-town development project in the southeastern United States was used as a case study to explore design considerations that promote landscape connectivity and facilitate biodiversity conservation. Based on geographic information system (GIS) and spatial statistical analysis (FRAGSTATS), this study attempts to quantify the landscape connectivity characterized by woodlands and hedgerows in southeastern United States where substantial areas with natural landscape are being urbanized due to the ever expanding real estate industry and high demand for new residential development. Results suggest that adding new hedgerows to the existing green infrastructure on site can significantly increase landscape connectivity thus improve ecosystem services due to the increase of habitat size and increased connectivity among habitat patches. In conclusion, this study shed lights on how to balance the needs of new urban development and eco-services provision by maintaining a higher level of landscape connectivity, thus will inform the design intervention aimed at achieving not only livability but also sustainability.

landscape architecture; landscape connectivity; ecosystem services; biodiversity conservation; landscape planning

TU986

A

1673-1530(2017)01-0066-16

10.14085/j.fjyl.2017.01.0066.16

2016-12-16

上海市浦江人才计划(15PJC092)；国家自然青年科学基金项目(51508391)

Fund Items: Swppooted by Shanghai Pujiang Program(Grant No. 15PJC092); National Natural Science Foundation (Youth Foundation) (Grant No. 51508391) are acknowledged.

(澳)陈思清/1973年生/男/湖北人/澳大利亚墨尔本大学设计学院高级讲师/博士生导师/澳大利亚TDG地理设计工作室创始人 / 美国奥本大学景观建筑硕士/中国科学院地理信息系统博士/主要从事生态城景观规划、地理设计、弹性城市和绿色基础设施等方面的教学、研究和实践工作(VIC 3010) Authors：

CHEN Siqing was born in Hubei in 1973. He is a Senior Lecturer, doctoral supervisor in Melbourne School of Design, The University of Melbourne, Australia. He is the founding director of the TGD GeoDesign Studio in Australia. He holds a Master of Landscape Architecture from Auburn University and a PhD of GIS from Chinese Academy of Sciences. His research, teaching and practice areas includes ecotown landscape planning, geodesign, resilient city and green infrastructure (VIC 3010).汪洁琼/1981年生/女/上海人/同济大学建筑与城市规划学院景观学系/同济大学建筑与城市规划学院生态智慧与城乡生态实践研究中心/高密度人居环境生态与节能教育部重点实验室/讲师、系主任助理(研究生教学)、硕士生导师、澳大利亚墨尔本大学博士、助教/研究方向为生态系统服务与空间形态机制、水生态、绿色基础设施等领域的教学、科研和工程实践(上海200092)

WANG Jie-qiong was born in Shanghai in 1981. She is a lecturer, postgraduate supervisor and also a Department Director Assistant in the College of Architecture and Urban Planning, Tongji University. She holds a PhD from The University of Melbourne and was a tutor and guest lecturer there. As a research member of the Center for Ecological Wisdom and Practice Research and the Key Laboratory of Ecology and Energy-saving Study of Dense Habitat (Tongji University), her research focuses on the ecosystem services and physical form, water ecology, green infrastructure (Shanghai 200092).译者简介：

王南/1984 年生/女/江苏南京人/同济大学环境科学与工程学院在读博士后/同济大学建筑与城市规划学院景观学系专业博士/研究方向为景观规划与设计理论与方法(上海200092) Translator：

WANG Nan was born in 1984 in Nanjing, Jiangsu

Province. She is a post-doctoral candidate of College of Environmental Science and Engineering, Tongji University. She holds a PhD of Landscape Architecture from College of Architecture and Urban Planning, Tongji University. Her research focuses on theories and methodology of landscape planning and design (Shanghai 200092).

修回日期：2017-01-23