تغییر کاربری زمین های شهری: نقش برنامه ریزی فضایی استراتژیک

Urban land-use change: The role of strategic spatial planning

Abstract

To date land-change science has devoted little attention to spatial policy and planning in urban landscapes despite the widely accepted premise that planning affects urban land change. This is primarily due to lack of relevant data and an underdeveloped theoretical understanding regarding the impact of spatial planning on urban land change. To be able to better analyse the role of spatial planning in urban development we need to distinguish: 1) the intentions expressed in the plans; 2) the means of implementation of the plans through governance processes and 3) the role of external conditions influencing implementation. Based on a synthesis of the current literature on how spatial planning is implemented in land-change models, and drawing from the literature on planning evaluation, we sketch a research agenda to further develop the understanding of these three components and their interconnections as well as their application in quantitative land-change modelling approaches for urban regions.

1 Introduction

Land change is one of the key processes of global environmental change (Magliocca et al., 2015; Turner II et al., 2007; Verburg et al., 2015). The studies on the topic have gradually advanced from a focus on patterns of land-use and land-cover change to an analysis of dynamic interactions within socio-ecological systems and the resulting impacts on, for example, ecosystem services and biodiversity (Rindfuss et al., 2004; Rounsevell et al., 2012). In this context, land change is understood as the result of interacting political/institutional, economic, cultural, technological and natural/spatial driving forces and the respective actors (Bürgi et al., 2004; Hersperger et al., 2010). Whereas data on economic and natural conditions have a long tradition in being used to explain land changes, researchers only recently started to pay attention to policies, plans, and regulations on land use, within their specific institutional and governmental contexts. Meta-analytical studies have emphasized the role of land-use policies and spatial planning as a major underlying driving factor for many different land-use change processes (van Vliet et al., 2016).

Compared with forest and agricultural related research, studies on land change in urban regions are so far a small part of land-system research (Geist et al., 2006; Magliocca et al., 2015; Seto et al., 2011). However, urban regions, which are also broadly defined as cities or metropolitan regions, are some of the most dynamic land-change systems worldwide. With strong further urbanization expected over the coming decades they will cover increasing areas of the earth surface and host the majority of the human population (Seto et al., 2012). At the same time, urban land change is not restricted to the core city, but includes many new urban-rural spaces functionally tied to the city (Brenner and Schmid, 2015) and has many impacts on rural hinterlands (Bren d’Amour et al., 2016) and hence deserves more attention in landchange science (Müller and Munroe, 2014).

A widely accepted premise is that, especially in urban regions, spatial planning – a multifaceted activity with many purposes, including project planning, master planning, land-use planning and strategic planning – influences patterns of land use and land cover (Couclelis, 2005). Amongst the many purposes of spatial planning, spatial planners and governments have been trying to steer urbanization processes with the aim of developing sustainable cities and regions (Albrechts et al., 2017; Collier et al., 2013). However, conceptualizing the role of planning in guiding land change is a great challenge (McNeill et al., 2014). This is partly due to the fact that research on the contribution of planning to land change is at the interface of two paradigms (Briassoulis, 2008; Hillier, 2007): planning scholars tend to stress contextuality and social construction of space; whereas land-change scientists incline to assume the existence of realities that can be objectively described and measured and lend themselves to generalizations. The difference between the two paradigms can be illustrated through how they position themselves in respect to the concepts of space and uncertainty. Many urban geographers and planning theoreticians conceive space as a social construct (Harvey, 2006; Lefebvre, 1991) and expect future urban development to be open ended. Uncertainty is inherent; to be expected and subject to exogenous factors (Graham and Healey, 1999). Land-change scientists quantify and model space by identifying correlation or causality between drivers and outcomes, and strive to identify sources and levels of uncertainty in landcover projections (Alexander et al., 2016). Research to bridge the two paradigms is sparse. Consequently, planning is not well integrated in quantitative land-change assessments.

پیش بینی رتبه جامعه – توزیع فراوانی تحت شرایط آب و هوایی فعلی و آینده

Predicting community rank‐abundance distributions under current and future climates

Understanding influences of environmental change on biodiversity requires consideration of more than just species richness. Here we present a novel framework for understanding possible changes in species’ abundance structures within communities under climate change. We demonstrate this using comprehensive survey and environmental data from 1748 woody plant communities across southeast Queensland, Australia, to model rank-abundance distributions (RADs) under current and future climates. Under current conditions, the models predicted RADs consistent with the region’s dominant vegetation types. We demonstrate that under a business as usual climate scenario, total abundance and richness may decline in subtropical rainforest and shrubby heath, and increase in dry sclerophyll forests. Despite these opposing trends, we predicted evenness in the distribution of abundances between species to increase in all vegetation types. By assessing the information rich, multidimensional RAD, we show that climate-driven changes to community abundance structures will likely vary depending on the current composition and environmental context.

1 Introduction

Understanding the drivers of diversity remains a key challenge in ecology, and is becoming increasingly important as organisms respond to rapid climate change (Parmesan et al. 1999, Walther et al. 2005). Almost all projections of diversity, both spatial and temporal, have focused on species richness derived from species occurrence data. For example, spatial models of richness across broad geographic extents have been developed for birds (Davies et al. 2007, Rahbek et al. 2007, Distler et al. 2015), reptiles (Lewin et al. 2016, Rodrigues et al. 2017), invertebrates (Lobo et al. 2002, Overton et al. 2009) and plants (Steinmann et al. 2009, Cramer and Verboom 2017, Fricker et al. 2015), with projections also made under future climates (Currie 2001, Thuiller et al. 2005). However, it has been suggested that richness alone is a poor proxy for diversity (Hurlbert 1971) as it fails to describe the abundance patterns that define ecological communities (Stirling and Wilsey 2001, Wilsey et al. 2005). Models of species richness rarely acknowledge that richness is intrinsically linked to, and constrained by, overall community abundance – there cannot be more species than the total number of individuals. While this is unlikely to be an issue in models using large grain sizes, it is relevant to data collected at plot scales (e.g. sub-hectare forest plots) where abundances of some species can be low, and adding or removing a relatively small number of individuals may greatly increase the chance of altering richness (Dornelas et al. 2011). Focusing on richness alone also fails to consider one of the universal laws of ecology: that virtually all communities contain many uncommon and few common species (Tokeshi 1993, McGill et al. 2007). A number of indices (i.e. Simpson’s and Shannon’s diversity) have been developed to compare community structure among sites, however these are known to oversimplify communitylevel abundance patterns and it has been recommended that full community-level abundance patterns are used when comparing across sites (see Matthews and Whittaker 2015, and references therein). Understanding the relative abundances of species (e.g. community evenness) provides additional useful information relating to important ecological processes, such as competition (Whittaker 1965, Rajaniemi 2011). However, due to influences from important environmental variables (e.g. fire, drought, storms) – sometimes with complicated interactions – species abundances can be difficult to predict accurately. A continued focus on richness alone may mask important changes to species abundance distributions under future climates. For example, even if richness remains relatively constant, the abundance of many species might nevertheless shift dramatically, with consequences for longer-term population persistence (Ehrlén and Morris 2015). The rank-abundance distribution (RAD), where all species in a community are ranked from most to least abundant, provides a more holistic representation of communities than richness alone (Whittaker 1965). Rank abundance distributions can be used to describe any community where the abundance of each species is recorded and can be summarised using three simple components: 1) the total number of individuals within a community (abundance), 2) the total number of species in the community (richness) and 3) the allocation of individuals among species (evenness). These three RAD components can be modelled sequentially as functions of environmental covariates (Foster and Dunstan 2010) allowing for spatial predictions that respect their inherent conditionality. The concept of the RAD has featured prominently in the literature since the 1940s (Fisher et al. 1943, Preston 1948), and a steady-state non-linear distribution of abundances within a community is a fundamental assumption of neutral theory (Hubbell 2001). There has been some concerted effort to examine the form of these distributions in more detail across large regions and environmental gradients (Dunstan and Foster 2011, Arellano et al. 2017), however their consideration in biogeography is lacking and there have been specific calls for greater consideration of RADs in largescale analyses (Matthews et al. 2017). Despite the enormous potential of abundance distributions to reveal new insights about the drivers of multiple community attributes, and for predicting and interpreting community responses to environmental change (Fig. 1), RADs have not yet been spatially projected across any terrestrial system, nor have temporal changes in RADs been predicted under future climates. In this study, we use an extensive woody plant community dataset from southeast Queensland, Australia, to project the three RAD components – abundance, richness and evenness – under current and future climates at a fine spatial resolution across this large and diverse region (vegetation ranging from subtropical rainforest to sclerophyll woodlands and shrublands). Given the region’s climate is expected to become warmer and drier in the future (Dowdy et al. 2015), our overall hypothesis is that substantial changes in RADs will be predicted under future climates, but that the nature of these changes will vary with environmental context. Specifically, in response to climate change we expect areas currently dominated by open sclerophyll forests and woodlands to experience ‘woody thickening’ (Bowman et al. 2001, Russell-Smith et al. 2004, Macinnis-Ng et al. 2011), resulting in increased overall abundance and richness (e.g. Fig. 1a).