A developed fuzzy-stochastic optimization for coordinating human activity and eco-environmental protection in a regional wetland ecosystem under uncertainties

Abstract

In this study, a developed fuzzy approximation mixed stochastic approach (DFAS) is proposed for a regional wetland ecosystem (RWE) management under uncertainty. DFAS can handle traditional objective non-determinacy (caused by natural element) and anthropogenic uncertainty (caused by artificial factor) expressed as probability distribution and fuzzy set in objective function or constraint; it also extend to reflect compromise of risk attitude/preference of the decision maker in decision-making process through introducing rough set theory (RST) and measure Me. The proposed approach can be applied to a practical RWE management of Yongnianwa wetland, located in north of China, where the natural ecosystem has been suffered severe degradation induced by disharmonious developing speeds between human activities and environment. Results of ecological effects of wetland ecosystem, water allocation patterns, pollution-mitigation schemes, and system benefit analysis can be acquired. The results indicate that wetland ecosystem can produce a numbers of positive effects to the pollution control and environmental protection, where the total excess pollution discharges (concluding TN, TP and BOD) would reduce 202 and 242 tone (LAV and UAV) at highest. Meanwhile, it finds that wetland system method deemed as an effective/appropriate technology can remove 67%, 72% and 88% TN, TP and BOD from wastewater, where water quality standard of effluent would be II, II and III for TN, TP and BOD at best. However, competitive relationships between water consumption from human activity and wetland protection can facilitate decision makers adjusting current water-environment policies with a more efficient/sustainable manner. Meanwhile, tradeoffs between economic benefit and system-failure risk under optimistic/pessimistic option can support generating a robust plan associated with risk control for RWE under uncertainties. All of these detections can avail local decision makers to generate a plan integrating socio-economic development and eco-environmental protection sustainably.

Introduction

In recent years, an eco-crisis due to ecosystem degeneration and human activity expansion has been of concern to many researchers and managers worldwide. Numbers of projects or ecological engineer techniques for recovering ecological functions have been carried out, which is perceived to produce significant effects on eco-crisis in the context of accelerated industrialization, urbanization, population growth today (Mensing et al., 1998). In general, numerous ecological engineer techniques can combine economic development and eco-environmental protection based on self-design and self-organization principle to intend a sustainable ecosystem for the benefit of both human society and natural environment, which can apply to river basin, agriculture, forest, coastal region and so on (Odum and Odum, 2003, Mitsch and Jørgensen, 2003, Mi et al., 2015, Aronson, 2016). All of these techniques (e.g., withdrawing cultivation, human activity restriction, ecological compensation mechanism, ecological irrigation district, wetland reserve contribution and adjustment of industry structure) can bring a synthetical effect on resolving many contradictions between human activities and eco-environment protections (You et al., 2014). Among them, wetland reserve contribution/protection can be a crucial ecological technique, which can promote ecological functions and economic values (such as flood control, aquifer replenishment, sediment retention, and water filtration) of wetland ecosystem to achieve integrity of soc-economic development and eco-environmental sustainability (Mensing et al., 1998, Millennium Ecosystems Assessment, 2005, Qu et al., 2011). However, in numbers of developing countries such as China, wetland protection has not been paid attention by policymakers yet due to various anthropogenic factors, which would result in ecological functions destruction, even leading adverse effects on human-public health, improvement of life standards and socioeconomic development (Li et al., 2011a, Li et al., 2011b, Zeng et al., 2014). Therefore, an effective planning/management strategy such as water-environment management plan can be advanced in wetland ecosystem, which can not only balance water competitions between various water users (e.g., human activity and ecology), but also facilitates decision makers adjusting current polices to obtain a contemplate and efficient plan for socio-economic development and eco-environmental sustainability (You et al., 2014).

However, in a regional wetland ecosystem (RWE) plan, numbers of impact factors such as land use pattern, biogeochemical/hydrological cycle and social-economic development would influence planning processes (Trepel and Palmeri, 2002, Lambin et al., 2003, Cai et al., 2009a, Tan et al., 2013, You et al., 2014). Meanwhile, a variety of uncertainties (e.g., random of stream flow, imprecision of economic data, uncertain diffusion and migration process of pollution, vague ecological mechanism, adjustment of natural policy and risk preference of decision maker), and their interactions may fortify the conflict laden issues of planning and management (Li et al., 2011b). All above issues require decision makers to contemplate a more comprehensive plan, with considering numbers of impact factors (e.g., economic, social and environmental), uncertainties and their interactions in a RWE plan.

In the past decades, many research works have been done to plan a wetland ecosystem (WE) through balancing contradictions between anthropogenic modification and ecosystem protection (Trepel and Palmeri, 2002, Dale and Polasky, 2007, Altunkaynak and Sen, 2007, Cai et al., 2009b; Sanon et al., 2012, Vidal-Legaz et al., 2013, Dorau et al., 2015, Yang et al., 2015, Zhou and Huang, 2011, Tan et al., 2015, Zeng et al., 2016); meanwhile numbers of efforts were proposed for supporting ecosystem planning system under uncertainties (Tsakiris and Spiliotis, 2004, Cai et al., 2011, Wang et al., 2011, Li et al., 2011b, Zhou et al., 2013, Huang et al., 2013, Dong et al., 2015). Among them, a majority of methods have been used for tackling the inherent uncertainty of objective (i.e., inartificial) conditions. Stochastic mathematical programming (SMP) approaches are advanced to tackle spatial and temporal random uncertainties (e.g., random stream flow) represented as chances or probabilities in an ecosystem planning problem (Calatrava, 2005, Zeng et al., 2014). Fuzzy programming (FP) can handle uncertainties due to anthropogenic errors in acquired data (e.g., imprecise economic data due to limited available data), which can be expressed as fuzzy sets (Nazemi et al., 2002, Lee and Chang, 2005). For example, Maqsood et al. (2005) have mixed fuzzy programming, interval-parameter programming and two-stage stochastic programming into a framework to plan water resources/ecological environment management under uncertainty. Altunkaynak and Sen (2007) have introduced fuzzy membership functions to evaluate the dynamic valuation of ecosystem services in the Lake Van, eastern Turkey. Zeng et al. (2015) have developed a fuzzy-quadratic water management model to plan regional sustainability of floodplain (wetland) ecosystem under uncertainties. However, in an WE planning system, another anthropogenic (subjective) uncertainties (fuzziness) such as risk attitudes/preferences are often affected by decision makers’ experiences and personality traits, which can not expressed as confidence levels primitively, leading conventional FP into dilemma. Therefore, an advanced measure Me is introduced handle such fuzziness, which can reflect the interactions between risk attitudes/preferences and expectations of objective functions in the processes of decision-making due to uncertain features with optimistic-pessimistic options. In general, based on optimal expected value for the objective function, measure Me can evaluate the degree that a fuzzy variable takes the values in an interval with different optimistic-pessimistic attitudes (Xu and Zhou, 2013). Nevertheless, since various format of fuzziness (concluding types of inartificial and anthropogenic fuzziness) exist in a realistic decision-making process synchronously, a certain feasible region (i.e., an interval with different optimistic-pessimistic attitudes) can not reflect all the possible situations completely (Xu and Zhou, 2013). For the sake of simulating the feasible regions more succinctly, rough sets theory (RST) is introduced to handle mixed vagueness presented in the human classification mechanism, where fuzzy decision-making models can be transformed into two approximation models to generate more accurate feasible regions and accessible results for decision makers (Pawlak, 1998, Yin and Wang, 2008). In the past decades, few works were concentrated on the SMP, FP, measure Me and RST into a framework to tackle multiple uncertainties in a hybrid format for RWE planning.

Therefore, a developed fuzzy approximation mixed stochastic method (DFAS) is proposed for a RWE planning and management under uncertainties, incorporating SMP, FP, measure Me and RST into a general frame. DFAS method cannot only provide an effective linkage between conflicting economic benefits and the associated penalties attributed to the violations of the pre-regulated policies, but also tackle various types of fuzziness (objective and subjective) in objective functions and constraints synchronously. Meanwhile, uncertainties in risk attitudes/risk preferences of decision makers can be handled by DFAS method to generate a feasible/accessible compromise between optimistic and pessimistic options. The developed method will be applied to a real case study of Yongnianwa wetland ecosystem (WE) of north China, where the vulnerable ecosystem has encountered environmental crisis (e.g., water deficit, land deterioration, soil erosion and water pollution) due to high-speed economic development, increased population growth, accelerated exploration and cultivation. Adverse impacts from human activities would facilitate decision makers adjusting current ecosystem management patterns and strategies; meanwhile, a variety of ecological protection approaches are established to remit contradictions between economic development and ecological protection. Satisfaction degrees for constraints and optimistic-pessimistic adjusting factors for objective functions can provide more robust supports for planning RWE under uncertainties with aim to sustainability of wetland ecosystem.