Adaptation and transformation planning for resilient social-ecological system in coastal wetland using spatial-temporal simulation.

Coastal wetlands comprise unique ecological systems such as tidal flats and wetlands coexisting with marine and terrestrial ecosystems. The Songdo wetlands in South Korea are adjacent to the Yellow Sea, and were once composed mainly of tidal flats, but as urbanization progressed, their social-ecological system changed. The social system created by land reclamation and development reduced the migratory bird population and the tidal flat area, damaging the ecological system. This study suggests adaptation and transformation plans by analyzing land use change and fragmentation of the Songdo wetlands using spatial-temporal simulation. System dynamics and GIS were used in the process of analyzing land use change through spatial-temporal simulation, and FRAGSTATS was used in the analysis of wetland fragmentation. Scenario 1 (current state maintenance) presents adaptation plans to increase the connectivity of wetland patches, since fragmentation has not progressed to the extent of wetland system collapse. In Scenario 2 (development acceleration), since the wetland system causes serious fragmentation in terms of area and shape, we propose transformation plans such as disaster response to the collapse of the ecological system and qualitative improvement of wildlife habitat. In Scenario 3 (wetland restoration), proposes transformation plans from the network and modularization perspective in response to quantitative restoration and morphological fragmentation of wetlands. The adaptation and transformation plans presented in this study can provide prediction results suitable for various contingencies such as the current state, acceleration of development, and wetlands restoration. This study is also meaningful in that it proposes plans from the perspective of resilience by predicting the change of the Songdo area, which is scheduled to be developed by 2030.

Simulation of Land Use Pattern Evolution from a Multi-Scenario Perspective: A Case Study of Suzhou City in Anhui Province, China.

Land use/land cover change is a frontier issue in the field of geography research. Taking Suzhou City in Anhui Province as the research case, based on thematic mapper /enhanced thematic mapper+ (TM/ETM+) remote sensing data from 1998 to 2018, through the transfer matrix model and modified conversion of land use and its effects at small region extent (CLUE-S) model, the simulation of the land use landscape pattern evolution was studied from a multi-scenario perspective. The results showed that in the past 20 years, landscape patterns have undergone spatial-temporal conversion, which was mainly manifested as the evolution from a cultivated land landscape and other agricultural land to construction land, and there was some transformation between other landscape types, but the transformation degree was not significant. The spatial autocorrelation factor was introduced to correct the CLUE-S model, and the Kappa index reached 0.83, indicating that the modified CLUE-S model had a good simulation accuracy. (I) In the cultivated land protection scenario, limiting the conversion of basic farmland use, and by 2028, the proportion of cultivated land increased by 5.23%, distributed in eastern Suzhou City; (II) in the economic development scenario, by 2028, the construction land area increased by 14.58%, and was distributed in the surrounding regions of the built-up areas; and (III) in the ecological protection scenario, by 2028, wood land, water, and other ecological protection land area increased, and were distributed in the central and eastern part of Suzhou City. Research can provide useful decision-making support for land use optimization and remediation.

SynTEG: a framework for temporal structured electronic health data simulation.

OBJECTIVE: Simulating electronic health record data offers an opportunity to resolve the tension between data sharing and patient privacy. Recent techniques based on generative adversarial networks have shown promise but neglect the temporal aspect of healthcare. We introduce a generative framework for simulating the trajectory of patients' diagnoses and measures to evaluate utility and privacy. MATERIALS AND METHODS: The framework simulates date-stamped diagnosis sequences based on a 2-stage process that 1) sequentially extracts temporal patterns from clinical visits and 2) generates synthetic data conditioned on the learned patterns. We designed 3 utility measures to characterize the extent to which the framework maintains feature correlations and temporal patterns in clinical events. We evaluated the framework with billing codes, represented as phenome-wide association study codes (phecodes), from over 500 000 Vanderbilt University Medical Center electronic health records. We further assessed the privacy risks based on membership inference and attribute disclosure attacks. RESULTS: The simulated temporal sequences exhibited similar characteristics to real sequences on the utility measures. Notably, diagnosis prediction models based on real versus synthetic temporal data exhibited an average relative difference in area under the ROC curve of 1.6% with standard deviation of 3.8% for 1276 phecodes. Additionally, the relative difference in the mean occurrence age and time between visits were 4.9% and 4.2%, respectively. The privacy risks in synthetic data, with respect to the membership and attribute inference were negligible. CONCLUSION: This investigation indicates that temporal diagnosis code sequences can be simulated in a manner that provides utility and respects privacy.

Testing for top-down cascading effects in a biomass-driven ecological network of soil invertebrates.

To investigate the structural changes of a food-web architecture, we considered real data coming from a soil food web in one abandoned pasture with former low-pressure agriculture management and we reproduced the corresponding ecological network within a multi-agent fully programmable modeling environment in order to simulate dynamically the cascading effects due to the removal of entire functional guilds.We performed several simulations differing from each other for the functional implications. At the first trophic level, we simulated a removal of the prey, that is, herbivores and microbivores, while at the second trophic level, we simulated a removal of the predators, that is, omnivores and carnivores. The five main guilds were removed either separately or in combination.The alteration in the food-web architecture induced by the removal of entire functional guilds was the highest when the entire second trophic level was removed, while the removal of all microbivores caused an alteration in the food-web structure of less than 5% of the total changes due to the removal of opportunistic and predatory species.Omnivores alone account for the highest shifts in time of the numerical abundances of the remaining species, providing computational evidence of the importance of the degree of omnivory in the stabilization of soil biota.

Evidence-based economic analysis demonstrates that ecosystem service benefits of water hyacinth management greatly exceed research and control costs.

Invasive species management can be a victim of its own success when decades of effective control cause memories of past harm to fade and raise questions of whether programs should continue. Economic analysis can be used to assess the efficiency of investing in invasive species control by comparing ecosystem service benefits to program costs, but only if appropriate data exist. We used a case study of water hyacinth (Eichhornia crassipes (Mart.) Solms), a nuisance floating aquatic plant, in Louisiana to demonstrate how comprehensive record-keeping supports economic analysis. Using long-term data sets, we developed empirical and spatio-temporal simulation models of intermediate complexity to project invasive species growth for control and no-control scenarios. For Louisiana, we estimated that peak plant cover would be 76% higher without the substantial growth rate suppression (84% reduction) that appeared due primarily to biological control agents. Our economic analysis revealed that combined biological and herbicide control programs, monitored over an unusually long time period (1975-2013), generated a benefit-cost ratio of about 34:1 derived from the relatively modest costs of $124 million ($2013) compared to the $4.2 billion ($2013) in benefits to anglers, waterfowl hunters, boating-dependent businesses, and water treatment facilities over the 38-year analysis period. This work adds to the literature by: (1) providing evidence of the effectiveness of water hyacinth biological control; (2) demonstrating use of parsimonious spatio-temporal models to estimate benefits of invasive species control; and (3) incorporating activity substitution into economic benefit transfer to avoid overstating benefits. Our study suggests that robust and cost-effective economic analysis is enabled by good record keeping and generalizable models that can demonstrate management effectiveness and promote social efficiency of invasive species control.