Most of the hydrological models developed in the last decades, and ordinarily used for decision support in water system, focus on the natural component in the water cycle whilst human activities are predominantly regarded as external forces that marginally impact the hydrological system. This is, however, contradict to the presence of human signature in many river basins worldwide which are prevalent nowadays, and whose increasing impacts are rivaling with the natural forces themselves in conditioning the natural processes and transforming the hydrosphere. The feedback between the human system and the natural ones also imply such process a complex co-evolutionary one, with the projected non-stationary climate change further exacerbating our ability to predict the future conditions of these systems. Therefore, there is a need to shift from traditional human-excluded modeling practices to a more integrated approach, where the human related entities have to be included in the modeling framework to address their internal feedback and the joint dynamics under changing conditions. Recent studies in water systems analysis are increasingly advocate such holistic view by treating the human and natural subsystems as a whole, the so-called Coupled Human and Natural Systems (CHNS), where characterizing and modeling co-evolution of CHNS is key to building reliable medium-to-long term projections and, ultimately, to designing management and adaptation strategies to mitigate water stress and crises under climate change context. To this end, we developed a decision-analytic framework, namely the DistriLake framework, to analyze and characterize the co-evolution of coupled human water systems (descriptive component), and to design and assess alternative management strategies (prescriptive component). The objective is to build a mathematical model of the co-evolving CHNS under changing climate and socio-economic conditions. The DistriLake integrates traditional hydrological process equations for the natural component, and behavioral models to characterize the human decision-making processes and their effects on the water cycle. Different climate and socio-economic scenarios will be used as boundary conditions. The behavioral modeling of stakeholders' decision-making process is the key component to approach the projection of such co-evolutionary process. The modeling framework is essentially based on Multi Agent Systems (MAS), which are state-of-the-art tools to characterize heterogeneous human agents, where agents' behaviors are modeled using normative approach assuming rational agents. The proposed modeling framework was applied on Lake Como water system located in north Italy as a pilot study area, and also as a representative coupled human water system. In the first application, the framework was shown to successfully capture current situation, and the system may, however, be exposed under a considerable risk under projected drought scenarios. Results show that timely co-adaptation to such changing conditions will be valuable to mitigate the negative impact from climate changes. In the second study, we applied the DistriLake framework to investigate the robustness of current system in face of deep uncertain scenarios. Results highlight the potential multi-stakeholders' conflicts implicit to traditional robustness based assessments, often with presumption of social-planner's view. The last study adopted the proposed modeling framework to assess the optional value of long-term climate forecast products in supporting farmers' crop planning decisions. Results show that integrating the end-users' decision making process into the assessment framework may provide more insightful thoughts on the poor adoption of the long-term forecast, given the current status of forecast quality.
La maggior parte dei modelli idrologici sviluppati negli ultimi decenni, e di solito utilizzati per il supporto decisionale nel sistema idrologico, si concentrano sulla componente naturale nel ciclo dell'acqua, mentre le attività umane sono prevalentemente considerato come disturbi esterni che influiscono marginalmente il sistema idrologico. Questo è, tuttavia, contraddice alla presenza di firma umano in molti bacini fluviali che sono prevalenti al giorno d'oggi, e il cui impatto crescente sono rivaleggiando con le forze naturali stessi per condizionare i processi naturali e trasformare l'idrosfera. Il feedback tra il sistema umano e naturali implicano anche tale processo un complesso co-evoluzione, con il cambiamento climatico non stazionari ulteriore esacerbando la nostra capacità di prevedere le future condizioni di questi sistemi. Pertanto, vi è la necessità di passare da pratiche di modellazione tradizionale umano-escluso ad una più approccio integrato, in cui le entità correlate umani devono essere incluso nella schema di modellazione per affrontare la loro feedback interno e le dinamiche congiunte in condizioni mutevoli. Recenti ricerche in analisi sistema idrico sono sempre sostengono tale visione olistica trattando i sottosistemi umani e naturali nel suo complesso, il cosiddetto Coupled Human Nature Systems (CHNS), dove la caratterizzazione e modellazione co-evoluzione di CHNS è la chiave per costruire affidabili proiezioni a medio-lungo termine e, in ultima analisi, a la progettazione di strategie di gestione e di adattamento per mitigare lo stress idrico e crisi sotto contesto del cambiamento climatico. Per questa ragione, abbiamo sviluppato un modello analitico-decisionale, cioè il DistriLake, per analizzare e caratterizzare la co-evoluzione dei sistema idrico umani accoppiati (componente descrittiva), e di progettare e valutare strategie di gestione alternativa (componente prescrittiva). L'obiettivo è quello di costruire un modello matematico delle CHNS co-evoluzione sotto cambiamento del clima e delle condizioni socio-economiche. Il DistriLake integra le equazioni di processo idrologica tradizionali per la componente naturale, e modelli di comportamento per caratterizzare i processi decisionali umani e dei loro effetti sul ciclo dell'acqua. Diversi scenari del climatico e socio-economici saranno utilizzati come condizioni al contorno. La modellazione del comportamento del processo decisionale delle parti interessate è il componente fondamentale di avvicinarsi alla proiezione di tale processo co-evolutivo. La schema di modellazione si basa essenzialmente sui Sistemi Multi-Agente (MAS), che sono state-of-the-art strumenti per caratterizzare agenti umani eterogenei, in cui i comportamenti degli agenti vengono modellati utilizzando approccio normativo assumendo agenti razionali. Il modello proposta è stata applicata sul sistema idrico Lago di Como si trova nel nord Italia come area di studio pilota, e anche come rappresentante CHNS. Nella prima applicazione, il framework è stato mostrato per catturare correttamente situazione attuale, e il sistema può, tuttavia, essere esposto sotto un rischio considerevole in scenari siccità proiettate. I risultati mostrano che puntuale co-adattamento a tali condizioni mutevoli sarà prezioso per mitigare l'impatto negativo dei cambiamenti climatici. Nel secondo studio, abbiamo applicato il DistriLake framework di indagare la robustezza del sistema attuale di fronte a scenari incerti profonde. I risultati mettono in evidenza i conflitti potenziali fra parti interessate implicito alle valutazioni tradizionali a base di robustezza, spesso con la presunzione di vista socio-planner. L'ultimo studio ha adottato il framework proposto per valutare il valore opzionale di prodotti di previsione del clima a lungo termine nel sostenere decisioni di pianificazione delle colture degli agricoltori. I risultati mostrano che l'integrazione di processo decisionale rendendo il utenti finali nella schema di valutazione può fornire pensieri più penetranti sui poveri adozione del lungo periodo di previsione, dato lo stato attuale della qualità del tempo.
Advancing coupled human-water systems analysis by agent-based modeling
LI, YU
Abstract
Most of the hydrological models developed in the last decades, and ordinarily used for decision support in water system, focus on the natural component in the water cycle whilst human activities are predominantly regarded as external forces that marginally impact the hydrological system. This is, however, contradict to the presence of human signature in many river basins worldwide which are prevalent nowadays, and whose increasing impacts are rivaling with the natural forces themselves in conditioning the natural processes and transforming the hydrosphere. The feedback between the human system and the natural ones also imply such process a complex co-evolutionary one, with the projected non-stationary climate change further exacerbating our ability to predict the future conditions of these systems. Therefore, there is a need to shift from traditional human-excluded modeling practices to a more integrated approach, where the human related entities have to be included in the modeling framework to address their internal feedback and the joint dynamics under changing conditions. Recent studies in water systems analysis are increasingly advocate such holistic view by treating the human and natural subsystems as a whole, the so-called Coupled Human and Natural Systems (CHNS), where characterizing and modeling co-evolution of CHNS is key to building reliable medium-to-long term projections and, ultimately, to designing management and adaptation strategies to mitigate water stress and crises under climate change context. To this end, we developed a decision-analytic framework, namely the DistriLake framework, to analyze and characterize the co-evolution of coupled human water systems (descriptive component), and to design and assess alternative management strategies (prescriptive component). The objective is to build a mathematical model of the co-evolving CHNS under changing climate and socio-economic conditions. The DistriLake integrates traditional hydrological process equations for the natural component, and behavioral models to characterize the human decision-making processes and their effects on the water cycle. Different climate and socio-economic scenarios will be used as boundary conditions. The behavioral modeling of stakeholders' decision-making process is the key component to approach the projection of such co-evolutionary process. The modeling framework is essentially based on Multi Agent Systems (MAS), which are state-of-the-art tools to characterize heterogeneous human agents, where agents' behaviors are modeled using normative approach assuming rational agents. The proposed modeling framework was applied on Lake Como water system located in north Italy as a pilot study area, and also as a representative coupled human water system. In the first application, the framework was shown to successfully capture current situation, and the system may, however, be exposed under a considerable risk under projected drought scenarios. Results show that timely co-adaptation to such changing conditions will be valuable to mitigate the negative impact from climate changes. In the second study, we applied the DistriLake framework to investigate the robustness of current system in face of deep uncertain scenarios. Results highlight the potential multi-stakeholders' conflicts implicit to traditional robustness based assessments, often with presumption of social-planner's view. The last study adopted the proposed modeling framework to assess the optional value of long-term climate forecast products in supporting farmers' crop planning decisions. Results show that integrating the end-users' decision making process into the assessment framework may provide more insightful thoughts on the poor adoption of the long-term forecast, given the current status of forecast quality.File | Dimensione | Formato | |
---|---|---|---|
Disertation_main.pdf
accessibile in internet per tutti
Descrizione: Dissertation on CHNS by Yu Li
Dimensione
13.2 MB
Formato
Adobe PDF
|
13.2 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/122634