Risk of fire occurrence in arid and semi-arid ecosystems of Iran: an investigation using Bayesian belief networks

Abstract Background: Fire occurrence may lead to a significant impactin many terrestrial ecosystems. This study attempted to evaluate the effects of fire on the water balance components in the Central Zagros, Iran. The study used two modeling frameworks, including WetSpass-M and Bayesian Belief Networks to investigate the effect of fire on the amount of runoff, groundwater recharge and evapotranspiration. The first part of the study was a water balance simulation at a monthly scale. In addition, a Bayesian belief networks was applied to explore and understand key issues affect in the water balance after fire. Calibration and validation of the WetSpass-M model was performed without considering the effect of fire (2000-2014) and then the model was run again to with the fire scenario by reducing manning roughness coefficient and increasing the θ coefficient. Results: Calibration and validation were performed before finalizing the simulation. A Nash-Sutcliff coefficient of 0.61 and 0.58 was obtained during the calibration and validation respectively. The analysis of the water balance components results depicted that fire has increased the amount of runoff and it has reduced the amount of groundwater recharge and actual evaporation especially in the sparse forest and poor, medium and good rangelands. Conclusions: Water balance components of each class, i.e. sparse forest, poor, medium and good rangelands were different under fire/non-fire scenarios. The percentage of change in the water balance components were used for comparison. The results of Bayesian model for post-fire scenario showed a significant increase in runoff due to reduced vegetation in the area. Both simulated groundwater recharge and surface flow have showed a reduction rate in the fire occurrence scenario. A similar conclusion was obtained from probabilistic Bayesian model due to reducing vegetation cover and surface changes. Actual evapotranspiration component for the poor rangeland has dropped off significantly. Therefore, there is a need for monitoring hydrologic dynamics of the lands with a high risk of burning.

Download Full-text

Path Associative Networks and Bayesian Belief Networks A career decision support tool

PsycEXTRA Dataset ◽

10.1037/e683322011-113 ◽

2000 ◽

Author(s):

Clarence C. Rohrbaugh

Keyword(s):

Decision Support ◽

Decision Support Tool ◽

Career Decision ◽

Bayesian Belief Networks ◽

Belief Networks ◽

Associative Networks ◽

Support Tool

Download Full-text

Faculty Opinions recommendation of Carbon cycle. The dominant role of semi-arid ecosystems in the trend and variability of the land CO₂ sink.

Faculty Opinions – Post-Publication Peer Review of the Biomedical Literature ◽

10.3410/f.725509014.793523368 ◽

2016 ◽

Author(s):

Dennis Baldocchi

Keyword(s):

Carbon Cycle ◽

Dominant Role ◽

Arid Ecosystems ◽

Semi Arid

Download Full-text

Using Bayesian network models to incorporate uncertainty in the economic analysis of pollution abatement measures under the water framework directive

Water Science & Technology Water Supply ◽

10.2166/ws.2005.0054 ◽

2005 ◽

Vol 5 (6) ◽

pp. 95-104 ◽

Cited By ~ 4

Author(s):

D.N. Barton ◽

T. Saloranta ◽

T.H. Bakken ◽

A. Lyche Solheim ◽

J. Moe ◽

...

Keyword(s):

Water Quality ◽

At Risk ◽

Economic Valuation ◽

Pollution Abatement ◽

Network Models ◽

Cost Effective ◽

Bayesian Belief Networks ◽

Probabilistic Interpretation ◽

Belief Networks ◽

Benefit Cost Analysis

The evaluation of water bodies “at risk” of not achieving the Water Framework Directive's (WFD) goal of “good status” begs the question of how big a risk is acceptable before a programme of measures should be implemented. Documentation of expert judgement and statistical uncertainty in pollution budgets and water quality modelling, combined with Monte Carlo simulation and Bayesian belief networks, make it possible to give a probabilistic interpretation of “at risk”. Combined with information on abatement costs, a cost-effective ranking of measures based on expected costs and effect can be undertaken. Combined with economic valuation of water quality, the definition of “disproportionate cost” of abatement measures compared to benefits of achieving “good status” can also be given a probabilistic interpretation. Explicit modelling of uncertainty helps visualize where research and consulting efforts are most critical for reducing uncertainty. Based on data from the Morsa catchment in South-Eastern Norway, this paper discusses the relative merits of using Bayesian belief networks when integrating biophysical modelling results in the benefit-cost analysis of derogations and cost-effectiveness ranking of abatement measures under the WFD.

Download Full-text

Using Bayesian Belief Networks to assess the influence of landscape connectivity on ecosystem service trade-offs and synergies in urban landscapes in the UK

Landscape Ecology ◽

10.1007/s10980-021-01307-6 ◽

2021 ◽

Author(s):

James D. Karimi ◽

Jim A. Harris ◽

Ron Corstanje

Keyword(s):

Functional Connectivity ◽

Ecosystem Service ◽

Urban Areas ◽

Landscape Connectivity ◽

Structural Characteristics ◽

Bird Species ◽

Bayesian Belief Networks ◽

Belief Networks ◽

Patch Area ◽

Trade Offs

Abstract Context Landscape connectivity is assumed to influence ecosystem service (ES) trade-offs and synergies. However, empirical studies of the effect of landscape connectivity on ES trade-offs and synergies are limited, especially in urban areas where the interactions between patterns and processes are complex. Objectives The objectives of this study were to use a Bayesian Belief Network approach to (1) assess whether functional connectivity drives ES trade-offs and synergies in urban areas and (2) assess the influence of connectivity on the supply of ESs. Methods We used circuit theory to model urban bird flow of P. major and C. caeruleus at a 2 m spatial resolution in Bedford, Luton and Milton Keynes, UK, and Bayesian Belief Networks (BBNs) to assess the sensitivity of ES trade-offs and synergies model outputs to landscape and patch structural characteristics (patch area, connectivity and bird species abundance). Results We found that functional connectivity was the most influential variable in determining two of three ES trade-offs and synergies. Patch area and connectivity exerted a strong influence on ES trade-offs and synergies. Low patch area and low to moderately low connectivity were associated with high levels of ES trade-offs and synergies. Conclusions This study demonstrates that landscape connectivity is an influential determinant of ES trade-offs and synergies and supports the conviction that larger and better-connected habitat patches increase ES provision. A BBN approach is proposed as a feasible method of ES trade-off and synergy prediction in complex landscapes. Our findings can prove to be informative for urban ES management.

Download Full-text