Identification of drought in Dhalai river watershed using MCDM and ANN models

Abstract

An innovative approach for drought identification is developed using Multi-Criteria Decision Making (MCDM) and Artificial Neural Network (ANN) models from surveyed drought parameter data around the Dhalai river watershed in Tripura hinterlands, India. Total eight drought parameters, i.e., precipitation, soil moisture, evapotranspiration, vegetation canopy, cropping pattern, temperature, cultivated land, and groundwater level were obtained from expert, literature and cultivator survey. Then, the Analytic Hierarchy Process (AHP) and Analytic Network Process (ANP) were used for weighting of parameters and Drought Index Identification (DII). Field data of weighted parameters in the meso scale Dhalai River watershed were collected and used to train the ANN model. The developed ANN model was used in the same watershed for identification of drought. Results indicate that the Limited-Memory Quasi-Newton algorithm was better than the commonly used training method. Results obtained from the ANN model shows the drought index developed from the study area ranges from 0.32 to 0.72. Overall analysis revealed that, with appropriate training, the ANN model can be used in the areas where the model is calibrated, or other areas where the range of input parameters is similar to the calibrated region for drought identification.

Unraveling water monitoring association towards weather attributes for response proportions data: A unit-Lindley learning

“…Variables integrated in the current study were previously chosen by other authors in the study of fog/humidity distribution. [6] developed a method for calculating diurnal patterns of air temperature, wind speed, global radiation and relative humidity, and validated it with data from different countries. Some other studies, as [41], have also demonstrated the relevance of scheduling appropriately the sampling frequency of climatic variables, in order to adequately estimate land surface fluxes.…”

“…In the case of rates and proportions processes, whereas the observed variable assumes values in the range (0, 1), there is a well-represented class of models, the unit distributions family, which deals with this type of sensor data, but are often univariate and not extended, in their inference, to some regression structures (see, e.g., [2][3][4][5]). Regression structures, in probabilistic modeling, can provide a flexible set of tools for examining such associations, while enabling, either, potentially confounding effects of other factors, or interaction effects for Statistical Process Control (SPC) tools [6].…”

Using multi criteria decision analysis in a geographical information system framework to assess drought risk

“…Very similarly, Wijitkosum (2018) , for the same watershed, used Fuzzy AHP- GIS to evaluate agricultural drought, the research indicated that the major factors causing droughts in the watershed were related to soil texture, fertility and salinity. Prior to this Aher et al. (2017) used Multi-Criteria Decision Making (MCDM) and Artificial Neural Network models to identify drought using eight drought parameters; precipitation, soil moisture, evapotranspiration, vegetation canopy, cropping pattern, temperature, cultivated land, and groundwater level.…”

Review of the Literature on Multicriteria Methods Applied in the Field of Public Security

“…The literature indicates that the MCDA methods are used and applied to solve decision problems in many areas, as [2,11,19]: information and communication technologies [22]; business intelligence [23]; environmental risk analysis [24,25]; environmental impact assessment and environmental sciences [26]; water resources management [27][28][29]; solid waste management [30,31]; remote sensing [32,33]; flood risk management [34,35]; health technology assessment [36,37]; health care [38]; transportation [39]; nanotechnology research [40]; climate change [41]; energy [42,43]; international politics and laws [44]; Human resources [45]; Investment decisions [46]; performance and benchmarking [47]; Supplier selection [48]; E-commerce and m-commerce [49]; Agriculture and horticulture [50]; Chemical and biochemical engineering [51]; Software evaluation [52]; Network selection [53]; Policy, social and education [54]; HVAC systems and small scale energy management [55].…”