Prediction of Impending Drought Scenarios Based on Surface and Subsurface Parameters in a Selected Region of Tropical Queensland, Australia

Droughts occur in all climatic regions around the world costing a large expense to global economies. Reasonably accurate prediction of drought events helps water managers proper planning for utilization of limited water resources and distribution of available waters to different sectors and avoid catastrophic consequences. Therefore, a means to create a simplistic approach for forecasting drought conditions with easily accessible parameters is highly desirable. This study proposes and evaluates newly developed accurate prediction models utilizing various hydrologic, meteorological, and geohydrology parameters along with the use of Artificial Neural Network (ANN) models with various forecast lead times. The present study develops a multitude of forecasting models to predict drought indices such as the Standard Precipitation Index with a lead-time of up to 6 months, and the Soil Moisture Index with a lead-time of 3 months. Furthermore, prediction models with the capability of approximating surface and groundwater storage levels including the Ross River Dam level have been developed with relatively high accuracy with a lead-time of 3 months. The results obtained from these models were compared to current values, revealing that ANN based approach can be used as a simple and effective predictive model that can be utilized for prediction of different aspects of drought scenarios in a typical study area like Townsville, North Queensland, Australia which had suffered severe recent drought conditions for almost six recent years (2014 to early 2019).

Abundance and habitat-suitability relationship deteriorate in fragmented forest landscapes:a case of Adinandra griffithii Dyer,a threatened endemic tree from Meghalaya in northeast India

Introduction:A strong positive‘abundance and habitat-suitability’relationship is crucial for conservation of species.Nevertheless,anthropogenic alteration of natural landscapes leading to land use and land cover change,habitat loss,and species extinctions(may)have putatively disturbed this relationship.Hence,it is important to study the nature of the relationship in such human influenced landscapes.Methods:In this study,we endeavored to understand the consistency of the relationship in the fragmented natural landscapes in the Khasi,Garo,and Jaintia hills of Meghalaya in northeast India,with Adinandra griffithii(an endangered endemic tree)as a model species.We reconstructed the distribution of its suitable habitats as a function of the remotely sensed vegetation phenology(i.e.,EVI data),using point occurrence data and ecological niche modeling(ENM)tool.Estimation of the abundance and habitat characterization was done through field surveys following standard methods.Results:The study revealed that remotely sensed landscape-level vegetation phenology could effectively discriminate the suitable and unsuitable habitats of threatened species.Linear regression model showed a weak positive correlation between abundance and predicted habitat suitability for adult trees indicating(plausible)deterioration in the relationship.However,sapling and seedling populations did not show a precise trend in this respect.Field-based studies revealed that removal of the species from the suitable habitats because of anthropogenic disturbances possibly weakened the abundance-suitability relationship.Conclusions:The findings of the study enjoin the need for re-establishment of the species in the suitable areas for its conservation and perpetuation.