Subscribe to RSS Feed

Tuesday, September 15th
5:20 PM

Downscaling and Testing of Remote-Sensing Derived Evapo-Transpiration Products in Agricultural Landscapes

James Moloney, The University of Sydney, Sydney Institute of Agriculture

5:20 PM - 5:40 PM

Evapo-transpiration (ET) presents a key component of managing water, not only for resources at the catchment scale, but also within agricultural landscapes. As almost two thirds of precipitation over land originates as ET, knowledge of ET provides insight into both rainfall patterns and plant water use, which in turn can inform land managers on biomass production and potential yields. Whilst there is an increasing array of spatial products available globally for use in monitoring ET (MODIS at 500m, METRIC at 30m etc.) they provide a suite of issues for fine resolution agricultural management. These problems include limited consideration of changes in land cover, which happens frequently within cropping scenarios due to rotations, and the validation of the products relies on stationary comparison sites at flux towers, many of which are located within natural landscapes, which are not representative of agricultural conditions. In this work we focus on MODIS ET and present a method for downscaling, incorporating both crop and soil information that can provide estimates of ET useful for within field management. We evaluate the ET product using multiple lines of evidence, including temporally using capacitance soil moisture probes, and spatially through crop yield monitor data and METRIC ET Values calculated from Landsat. We test the approach using a case study in south-eastern Australia, the Muttama creek catchment, a 1025 km2 sub catchment within the Murray-Darling basin in eastern Australia with approximately 620 mm of annual rainfall and predominately dryland cropping and grazing land use.

Thursday, September 17th
1:20 PM

Application of a Hybrid Model in Assessing Northern Peatlands Vulnerability for Various Climate Change Scenarios

Georgii Alexandrov, IAPRAS, Russia

1:20 PM - 1:40 PM

The long periods of dry weather reduce the groundwater level and make northern peatlands vulnerable to peatfires. Increasing frequency of such periods may jeopardise normal functioning of these ecosystems under climate change. Here I apply a hybrid model to quantify the area of northern peatlands that become more vulnerable to peatfires under a given climate change scenario. The hybrid model provides an analytical solution of partial differential equations representing an impeded drainage model. The impeded drainage model describes groundwater movement within a watershed partly covered by peatlands that makes it possible to simulate the groundwater level within the watershed for given geomorphological conditions and climatic conditions projected by the Earth System Models. Applying the hybrid models to global fields of monthly temperature and precipitation produced as the part of CMIP6 (sixth phase of the climate model intercomparison project) results in an ensemble-based assessment of the risk imposed by climate change.

3:00 PM

Developing Training Sets for Hydrological Prediction Based on Supervised Classification

Marina Erechtchoukova, York University, Canada

3:00 PM - 3:20 PM

In support of watershed management decisions, trustworthy and efficient predictive models are required. Classification machine learning algorithms representing a data-driven approach to hydrological model development had become increasingly popular in recent years. The specific of their utilization, however, is in the necessity to develop a reliable model on data from the recent past which will be applied to data collected in the future. Incorporation of generated predictions in management decisions is only possible after the assessment of their reliability, traditionally associated with the estimates of the model generalization error. Since a model is created by training (e.g., calibration of) a machine learning algorithm on a representative data set, the issue of transforming the spatially and temporally dispersed data into training and testing sets becomes critical for developing a predictive tool. Just evaluating the model performance on data samples unseen on the training step is insufficient for model validation. This study was focused on investigation of the approaches to building training sets which ensure models reliable for short-term prediction of flood events for extended lead time intervals in a watershed with a flashy response to precipitation. The original computational scheme was based on the framework incorporating time-delay embedding applied to time-series data from all observation sites of a watershed. Stratified random sampling was compared with chronological splits of complete data sets and stratified data reflected distinct classes of hydrological events. While stratified random sampling provides average estimates of model performance, practical needs dictate the necessity to choose a model with better performance on events occurring outside of the time interval used for training the model. The computational experiments were conducted on data collected during years with different hydrological characteristics. The results of this study are presented in this paper.