Conference Presentations

Based on over 70 site-years of Ameriflux sites observations, water, whether in atmosphere (VPD), soil, stream, or water table plays in modulating carbon flux variation, despite a general overall lack of water-limitation in the biome.

A new ensemble modeling framework based on artificial neural network nonlinear autoregressive networks was developed to predict spatio-temporal fluctuations of groundwater in the densely irrigated area of Wisconsin central sands (WCS).

Evapotranspiration (ET) values were modeled from the flight data using the High-Resolution Mapping of Evapotranspiration (HRMET) model. HRMET-derived ET maps are compared to ET estimates from an Eddy Covariance system within the flight domain. Ongoing results will be used to develop best practices and assess trade-offs for integrating RPAs as decision support tools for irrigation and water management.

The data comparison between potatoes and pine evapotranspiration (ET) shows that when there is a big rainfall or irrigation event, the differences between already available ET measurement based on remote sensing models and actual ET measurements are small. However, when soil is dry, ET measurement from Eddy covariance flux tower is more accurate.

For a wet year, pines ET is higher than potatoes. In dry years when pine ET is more moisture limited, pine evapotranspiration could be lower, according to previous studies.

An algorithm is developed that can accurately predict and forecast farm-scale regional daily out to 3 days. Daily ET forecast (3 days) model based on random forest (RF) has R2 and RMSE of 0.72 mm and 0.76mm respectively while recurrent neural network (RNN) ensemble forecast model was able to forecast 3 days ET with R2 and RMSE of 0.71 mm and 0.78 mm respectively.

Remotely piloted aircraft (RPA) was used to generate high-resolution maps of crop water stress using remotely sensed thermal and multi-spectral RPA imagery in the Central Sands region of Wisconsin.

We can better quantify evapotranspiration (ET) to improve our understanding of the Central Sands water cycle and improve irrigation demand forecasting.

A new ensemble modeling framework based on artificial neural network nonlinear autoregressive networks was developed to predict spatio-temporal fluctuations of groundwater in the densely irrigated area of Wisconsin central sands (WCS).

To better understand the historical and potential hydrological impacts of pine-to-potato conversion across the northern Great Lakes region, a process-based AgroIBIS-VSF model was developed using lysimetry, eddy covariance, phenological, and physiological data from several sites in Wisconsin and Minnesota.