PyEt - a Python package to estimate potential and reference evapotranspiration

Evapotranspiration (ET) is a major component of the hydrological cycle and accurate estimates of the flux are important to the water and agricultural sector, among others. Due to difficulties in the direct observation of ET in the field, the flux is often estimated from other meteorological data using empirical formulas. There is a wide variety of such formulas, with different levels of input data and parameter requirements. While some Python packages are available in the Python ecosystem for these tasks, they typically focus on one specific formula or data type. The goal of PyEt is to provide a Python package for the estimation of ET that works with many different data types, is well documented and tested, and simple to use. The source code is hosted at GitHub (https://github.com/phydrus/PyEt) and Pypi can be used to install the package. PyEt currently contains nine different methods to estimate ET and various methods to estimate surface and aerodynamic resistance. The methods are tested against other open source data to ensure proper functioning of the methods. While the methods currently are only implemented for 1D data (e.g. time series data), future work will focus on enabling the methods on 2D and 3D data as well (such as Numpy Arrays, XArray, and NetCDF files). The package allows hydrologists to compute and compare evapotranspiration estimates using different approaches with minimum effort. The presentation will focus on the problems associated with reproducibility in ET estimation and linkage with existing Python libraries to perform complex sensitivity and uncertainty analyses.