Insights into the function of rivers and watersheds often emerge when we quantify riverine solute fluxes. Many flux estimation methods are available, and the best method for each dataset may depend on the solute of interest, the land use and hydrology of the watershed, and the site position within the river network. Here we present a new R software package called loadflex that implements several prominent methods for flux estimation, including regressions, period-weighted methods, and a recently developed approach called the composite method. To demonstrate, we use loadflex to quickly analyze data from the Lamprey River, New Hampshire, where two large floods in 2006-7 are hypothesized to have driven a long-term shift in nitrate fluxes. Several competing estimation methods each give believable flux estimates, and yet (1) they yield different answers for whether and how the floods altered nitrate fluxes, and (2) the best method differs for main-stem versus tributary sites. Our R package simplifies the process of comparing flux estimation methods and drawing conclusions such as these, ultimately allowing researchers to estimate riverine fluxes with greater ease and accuracy.
Appling A.P., Leon M.C., McDowell W.H. (2015): Optimizing watershed flux estimates: the R package ‘loadflex’ . Society for Fresh Water Science.