**8. Modeling phosphorus transport to water bodies and phosphorus index**

According to the conceptual model and hillslope hydrology, it is vital to identify significant source locations, which are areas with disproportionately high P-losses, to efficiently manage P-movement regularly at the field and watershed scales. Soil and water assessment tools like the Soil and Water Assessment Tool [100] have been developed to identify critical source locations of P-migration (SWAT).

While various developments in modeling, including graphical user interfaces and geographic information system (GIS) layers, have been made in the field, the fundamental methodologies to modeling P-movement have remained essentially unchanged [101]. Land use, soil texture, and topography are the primary inputs for most P-transport models, but other factors like management techniques are also widely used as data. There are three preliminary modeling approaches: processed-based, export-coefficient, and statistical models [102].

For the original P-index, Lemunyon and Gilbert [103] set out to assess the risk of P-transport to water bodies, identify the essential components that drive P-loss, and assist in selecting management measures that reduce P-loss. It has evolved significantly since its introduction in 1993 [103] from being a critical source area identification to now serving as best management practice selectors in manure application scheduling tools, manure application rate calculators, and regulatory mechanisms of some states in the United States of America [104]. In the United States and Europe, P-indices have included as many as 34 variables from each location [105]. Runoff class, soil erosion, irrigation erosion, soil P-test, and P-fertilizer application rate and technique, organic P-source application rate, and method of organic P-source application were the initial eight parameters to evaluate in the original P-index, which was updated in 2010.

Some site factors may be more important than others in influencing P-migration from the site. The weighting factor for each site attribute was determined by professional judgment and historical experience. To get a weighted score for a site characteristic, the P-loss rating value is first multiplied by the site characteristic weighing factor. P-loss is less likely in fields with low P-indices, while P-loss is more likely in areas with high P-indices.
