RUSLE2 has evolved from a series of previous erosion prediction technologies.  The USLE was entirely an empirically based equation and was limited in its application to conditions where experimental data were available for deriving factor values.
A major advancement in RUSLE1 was the use of subfactor relationships to compute C factor values from basis features of cover-management systems, which allowed RUSLE2 to be applied tofar more conditions than the USLE.  While RUSLE1 retained the basic structure of the USLE, process-based relationships were added where empirical data and relationships were inadequate, such as computing the effect of strip cropping for modern conservation tillage systems.
RUSLE2 is another major advancement over RUSLE2.  While RUSLE2 uses the USLE basic formulation of the unit plot, the mathematics of RUSLE2 are on a daily basis.  Subfactor relationships are also used in RUSLE2 but these relationships have been improved from RUSLE1, a new ridge subfactor has been added, and the deposition equations have been extended to consider sediment characteristics and how deposition changes these characteristics.  It includes new relationships for handling residue, including resurfacing of residue by implements like field cultivators.
The major change visible change in RUSLE2 is its new, modern graphical user interface that is easy to use, but is extremely powerful in the information that it displays and the types of situations that it can represent.  RUSLE2 is a vary powerful model yet it uses very simple, easy to obtain inputs.
RUSLE2 computes both temporal and spatially variable effects such as the effect of soil and management varying along a hillslope.
The RUSLE2 developers claim that it is the best available technology for conservation planning at the local field office level.  Its science is totally modern and is just as process-based as any other model and includes new science not available in any other model.
RUSLE2 brings together of the empirically based models like the USLE/RUSLE1 and the process based models like WEPP.
RUSLE2 does not compute ephemeral gully erosion at present, but capability is being planned for RUSLE2.  RUSLE2 will be the erosion prediction tool of choice for the next several years.
.