Table 1.

Summary of the explanatory variables and their statistical difference among the three clusters of monitoring stations. For land use, geology, and physiography variables, the median of annual values in 2007–2018 was used. Each feature was separately evaluated against the response variable using the one-way analysis of variance (ANOVA) test. A feature is statistically different among the three clusters if the reported p-value is less than 0.1 (bold).

Explanatory variable	Definition	ANOVA p-value
Land use (n = 4)
Natural_pct	Fraction of natural land, percent	0.00013
Crop_pct	Fraction of cropland, percent	0.023
Pasture_pct	Fraction of pastureland, percent	0.82
Hay_pct	Fraction of hay land, percent	0.27
Geology (n = 1)
Carb_pct	Fraction of carbonate geology, percent	0.089
Physiography (n = 5)
Appalachian_pct	Fraction of Appalachian physiography, percent	0.12
BlueRidge_pct	Fraction of Blue Ridge physiography, percent	0.78
ValleyRidge_pct	Fraction of Valley and Ridge physiography, percent	0.69
Piedmont_pct	Fraction of Piedmont physiography, percent	0.13
Coastal_pct	Fraction of Coastal Plain physiography, percent	0.61
Mann-Kendall (MK) trends of N source (n = 6)
PointSource MK	MK trend of point source for the period of 2007–2018, scaled by the period-of-record median	0.33
Deposition MK	MK trend of atmospheric deposition for the period of 1997–2018, scaled by the period-of-record median	0.038
Fertilizer MK	MK trend of agricultural fertilizer for the period of 1997–2018, scaled by the period-of-record median	0.036
Manure MK	MK trend of manure for the period of 1997–2018, scaled by the period-of-record median	0.28
AgInput_MK	MK trend of agricultural input for the period of 1997–2018, scaled by the period-of-record median	0.81
AgSurplus_MK	MK trend of agricultural surplus for the period of 1997–2018, scaled by the period-of-record median	0.14