ORIGINAL RESEARCH
Evaluation of Mathematical Models in Nitrogen Transfer to Overland Flow Subjected to Simulated Rainfall
,

,

More details
Hide details
1
College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing, China

2
College of Agricultural Engineering, Hohai University, Nanjing, China

3
Changjiang River Scientiﬁc Research Institute, Wuhan, China

Submission date: 2018-12-06

Final revision date: 2019-03-12

Acceptance date: 2019-03-28

Online publication date: 2019-10-04

Publication date: 2020-01-16

Corresponding author
Zhanyu Zhang

College of Water Conservancy and Hydropower Engineering, Hohai university, China

Pol. J. Environ. Stud. 2020;29(2):1421-1434

KEYWORDS
TOPICS
ABSTRACT
Nitrogen (N) transfer to runoff contributed to nutrient loss and water pollution. Experiments were conducted to study the ammonia nitrogen (NH4-N), nitrate nitrogen (NO3-N) and total nitrogen (TN) transfer from loam soil to runoff in response to various rainfall intensities (RIs) (0.4±0.02, 1.0±0.04, and 1.8±0.11 mm min-1) and slope gradients (SGs) (5°, 10°, 15° and 20°). A typical mathematical model based on effective mixing depth (hm) and a refined model which replaced the time-average hm in this typical model with a time-increasing hm were both applied to predict N transfer to runoff. These models were verified with experimental data to evaluate the applications in simulations of surface N dynamics. NH4-N and TN concentrations in overland flow presented large deviations but NO3-N concentration highly declined from the initiation of runoff and then stabilized with slight deviations. The effective mixing depth deduced from fitted results coincided positively with RI but negatively with SG. The linear regressions between model prediction and experimental results revealed better agreements for NO3-N (r2 = 0.696; Slope = 1.1617) than NH4-N (r2 = 0.2538; slope = 0.7916) and TN (r2 = 0.224; slope = 0.6658). The refined model showed improved performance compared with the original model for the NO3-N (r2 = 0.8267; slope = 0.9996; intercept = -0.2675 versus r2 = 0.696; slope = 1.1617; intercept = -0.0438).
 eISSN: 2083-5906 ISSN: 1230-1485