Simulating pesticides in ditches to assess ecological risk (SPIDER): II. Benchmarking for the drainage model |
| |
Authors: | Renaud Fabrice G Brown Colin D |
| |
Affiliation: | Cranfield University, School of Applied Sciences, Natural Resources Department, Integrated Environmental Systems Institute, Cranfield, Bedfordshire, MK43 0AL, UK. renaud@ehs.unu.edu |
| |
Abstract: | SPIDER (simulating pesticides in ditches to assess ecological risk) is a locally distributed, capacitance-based model that accounts for pesticide entry into surface water bodies via spray drift, surface runoff, interlayer flow and drainage. SPIDER was developed for application to small agricultural catchments. Transport of pesticide from site of application to surface water via subsurface field drains is one of the major routes of entry to surface water. Several pesticide fate models describe transfer of pesticide via drainflow, notably MACRO which has been evaluated against field data in several studies. The capacity of SPIDER to simulate drainflow and pesticide concentration in drain water was evaluated against two datasets that had been used previously to evaluate MACRO independently of this study: a plot experiment at Cockle Park and a field experiment at Maidwell, both located in the UK. In both circumstances, SPIDER was able to reproduce drain hydrographs relatively well with no or limited calibration. At Cockle Park, simulated and observed drainflow over the season were 240 and 278 mm, respectively with a Nash and Sutcliffe model efficiency (NSME) coefficient of 0.32 whilst at Maidwell they were 259 and 296 mm, respectively with a NSME coefficient of 0.55. Prediction of maximum isoproturon concentration at Cockle Park by SPIDER and MACRO were 5.3 and 13.1 microg L(- 1) respectively compared to the 3.8 microg L(- 1) measured in the field, whilst pesticide load to drains over the season were 0.22 and 1.53 g, respectively, compared to an observed load of 0.35 g. Maximum sulfosulfuron concentration at Maidwell were 2.3, 3.9 and 5.4 microg L(- 1) for observed and as simulated by SPIDER and MACRO, respectively and pesticide loading to drains of the season was 0.77, 5.61, 4.77 g, respectively. Results from the sensitivity analysis showed that the sensitivity of SPIDER compared favourably to that of several other capacity models but was more sensitive than MACRO to variations in input parameters. SPIDER is currently being tested at the catchment scale. |
| |
Keywords: | Pesticide Drainage Model Simulation Calibration |
本文献已被 ScienceDirect PubMed 等数据库收录! |
|