谢梅香, 张展羽, 张平仓, 徐金鑫, 林庆明. 紫色土坡耕地硝态氮的迁移流失规律及其数值模拟[J]. 农业工程学报, 2018, 34(19): 147-154. DOI: 10.11975/j.issn.1002-6819.2018.19.019
    引用本文: 谢梅香, 张展羽, 张平仓, 徐金鑫, 林庆明. 紫色土坡耕地硝态氮的迁移流失规律及其数值模拟[J]. 农业工程学报, 2018, 34(19): 147-154. DOI: 10.11975/j.issn.1002-6819.2018.19.019
    Xie Meixiang, Zhang Zhanyu, Zhang Pingcang, Xu Jinxin, Lin Qingming. Law of nitrate transfer and loss in purple sloping farmland and its numerical simulation[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(19): 147-154. DOI: 10.11975/j.issn.1002-6819.2018.19.019
    Citation: Xie Meixiang, Zhang Zhanyu, Zhang Pingcang, Xu Jinxin, Lin Qingming. Law of nitrate transfer and loss in purple sloping farmland and its numerical simulation[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(19): 147-154. DOI: 10.11975/j.issn.1002-6819.2018.19.019

    紫色土坡耕地硝态氮的迁移流失规律及其数值模拟

    Law of nitrate transfer and loss in purple sloping farmland and its numerical simulation

    • 摘要: 为探究紫色土坡耕地硝态氮迁移流失过程,通过室内模拟试验,并结合数学模型,研究在不同雨强(0.4,1.0,1.8 mm/min)和坡度(5?,15?,20?)下硝态氮分别随地表径流和壤中流迁移而流失的特征。结果表明:随地表径流和壤中流迁移的硝态氮流失浓度随时间分别呈现指数下降和线性上升趋势;随壤中流流失的浓度是地表径流携带的19~72倍,在小雨强下壤中流携带流失负荷大于随地表径流流失负荷,但随雨强增大,硝态氮流失负荷通过地表径流流失的比例由17.3%增大至66.0%,大雨强下硝态氮主要通过地表径流流失;与实测数据比较分析,有效混合深度模型在随地表径流流失的硝态氮模拟中精度评价指标Nash-Suttcliffe系数ENS和决定系数R2达到0.590和0.826 7,而对流弥散方程在壤中流携带硝氮流失的过程模拟中ENS和R2达到0.792和0.842 6,取得较好的模拟结果。该研究为紫色土坡耕地硝态氮迁移流失机理研究提供依据和参考。

       

      Abstract: Abstract: The nitrate transfer and loss plays a critical role in groundwater contamination. Specially, the purple soil sloping field accounts for large areas of farmlands in China and the nitrate transport and loss in purple soils causes serious pollution towards waterbodies. To research the transfer and loss of nitrate in sloping field of purple soil, we conducted laboratory experiments using soil tanks and artificial rainfall device to study nitrate loss features by surface flow (SF) and subsurface flow (SSF) subjected to various precipitation intensities and slope gradients. In this study, 3 precipitation intensities (0.4, 1.0, 1.8 mm/min) coupled with 3 slope gradients (5°, 15°, 20°) were used, and totally 9 treatments were conducted with 2 repetition. Besides, numerical modelling approach was also applied to investigate the nitrate transfer and loss characteristics of purple soil in sloping field. The modified effective mixing depth model and convective-dispersion equation were applied in simulations of nitrate loss via SF and SSF, respectively, with the effective mixing model was modified by a time-increasing effective mixing depth and convective-dispersion equation was solved by HYDRUS-2D software. The results showed: 1) exponential decrease between nitrate concentration and time through SF and linear increment through SSF. 2) The loss concentration of nitrate in SSF was 19-72 folds more than that in SF, and the nitrate loss cumulative loss loads through SF and SSF both presented linear increments with time. Additionally, the variabilities of measured nitrate concentration in SSF were much larger than that in SF. 3) The nitrate loss load presented linear increment over time subjected to all treatments and the proportion of nitrate loss load in SF increased with increasing precipitation intensities. As a result, the loss load of nitrate was mainly through SF in response to large precipitation intensity, but nitrate loss was mainly through SSF subjected to low precipitation intensity. In detail, the proportion of nitrate loss through SF increased from 17.3% to 66.0% as response to increasing rainfall intensity from 0.4 to 1.8 mm/min. 4) The precipitation intensity was a very influential factor for nitrate nitrogen loss, while the impact of slope gradient on nitrate loss showed no consistent pattern. 5) The linear regressions between model prediction results and experimental data and evaluation index of accuracy for simulation results both revealed good agreements for nitrate transfer and loss through SF and SSF, respectively. The mean absolute error (MAE), root mean square error (RMSE), coefficient of Nash-Suttcliffe ENS and R2 reached 0.872 mg/L, 1.009 mg/L, 0.590 and 0.826 7, respectively, for nitrate loss via SF. Similarly, the MAE, RMSE, ENS and R2 of prediction for subsurface nitrate loss reached 9.889 mg/L, 13.084 mg/L, 0.792 and 0.842 6, respectively. This study provided better understanding for nitrate transfer and loss mechanism of purple soil in sloping farmland.

       

    /

    返回文章
    返回