Evaluation of the DNDC model for simulating soil temperature, moisture and respiration from monoculture and rotational corn, soybean and winter wheat in Canada

Li, Zhuoting; Yang, Jingyi; Drury, C. F.; Yang, Xueming; Reynolds, W. D.; Li, Yong; Hu, Chunsheng

doi:10.1016/j.ecolmodel.2017.07.013

Cited by 32 publications

(16 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the evaluation of Random Forest model performance, we used five model evaluation statistics including correlation coefficients of the “spearman” correlation test, mean error, root mean square error, index of agreement, and modeling efficiency (Supporting Information Table S1). For detailed descriptions on the features of each evaluation statistic please see previous publications (Li et al., ; Yang, Yang, Liu, & Hoogenboom, ). Each evaluation statistic reflects a specific aspect of the model performance, but the combination of five evaluation statistics helps quantify the overall model performance.…”

Section: Methodsmentioning

confidence: 99%

Constraining estimates of global soil respiration by quantifying sources of variability

Jian

Steele

Thomas

et al. 2018

Global Change Biology

View full text Add to dashboard Cite

Quantifying global soil respiration (R ) and its response to temperature change are critical for predicting the turnover of terrestrial carbon stocks and their feedbacks to climate change. Currently, estimates of R range from 68 to 98 Pg C year , causing considerable uncertainty in the global carbon budget. We argue the source of this variability lies in the upscaling assumptions regarding the model format, data timescales, and precipitation component. To quantify the variability and constrain R , we developed R models using Random Forest and exponential models, and used different timescales (daily, monthly, and annual) of soil respiration (R ) and climate data to predict R . From the resulting R estimates (range = 66.62-100.72 Pg), we calculated variability associated with each assumption. Among model formats, using monthly R data rather than annual data decreased R by 7.43-9.46 Pg; however, R calculated from daily R data was only 1.83 Pg lower than the R from monthly data. Using mean annual precipitation and temperature data instead of monthly data caused +4.84 and -4.36 Pg C differences, respectively. If the timescale of R data is constant, R estimated by the first-order exponential (93.2 Pg) was greater than the Random Forest (78.76 Pg) or second-order exponential (76.18 Pg) estimates. These results highlight the importance of variation at subannual timescales for upscaling to R The results indicated R is lower than in recent papers and the current benchmark for land models (98 Pg C year ), and thus may change the predicted rates of terrestrial carbon turnover and the carbon to climate feedback as global temperatures rise.

show abstract

Section: Methodsmentioning

confidence: 99%

Constraining estimates of global soil respiration by quantifying sources of variability

Jian

Steele

Thomas

et al. 2018

Global Change Biology

View full text Add to dashboard Cite

show abstract

“…They include mean error (ME), root mean squared error (RMSE), relative root mean squared error (rRMSE), model efficiency (ME) and the coefficient of determination (R 2 ). Since using one of the metrics is not sufficient, a combination of the metrics gives a better response on the performance of the model as noted by Li et al (2017).…”

Section: Model Accuracy Determinationmentioning

confidence: 99%

Parameterization, calibration and validation of the DNDC model for carbon dioxide, nitrous oxide and maize crop performance estimation in East Africa

Macharia

Ngetich

Shisanya

2021

Heliyon

View full text Add to dashboard Cite

Dynamic biogeochemical models are crucial tools for simulating the complex interaction between soils, climate and plants; thus the need for improving understanding of nutrient cycling and reduction of greenhouse gases (GHG) from the environment. This study aimed to calibrate and validate the DeNitrification-DeComposition (DNDC) model for soil moisture, temperature, respiration, nitrous oxide and maize crop growth simulation in drier sub-humid parts of the central highlands of Kenya. We measured soil GHG fluxes from a maize field under four different soil fertility management practices for one year using static chambers and gas chromatography. Using experimental data collected from four management practices during GHG sampling period, we parameterized the DNDC model. The results indicate that the DNDC model simulates daily and annual soil moisture, soil temperature, soil respiration (CO 2 ), nitrous oxide (N 2 O), N 2 O yield-scaled emissions (YSE), N 2 O emission factors (EFs) and maize crop growth with a high degree of fitness. However, the DNDC simulations slightly underestimated soil temperature (2-6%), crop growth (2-45%) and N 2 O emissions (5-23%). The simulation overestimated soil moisture (9-17%) and CO 2 emissions (3-10%). It however, perfectly simulated YSE and EFs. Compared to the observed/measured annual GHG trends, the simulation results were relatively good, with an almost perfect fitting of emission peaks during soil rewetting at the onset of rains, coinciding with soil fertilisation. These findings provide reliable information in selecting best farm management practice, which simultaneously improves agricultural productivity and reduces GHG emissions, thus permitting climate-smart agriculture. The good DNDC simulated YSE and EFs values (Tier III) provide cheaper and reliable ways of filling the huge GHG data gap, reducing uncertainties in national GHG inventories and result to efficient targeting of mitigation measures in sub-Saharan Africa.

show abstract

“…As a result, a model that considers a single correlation between meteorological parameters and the ST across different sites will suffer from obvious limitations. Second, the input of hourly soil temperature can help the model to obtain high-resolution simulation results in related studies such as using terrestrial biogeochemical models to simulate the dynamic changes of N 2 O, CO, N 2 , and CH 4 in soil [22][23][24], simulating soil respiration by land surface models [25], and quantifying hydrological and biological processes in hydrological models [26,27]. However, most of the current studies have focused on daily or monthly ST predictions, and hourly ST predictions remain scarce.…”

Section: Introductionmentioning

confidence: 99%

Modeling Hourly Soil Temperature Using Deep BiLSTM Neural Network

2020

View full text Add to dashboard Cite

Soil temperature (ST) plays a key role in the processes and functions of almost all ecosystems, and is also an essential parameter for various applications such as agricultural production, geothermal development, and their utilization. Although numerous machine learning models have been used in the prediction of ST, and good results have been obtained, most of the current studies have focused on daily or monthly ST predictions, while hourly ST predictions are scarce. This paper presents a novel scheme for forecasting the hourly ST using weather forecast data. The method considers the hourly ST prediction to be the superposition of two parts, namely, the daily average ST prediction and the ST amplitude (the difference between the hourly ST and the daily average ST) prediction. According to the results of correlation analysis, we selected nine meteorological parameters and combined two temporal parameters as the input vectors for predicting the daily average ST. For the task of predicting the ST amplitude, seven meteorological parameters and one temporal parameter were selected as the inputs. Two submodels were constructed using a deep bidirectional long short-term memory network (BiLSTM). For the task of hourly ST prediction at five different soil depths at 30 sites, which are located in 5 common climates in the United States, the results showed the method proposed in this paper performs best at all depths for 30 stations (100% of all) for the root mean square error (RMSE), 27 stations (90% of all) for the mean absolute error (MAE), and 30 stations (100% of all) for the coefficient of determination (R2), respectively. Moreover, the method adopted in this study displays a stronger ST prediction ability than the traditional methods under all climate types involved in the experiment, the hourly ST produced by it can be used as a driving parameter for high-resolution biogeochemical models, land surface models and hydrological models and can provide ideas for an analysis of other time series data.

show abstract

Evaluation of the DNDC model for simulating soil temperature, moisture and respiration from monoculture and rotational corn, soybean and winter wheat in Canada

Cited by 32 publications

References 39 publications

Constraining estimates of global soil respiration by quantifying sources of variability

Constraining estimates of global soil respiration by quantifying sources of variability

Parameterization, calibration and validation of the DNDC model for carbon dioxide, nitrous oxide and maize crop performance estimation in East Africa

Modeling Hourly Soil Temperature Using Deep BiLSTM Neural Network

Contact Info

Product

Resources

About