Prediction of exchangeable potassium in soil through mid-infrared spectroscopy and deep learning:From prediction to explainability

The ability to characterize rapidly and repeatedly exchangeable potassium(Kex)content in the soil is essential for optimizing remediation of radiocaesium contamination in agriculture.In this paper,we show how this can be now achieved using a Convolutional Neural Network(CNN)model trained on a large Mid-Infrared(MIR)soil spectral library(40,000 samples with Kex determined with 1 M NH4OAc,pH 7),compiled by the National Soil Survey Center of the United States Department of Agriculture.Using Partial Least Squares Regression as a base-line,we found that our implemented CNN leads to a significantly higher prediction performance of Kex when a large amount of data is available(10000),increasing the coefficient of determination from 0.64 to 0.79,and reducing the Mean Absolute Percentage Error from 135%to 31%.Furthermore,in order to provide end-users with required interpretive keys,we implemented the GradientShap algorithm to identify the spectral regions considered important by the model for predicting Kex.Used in the context of the implemented CNN on various Soil Taxonomy Orders,it allowed(i)to relate the important spectral features to domain knowledge and(ii)to demonstrate that including all Soil Taxonomy Orders in CNN-based modeling is beneficial as spectral features learned can be reused across different,sometimes underrepresented orders.

Mitigating greenhouse gas emissions from croplands and pasturelands—climate-smart agriculture

One of the main challenges facing humankind is ensuring food security for a rapidly growing population with lower environmental footprints under changing climate. Environmental unsustainability of agro-food systems is multi-faced,but alteration of biogeochemical cycles (e.g., nitrogen (N)and phosphorus (P) cycles) and emissions of greenhouse gases (GHGs) to the atmosphere have been reported as one of the main disruptive forces over safe-operating space of planetary boundaries (Springmann et al., 2018).

Effects of the nitrification inhibitor nitrapyrin and tillage practices on yield-scaled nitrous oxide emission from a maize field in Iran

Nitrification inhibitors can effectively decrease nitrification rates and nitrous oxide(N_(2)O)emission while increasing crop yield under certain conditions.However,there is no information available on the effects of nitrification inhibitors and tillage practices on N_(2)O emissions from maize cropping in Iran.To study how tillage practices and nitrapyrin(a nitrification inhibitor)affect N_(2)O emission,a split factorial experiment using a completely randomized block design with three replications was carried out in Northeast Iran,which has a cold semiarid climate.Two main plots were created with conventional tillage and minimum tillage levels,and two nitrogen(N)fertilizer(urea)management systems(with and without nitrapyrin application)were created as subplots.Tillage level did not have any significant effect on soil ammonium(NH_(4)^(+))and nitrate(NO_(3)^(-))concentrations,cumulative amount and yield-scaled N_(2)O emission,and aboveground biomass of maize,whereas nitrapyrin application showed significant effect.Nitrapyrin application significantly reduced the cumulative amount of N_(2)O emission by 41%and 32%in conventional tillage and minimum tillage practices,respectively.A reduction in soil NO_(3)^(-)concentration by nitrapyrin was also observed.The average yield-scaled N_(2)O emission was 13.6 g N_(2)O-N kg^(-1)N uptake in both tillage systems without nitrapyrin application and was significantly reduced to 7.9 and 8.2 g N_(2)O-N kg^(-1)N uptake upon the application of nitrapyrin in minimum tillage and conventional tillage practices,respectively.Additionally,nitrapyrin application increased maize biomass yield by 4%and 13%in the minimum tillage and conventional tillage systems,respectively.Our results indicate that nitrapyrin has a potential role in reducing N_(2)O emission from agricultural systems where urea fertilizers are broadcasted,which is common in Iran due to the practice of traditional farming.

Effects of the nitrification inhibitor nitrapyrin and the plant growth regulator gibberellic acid on yield-scale nitrous oxide emission in maize fields under hot climatic conditions

Nitrification inhibitors are widely used in agriculture to mitigate nitrous oxide(N_(2)O)emission and increase crop yield.However,no concrete information on their mitigation of N_(2)O emission is available under soil and environmental conditions as in Pakistan.A field experiment was established using a silt clay loam soil from Peshawar,Pakistan,to study the effect of urea applied in combination with a nitrification inhibitor,nitrapyrin(2-chloro-6-tri-chloromethyl pyridine),and/or a plant growth regulator,gibberellic acid(GA_3),on N_(2)O emission and the nitrogen(N)uptake efficiency of maize.The experimental design was a randomized complete block with five treatments in four replicates:control with no N(CK),urea(200 kg N ha^(-1))alone,urea in combination with nitrapyrin(700 g ha^(-1)),urea in combination with GA_3(60 g ha^(-1)),and urea in combination with nitrapyrin and GA_3.The N_(2)O emission,yield,N response efficiency,and total N uptake were measured during the experimental period.The treatment with urea and nitrapyrin reduced total N_(2)O emission by 39%–43%and decreased yield-scaled N_(2)O emission by 47%–52%,relative to the treatment with urea alone.The maize plant biomass,grain yield,and total N uptake increased significantly by 23%,17%,and 15%,respectively,in the treatment with urea and nitrapyrin,relative to the treatment with urea alone,which was possibly due to N saving,lower N loss,and increased N uptake in the form of ammonium;they were further enhanced in the treatment with urea,nitrapyrin,and GA_3 by 27%,36%,and 25%,respectively,probably because of the stimulating effect of GA_3 on plant growth and development and the reduction in biotic and abiotic stresses.These results suggest that applying urea in combination with nitrapyrin and GA_3 has the potential to mitigate N_(2)O emission,improve N response efficiency,and increase maize yield.

Nitrapyrin effectiveness in reducing nitrous oxide emissions decreases at low doses of urea in an Andosol

In the tropics,frequent nitrogen(N)fertilization of grazing areas can potentially increase nitrous oxide(N_(2)O)emissions.The application of nitrification inhibitors has been reported as an effective management practice for potentially reducing N loss from the soil-plant system and improving N use efficiency(NUE).The aim of this study was to determine the effect of the co-application of nitrapyrin(a nitrification inhibitor,NI)and urea in a tropical Andosol on the behavior of N and the emissions of N_(2)O from autotrophic and heterotrophic nitrification.A greenhouse experiment was performed using a soil(pH 5.9,organic matter content 78 g kg^(-1),and N 5.6 g kg^(-1))sown with Cynodon nlemfuensis at 60%water-filled pore space to quantify total N_(2)O emissions,N_(2)O derived from fertilizer,soil ammonium(NH_(4)^(+))and nitrate(NO_(3)^(-)),and NUE.The study included treatments that received deionized water only(control,NI).No significant differences were observed in soil NH_(4)^(+)content between the UR and UR+NI treatments,probably because of soil mineralization and NO_(3)^(-)produced by heterotrophic nitrification,which is not effectively inhibited by nitrapyrin.After 56 d,N_(2)O emissions in UR(0.51±0.12 mg N_(2)O-N concluded that the soil organic N mineralization and heterotrophic nitrification are the main processes of NH_(4)^(+)and NO_(3)^(-)production.Additionally,it was found that N_(2)O emissions were partially a consequence of the direct oxidation of the soil's organic N via heterotrophic nitrification coupled to denitrification.Finally,the results suggest that nitrapyrin would likely exert significant mitigation on N_(2)O emissions only if a substantial N surplus exists in soils with high organic matter content.

Nitrification inhibitor nitrapyrin does not affect yield-scaled nitrous oxide emissions in a tropical grassland

Urea is the most common nitrogen(N)fertilizer used in the tropics but it has the risk of high gaseous nitrogen(N)losses.Use of nitrification inhibitor has been suggested as a potential mitigation measure for gaseous N losses in N fertilizer-applied fields.In a field trial on a tropical Andosol pastureland in Costa Rica,gaseous emissions of ammonia(NH_(3))and nitrous oxide(N_(2)O)and grass yield were quantified from plots treated with urea(U;41.7 kg N ha^(-1)application^(-1))and urea plus the nitrification inhibitor nitrapyrin(U+NI;41.7 kg N ha^(-1)application^(-1)and 350 g of nitrapyrin for each 100 kg of N applied)and control plots(without U and NI)over a six-month period(rainy season).Volatilization of NH_(3)(August to November)in U(7.4%±1.3%of N applied)and U+NI(8.1%±0.9%of N applied)were not significantly different(P>0.05).Emissions of N_(2)O in U and U+NI from June to November were significantly different(P<0.05)only in October,when N_(2)O emission in U+NI was higher than that in U.Yield and crude protein production of grass were significantly higher(P<0.05)in U and U+NI than in the control plots,but they were not significantly different between U and U+NI.There was no significant difference in yield-scaled N_(2)O emission between U(0.31±0.10 g N kg^(-1)dry matter)and U+NI(0.47±0.10 g N kg^(-1)dry matter).The results suggest that nitrapyrin is not a viable mitigation option for gaseous N losses under typical N fertilizer application practices of pasturelands at the study site.

A simple and easy method to measure ammonia volatilization:Accuracy under field conditions

Field studies on soil ammonia(NH_(3))volatilization are restricted in many countries owing to the high costs commonly demanded for accurate quantification.We assessed the accuracy of a simple,open chamber design to capture NH_(3)under field conditions,as affected by different chamber placement schemes.Urea-15 N was surface applied to lysimeters installed in the spaces between maize rows.Open chambers made from plastic bottles were installed on each lysimeter with variations in i)N rates(3,8,13,and 18 g m^(-2)),ii)the height of the chamber above the soil surface(0,5,and 10 mm),and iii)chamber relocation(static vs.dynamic).Reference lysimeters without chambers were used to measure NH_(3)losses by^(15)N-balance.Losses of NH_(3)-N accounted for more than 50%of the applied N.Relocation of the chambers had no impact on their NH_(3)-trapping efficiencies,proving to be an unnecessary procedure.Variation in the height of the chambers above the soil surface affected the capture of NH_(3),but the results still maintained high linearity with the NH_(3)losses quantified by the reference method(R^(2)>0.98).When the same placement scheme used in the introductory study describing the chamber was utilized(static and touching the soil surface),we found a trapping efficiency of 60%,which was very similar to that(57%)obtained in the previous study.Our results show that this simple,open chamber design can be used with satisfactory accuracy under field conditions,provided that simple,standardized procedures are warranted.

Assessment of deforestation impact on soil erosion in loess formation using 137Cs method (case study: Golestan Province, Iran)

Golestan,a province in the North-East of Iran,is characterized by high coverage of loess deposits.Since 1963,the area has experienced approximately 200,000 ha deforestation due to land-use changes in agriculture and increasing demand for wood.Approximately,110,000 ha of the clear-cut lands are under dry-farming,mainly for wheat cropping,and about 86,000 ha have been reforested.This IAEA funded project is the first attempt to use nuclear techniques in the East of Hircanian Forest for determination of on-site impacts of deforestation due to two land-use changes(i.e.dry farming and reforestation).Practicing long-term dry-farming led to 60%soil losses with a mean rate of 2 mm per year.The net erosion rate of croplands on loess deposits in the study area was 32.27 t ha-1 yr-1.Reforestation,cultivation of even-aged Cypress trees since 1993,showed 13 to 60 percent effectiveness in soil conservation.Dry-farming land use resulted in the loss of 95 t ha-1 soil organic carbon(SOC)stock at a mean rate of 1.7 t ha-1 over 54 years.Cultivating Cypress trees successfully restored the SOC content by 100%compared with the SOC in original forests.The conversion of dry-farming lands to orchards of olive trees since 2004,brought more income for farmers but were less effective in soil conservation because of low canopy cover of olive trees.Our data provide key information and guidance for land users and decisionmakers about implementing strategic and sustainable conservation practices to restore degraded land.

Use of geochemical fingerprints to trace sediment sources in an agricultural catchment of Argentina

Soil erosion and associated sediment redistribution are key environmental problems in Central Argentina.Specific land uses and management practices,such as intensive grazing and crop cultivation,are considered to be significantly driving and accelerating these processes.This research focuses on the identification of suitable soil tracers from hot spots of land degradation and sediment fate in an agricultural catchment of central Argentina with erodible loess soils.Using Energy Dispersive X-Ray Fluorescence(EDXRF),elemental concentrations were determined and later used as soil tracers for geochemical characterization.The best set of tracers were identified using two artificial mixtures composed of known proportions of soil sources collected from different lands having contrasting soil uses.Barium,calcium,iron,phosphorus,and titanium were identified for obtaining the best suitable reconstruction of source proportions in the laboratory-prepared artificial mixtures.Then,these elements,as well as the total organic carbon,were applied for pinpointing critical hot spots of erosion within the studied catchment.Feedlots were identified to be the main source of sediments,river banks and dirt roads together are the second most important source.This investigation provides key information for optimizing soil conservation strategies and selecting land management practices and land uses which do not generate great contribution of sediment,preventing pollution of the waterways of the region.

Optimizing the use of open chambers to measure ammonia volatilization in field plots amended with urea

Measuring ammonia(NH_(3))volatilization from urea-fertilized soils is crucial for evaluation of practices that reduce gaseous nitrogen(N)losses in agriculture.The small area of chambers used for NH_(3)volatilization measurements compared with the size of field plots may cause significant errors if inadequate sampling strategies are adopted.Our aims were:i)to investigate the effect of using multiple open chambers on the variability in the measurement of NH_(3)volatilization in urea-amended field plots and ii)to define the critical period of NH_(3)-N losses during which the concentration of sampling effort is capable of reducing uncertainty.The use of only one chamber covering 0.015%of the plot(51.84 m^(2))generates a value of NH_(3)-N loss within an expected margin of error of 30%around the true mean.To reduce the error margin by half(15%),3–7 chambers were required with a mean of 5 chambers per plot.Concentrating the sampling efforts in the first two weeks after urea application,which is usually the most critical period of N losses and associated errors,represents an efficient strategy to lessen uncertainty in the measurements of NH_(3)volatilization.This strategy enhances the power of detection of NH_(3)-N loss abatement in field experiments using chambers.