Journal Paper Digests 2018 #18
- Optimum soil water content sensors placement for surface drip irrigation scheduling in layered soils
- Multi-stage committee based extreme learning machine model incorporating the influence of climate parameters and seasonality on drought forecasting
- An Integrated View of Complex Landscapes: A Big Data-Model Integration Approach to Transdisciplinary Science
- The Interactive Role of Wind and Water in Functioning of Drylands: What Does the Future Hold?
- Soil carbon dynamics in wheat plots established on grassland in 1911 as influenced by nitrogen and phosphorus fertilizers
- Spatial modelling with Euclidean distance fields and machine learning
- Testing soil phosphorus in a depleting P scenario: an accelerated soil mining experiment
- Simultaneous determination of soil bulk density and water content: a heat pulse-based method
- High-Resolution Shortwave Infrared Imaging of Water Infiltration into Dry Soil
- Measurement and Partitioning of Evapotranspiration for Application to Vadose Zone Studies
Optimum soil water content sensors placement for surface drip irrigation scheduling in layered soils
Authors: Soulis, KX; Elmaloglou, S
Source: COMPUTERS AND ELECTRONICS IN AGRICULTURE, 152 1-8; SEP 2018
Abstract: Efficient irrigation management requires a sound information basis; therefore, various environmental measurements are currently used in irrigation scheduling. Among other technics, the recent progress in electromagnetic sensors technologies promoted the development of automated irrigation scheduling systems based on soil water content sensors with very promising results in terms of water savings. However, a key factor for the adequate performance of such systems is proper placement of soil moisture sensors. Up to now, sensor placement guidelines are fragmentary or empirically determined from site and crop specific experiments. This study aims to extend the findings of previous studies investigating the issue of proper positioning of water content sensors in drip irrigation scheduling systems in uniform soils for the case of layered soils. In this context the representativeness of soil water content sensors’ readings and the existence of Time Stable Representative Positions (TSRP) are investigated using a specially developed mathematical model. The use of soil water content probes that are able to monitor soil water content at various depths is also evaluated. It was found that in contrast to the previous findings concerning uniform soils, in the case of layered soils it is not possible to precisely monitor the average soil water content temporal variation in the root zone using a single sensor; however, it is feasible to achieve this using a pair of sensors. Furthermore, common optimum positions for a pair of sensors providing representative soil water content readings independently from the prevailing conditions and the irrigation system configuration can be identified. It was also found that soil water content probes covering the average rooting depth and penetrating both soil layers are also able to provide representative soil water content readings during the whole duration of the irrigation cycle. The above results represent a further step towards the development of general guidelines for sensor placement in soil water content based surface drip irrigation scheduling systems.
Multi-stage committee based extreme learning machine model incorporating the influence of climate parameters and seasonality on drought forecasting
Authors: Ali, M; Deo, RC; Downs, NJ; Maraseni, T
Source: COMPUTERS AND ELECTRONICS IN AGRICULTURE, 152 149-165; SEP 2018
Abstract: Drought forewarning is an important decisive task since drought is perceived a recurrent feature of climate variability and climate change leading to catastrophic consequences for agriculture, ecosystem sustainability, and food and water scarcity. This study designs and evaluates a soft-computing drought modelling framework in context of Pakistan, a drought-stricken nation, by means of a committee extreme learning machine (Comm-ELM) model in respect to a committee particle swarm optimization-adaptive neuro fuzzy inference system (Comm-PSO-ANFIS) and committee multiple linear regression (Comm-MLR) model applied to forecast monthly standardized precipitation index (SPI). The proposed Comm-ELM model incorporates historical monthly rainfall, temperature, humidity, Southern Oscillation Index (SOI) at monthly lag (t - 1) and the respective month (i.e., periodicity factor) as the explanatory variable for the drought’s behaviour defined by SPI. The model accuracy is assessed by root mean squared error, mean absolute error, correlation coefficient, Willmott’s index, Nash-Sutcliffe efficiency and Legates McCabe’s index in the independent test dataset. With the incorporation of periodicity as an input factor, the performance of the Comm-ELM model for Islamabad, Multan and Dera Ismail Khan (D. I. Khan) as the test stations, was remarkably improved in respect to the Comm-PSO-ANFIS and Comm-MLR model. Other than the superiority of Comm-ELM over the alternative models tested for monthly SPI forecasting, we also highlight the importance of the periodicity cycle as a pertinent predictor variable in a drought forecasting model. The results ascertain that the model accuracy scales with geographic factors, due to the complexity of drought phenomenon and its relationship with the different inputs and data attributes that can affect the overall evolution of a drought event. The findings of this study has important implications for agricultural decision-making where future knowledge of drought can be used to develop climate risk mitigation strategies for better crop management.
An Integrated View of Complex Landscapes: A Big Data-Model Integration Approach to Transdisciplinary Science
Authors: Peters, DPC; Burruss, ND; Rodriguez, LL; Mcvey, DS; Elias, EH; Pelzel-Mccluskey, AM; Derner, JD; Schrader, TS; Yao, J; Pauszek, SJ; Lombard, J; Archer, SR; Bestelmeyer, BT; Browning, DM; Brungard, CW; Hatfield, JL; Hanan, NP; Herrick, JE; Okin, GS; Sala, OE; Savoy, H; Vivoni, ER
Source: BIOSCIENCE, 68 (9):653-669; SEP 2018
Abstract: The Earth is a complex system comprising many interacting spatial and temporal scales. We developed a transdisciplinary data-model integration (TDMI) approach to understand, predict; and manage for these complex dynamics that focuses on spatiotemporal modeling and cross-scale interactions. Our approach employs human-centered machine-learning strategies supported by a data science integration system (DSIS). Applied to ecological problems, our approach integrates knowledge and data on (a) biological processes, (b) spatial heterogeneity in the land surface template, and (c) variability in environmental drivers using data and knowledge drawn from multiple lines of evidence (cc., observations, experimental manipulations, analytical and numerical models, products from imagery, conceptual model reasoning, and theory). We apply this transdisciplinary approach to a suite of increasingly complex ecologically relevant problems and then discuss how information management systems will need to evolve into DSIS to allow other transdisciplinary questions to be addressed in the future.
The Interactive Role of Wind and Water in Functioning of Drylands: What Does the Future Hold?
Authors: Okin, GS; Sala, OE; Vivoni, ER; Zhang, JZ; Bhattachan, A
Source: BIOSCIENCE, 68 (9):670-677; SEP 2018
Abstract: Feedback mechanisms between abiotic and biotic processes in dryland ecosystems lead to a strong sensitivity to interannual variations in climate. Under a future regime of higher temperatures but potentially increased rainfall variability drylands are anticipated to experience changes in wind and water transport that will alter plant community composition and feedback on landscape connectivity Here, we present a conceptual framework for understanding the coupling of vegetation productivity, aeolian transport, and hydrologic connectivity under anticipated changes in future climate, which suggests that a more extreme climatic regime will lead to more connected landscapes with attendant losses in soil, nutrient, and water resources. When enhanced connectivity triggers state changes, irreversible changes in ecosystem functioning can occur, with implications for the future of global drylands.
Soil carbon dynamics in wheat plots established on grassland in 1911 as influenced by nitrogen and phosphorus fertilizers
Authors: Karimi, R; Janzen, HH; Smith, EG; Ellert, BH; Krobel, R
Source: CANADIAN JOURNAL OF SOIL SCIENCE, 98 (3):580-583; SEP 2018
Abstract: Soil organic carbon (SOC) changes slowly, and final management influences can be measured only after decades. Analysis of archived samples from a site established on grassland in 1911 showed that SOC, under wheat systems, approached steady state after several decades, and that its amount reflected the inputs of residue C.
Spatial modelling with Euclidean distance fields and machine learning
Authors: Behrens, T; Schmidt, K; Rossel, RAV; Gries, P; Scholten, T; MacMillan, RA
Source: EUROPEAN JOURNAL OF SOIL SCIENCE, 69 (5):757-770; SEP 2018
Abstract: This study introduces a hybrid spatial modelling framework, which accounts for spatial non-stationarity, spatial autocorrelation and environmental correlation. A set of geographic spatially autocorrelated Euclidean distance fields (EDF) was used to provide additional spatially relevant predictors to the environmental covariates commonly used for mapping. The approach was used in combination with machine-learning methods, so we called the method Euclidean distance fields in machine-learning (EDM). This method provides advantages over other prediction methods that integrate spatial dependence and state factor models, for example, regression kriging (RK) and geographically weighted regression (GWR). We used seven generic (EDFs) and several commonly used predictors with different regression algorithms in two digital soil mapping (DSM) case studies and compared the results to those achieved with ordinary kriging (OK), RK and GWR as well as the multiscale methods ConMap, ConStat and contextual spatial modelling (CSM). The algorithms tested in EDM were a linear model, bagged multivariate adaptive regression splines (MARS), radial basis function support vector machines (SVM), Cubist, random forest (RF) and a neural network (NN) ensemble. The study demonstrated that DSM with EDM provided results comparable to RK and to the contextual multiscale methods. Best results were obtained with Cubist, RF and bagged MARS. Because the tree-based approaches produce discontinuous response surfaces, the resulting maps can show visible artefacts when only the EDFs are used as predictors (i.e. no additional environmental covariates). Artefacts were not obvious for SVM and NN and to a lesser extent bagged MARS. An advantage of EDM is that it accounts for spatial non-stationarity and spatial autocorrelation when using a small set of additional predictors. The EDM is a new method that provides a practical alternative to more conventional spatial modelling and thus it enhances the DSM toolbox.
Testing soil phosphorus in a depleting P scenario: an accelerated soil mining experiment
Authors: Nawara, S; van Dael, T; De Cooman, E; Elsen, A; Merckx, R; Smolders, E; Amery, F
Source: EUROPEAN JOURNAL OF SOIL SCIENCE, 69 (5):804-815; SEP 2018
Abstract: The reduced use of phosphorus (P) fertilizer in fertile soil has reverted the P balance to negative in some regions. It is unclear how long current soil P stocks will ensure adequate P supply to crops. In addition, it is unknown if current soil tests for available P describe bioavailable P adequately in soil where P is becoming depleted. We set up an accelerated soil P mining test to address these questions. Perennial ryegrass (Lolium perenne, Melpetra tetra) was grown for 2years in a greenhouse on 5-cm-deep soil layers of eight contrasting soils with periodic grass clipping. Each soil was split into four fertilizer treatments (i.e. no P (-P) and adequate P (+P)) and two nitrogen levels, the latter to alter the rate of P uptake. The long-term P mining induced P-related yield losses in seven of the 16 soil treatments. The cumulative uptake of shoot P at which yield loss started to exceed 10% (-P versus +P) varied over a small range of 37-74mg Pkg(-1) soil among the soils. This critical cumulative P uptake (CCP) was related to the soil P content prior to mining measured by five soil P tests (ammonium oxalate, ammonium lactate (AL), Olsen P, 0.01m CaCl2 and the diffusive gradient in thin film technique (DGT)); the largest R-2 values were observed for P-AL (R-2=0.72) and P-DGT (R-2=0.73). However, none of the tests was diagnostic for yield loss during the depletion period. Increased N supply accelerated growth and rates of P uptake and decreased the CCP by a factor of 1.7 on average, illustrating the effect of the rate of biomass production. The CCP values obtained in the treatment with reduced N fertilizer application are likely to be the most relevant for the field and suggest that current stocks allow adequate P supply for arable crops for 3-8years under zero P application (0-23cm) in soils similar to those tested. The lack of a successful diagnosis for P deficiency during this depletion experiment calls for further calibration of soil tests for available P in the field.
Simultaneous determination of soil bulk density and water content: a heat pulse-based method
Authors: Lu, Y; Horton, R; Ren, T
Source: EUROPEAN JOURNAL OF SOIL SCIENCE, 69 (5):947-952; SEP 2018
Abstract: Soil bulk density ((b)) and volumetric water content () determine the volume fractions of soil solids, water and air, and influence mass and energy transfer in soil. It is desirable to monitor (b) and concurrently and non-destructively. We present a heat pulse-based method for simultaneous determination of (b) and from soil thermal properties. The method uses equations that relate (b) and to soil volumetric heat capacity (C) and to soil thermal conductivity (). We developed a three-step procedure to calculate (b) and from C and measured by a heat pulse sensor, with soil texture and specific heat of soil solids known a priori. Laboratory evaluation of soil samples with various textures showed that the three-step method provided reliable estimates of (b) and at values greater than the critical water content ((c)) when started to respond notably to increases in . This method provides a new way to determine (b) and simultaneously with heat pulse sensors.HighlightsWe developed an approach to determine soil bulk density (.. b) and water content (..) simultaneously with a heat pulse sensor. We estimated.. b and.. from soil thermal properties based on heat capacity and thermal conductivity models. The new approach provided reliable.. b and.. values at water contents >.. c, the critical value. At.. <.. c, the approach gave unstable results because soil thermal conductivity was insensitive to.. b.
High-Resolution Shortwave Infrared Imaging of Water Infiltration into Dry Soil
Authors: Sadeghi, M; Sheng, WY; Babaeian, E; Tuller, M; Jones, SB
Source: VADOSE ZONE JOURNAL, 16 (13):NIL_91-NIL_100; DEC 2017
Abstract: A novel proximal sensing framework for high-resolution soil water content profile retrieval under laboratory conditions has been developed. Constant-head upward-flow experiments were conducted for a number of soils that cover a wide textural range. The soils were packed into quartz Hele-Shaw cells, and the profile was imaged at high temporal frequency with a short-wave infrared (SWIR) camera in the 900- to 1700-nm electromagnetic domain during upward infiltration of water. The SWIR reflectance recorded for each spatial pixel was converted to soil water content with a recently developed linear physical model. Because of the linearity of the model, its parameters were assumed to be identical at both the pixel and column scales, and this allowed simple self-calibration during the experiment. The obtained water content profiles were in good agreement with soil water content data measured independently with a recently developed time domain reflectometry array with 1-cm depth resolution. In addition, the accuracy of the soil water content profiles was verified based on the water mass balance. The high-spatiotemporal-resolution SWIR reflectance-derived water content profiles allow calculation of water flux densities, which provides a potential new avenue for the rapid estimation of soil hydraulic properties and processes via inverse numerical or analytical modeling.
Measurement and Partitioning of Evapotranspiration for Application to Vadose Zone Studies
Authors: Anderson, RG; Zhang, XD; Skaggs, TH
Source: VADOSE ZONE JOURNAL, 16 (13):NIL_82-NIL_90; DEC 2017
Abstract: Partitioning evapotranspiration (ET) into its constituent components, evaporation (E) and transpiration (T), is important for numerous hydrological purposes including assessing impacts of management practices on water use efficiency and improved validation of vadose zone models that parameterize E and T separately. However, most long-established observational techniques have short observational timescales and spatial footprints, raising questions about the representativeness of these measurements. In the past 15 yr, new approaches have allowed ET partitioning at spatial scales ranging from the pedon to the globe and at long timescales. In this update, we review some recent methodological developments for partitioning ET. These include micrometeorological approaches involving the flux variance partitioning of high-frequency eddy covariance observations and proxies for photosynthesis and transpiration such as measurements of isotopic fractionation and carbonyl sulfide uptake. We discuss advances in partitioning the energy balance between canopy and soil using remote sensing. We conclude that the flux variance partitioning with raw eddy covariance data and the two-source energy balance approaches with remote sensing platforms may have the greatest potential for partitioning ET, in part because large public repositories of eddy covariance and satellite data could be readily reprocessed to partition ET.