Identifying artificially drained pasture soils using machine learning and Earth observation imagery

Abstract. In many areas of the globe, the installation of artificial drains on naturally poorly drained soils is a necessary part of farm management. Identifying the location of artificially drained areas is an important step in achieving environmentally sustainable agricultural production. However, in many regions, data on the presence or the distribution of artificial drainage systems are rare. We outline an approach to identify artificially drained soils using Earth observation (EO) satellite imagery and digital elevation data. The method exploits the contrasting phenology of grass during a peak growth stage to identify artificially drained and undrained soils. Two machine-learning techniques, support vector machine and random forest, were tested. Classification accuracy up to 91% was achieved using photointerpreted accuracy points using higher resolution satellite imagery. Additional investigations would be required to establish whether the drained conditions identified were a result of artificial drainage or from naturally well-drained soils occurring within larger soil units. Herein, the Republic of Ireland is used as a test case. Based on our findings, the area of artificially drained grassland within the study area could be revised upward, with 44% (or ∼345  ,  000  ha) of pasture currently classed as “poorly drained” identified as “artificially drained.” At one location, a change in the modeled drainage condition at field level was demonstrated following drain installation. The presented method demonstrates the ability of EO satellites to quickly and accurately map field drainage status at farm management scales over a wide area. This has the potential to improve management decisions at local scales, but also has implications in terms of national policy development and regulation in areas such as water quality and climate change mitigation.

[1]  Reamonn Fealy,et al.  Assessing the role of artificially drained agricultural land for climate change mitigation in Ireland , 2018 .

[2]  J. A. Schell,et al.  Monitoring vegetation systems in the great plains with ERTS , 1973 .

[3]  Anne Gobin,et al.  Logistic Modeling to Spatially Predict the Probability of Soil Drainage Classes , 2002 .

[4]  Elizabeth Pattey,et al.  Mapping within-field soil drainage using remote sensing, DEM and apparent soil electrical conductivity , 2008 .

[5]  Nicholas M. Holden,et al.  The relationship between farmer opinion of suitable conditions for nutrient application, soil moisture deficit and weather , 2016 .

[6]  Neil Flood,et al.  Comparing Sentinel-2A and Landsat 7 and 8 Using Surface Reflectance over Australia , 2017, Remote. Sens..

[7]  David G. Rossiter,et al.  Technical Note: Statistical methods for accuracy assesment of classified thematic maps , 2004 .

[8]  Jungho Im,et al.  Support vector machines in remote sensing: A review , 2011 .

[9]  B. Minasny,et al.  On digital soil mapping , 2003 .

[10]  Jan M. H. Hendrickx,et al.  Remote sensing for soil map unit boundary detection , 2014 .

[11]  Owen Fenton,et al.  SPATIAL AND TEMPORAL VARIATIONS OF NUTRIENT LOADS IN OVERLAND FLOW AND SUBSURFACE DRAINAGE FROM A MARGINAL LAND SITE IN SOUTH-EAST IRELAND , 2022, Biology and Environment: Proceedings of the Royal Irish Academy.

[12]  Russell G. Congalton,et al.  A review of assessing the accuracy of classifications of remotely sensed data , 1991 .

[13]  pTools.  Building Control Management System , 2014 .

[14]  R. W. Skaggs,et al.  Hydrologic and water quality impacts of agricultural drainage , 1994 .

[15]  James C. Bell,et al.  Soil drainage class probability mapping using a soil-landscape model , 1994 .

[16]  Wanglu Peng,et al.  Delineating patterns of soil drainage class on bare soils using remote sensing analyses , 2003 .

[17]  AshrafM. Irfan,et al.  Model prediction of soil drainage classes over a large area using a limited number of field samples: A case study in the province of Nova Scotia, Canada , 2013 .

[18]  Kevin F. Smith,et al.  The effects of waterlogging on growth, photosynthesis and biomass allocation in perennial ryegrass (Lolium perenne L.) genotypes with contrasting root development , 2003, The Journal of Agricultural Science.

[19]  Paul Leahy,et al.  Carbon sequestration determined using farm scale carbon balance and eddy covariance , 2007 .

[20]  Ingmar Nitze,et al.  The Irish Land Mapping Observatory: Mapping and Monitoring Land Cover, Use and Change , 2017 .

[21]  Mogens Humlekrog Greve,et al.  Predicting artificially drained areas by means of a selective model ensemble , 2018, Geoderma.

[22]  S F Thornton,et al.  Influence of artificial drainage system design on the nitrogen attenuation potential of gley soils: Evidence from hydrochemical and isotope studies under field-scale conditions. , 2018, Journal of environmental management.

[23]  A. Brereton,et al.  Drumlin Soils ? The Depression of Herbage Yield by Shallow Water Table Depth , 1988 .

[24]  M. Greve,et al.  Artificial neural networks and decision tree classification for predicting soil drainage classes in Denmark , 2017, Geoderma.

[25]  Luis Alonso,et al.  Evaluation of Sentinel-2 Red-Edge Bands for Empirical Estimation of Green LAI and Chlorophyll Content , 2011, Sensors.

[26]  V. L. Mulder,et al.  The use of remote sensing in soil and terrain mapping — A review , 2011 .

[27]  J Humphreys,et al.  Phosphorus and nitrogen losses from temperate permanent grassland on clay-loam soil after the installation of artificial mole and gravel mole drainage. , 2019, The Science of the total environment.

[28]  Robert S. Freeland,et al.  Overall results and key findings on the use of UAV visible-color, multispectral, and thermal infrared imagery to map agricultural drainage pipes , 2020 .

[29]  N. Holden,et al.  The effect of local climate and soil drainage on the environmental impact of grass-based milk production , 2017, The International Journal of Life Cycle Assessment.

[30]  J. V. Soares,et al.  HAND, a new terrain descriptor using SRTM-DEM: Mapping terra-firme rainforest environments in Amazonia , 2008 .

[31]  Rogier P.O. Schulte,et al.  A review of the role of excess soil moisture conditions in constraining farm practices under Atlantic conditions , 2012 .

[32]  H. Kobryn,et al.  Remote sensing for assessing the zone of benefit where deep drains improve productivity of land affected by shallow saline groundwater. , 2015, Journal of environmental management.

[33]  C. Thorne,et al.  Quantitative analysis of land surface topography , 1987 .

[34]  B. Gao.  NDWI—A normalized difference water index for remote sensing of vegetation liquid water from space , 1996 .

[35]  E. R. Levine,et al.  Predicting Soil Drainage Class Using Remotely Sensed and Digital Elevation Data , 1997 .

[36]  Nicholas M. Holden,et al.  Simulation of the influence of poor soil drainage on grass‐based dairy production systems in Ireland , 2008 .

[37]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[38]  K. Beven,et al.  A physically based, variable contributing area model of basin hydrology , 1979 .

[39]  Rogier P.O. Schulte,et al.  A note on the Hybrid Soil Moisture Deficit Model v2.0 , 2015 .

[40]  C. Woodcock,et al.  Improvement and expansion of the Fmask algorithm: cloud, cloud shadow, and snow detection for Landsats 4–7, 8, and Sentinel 2 images , 2015 .

[41]  T. McCarthy,et al.  The agricultural impact of the 2015–2016 floods in Ireland as mapped through Sentinel 1 satellite imagery , 2019, Irish Journal of Agricultural and Food Research.

[42]  Laurence Shalloo,et al.  Comparison of a pasture-based system of milk production on a high rainfall, heavy-clay soil with that on a lower rainfall, free-draining soil , 2004 .

[43]  M. I. Dragila,et al.  Ground-based magnetic surveys as a new technique to locate subsurface drainage pipes: A case study , 2005 .

[44]  Feng Gao,et al.  LEDAPS Calibration, Reflectance, Atmospheric Correction Preprocessing Code, Version 2 , 2013 .

[45]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[46]  O. Fenton,et al.  Greenhouse gas emissions from temperate permanent grassland on clay-loam soil following the installation of artificial drainage , 2019, Agriculture, Ecosystems & Environment.

[47]  Simon Kraatz,et al.  Identifying Subsurface Drainage using Satellite Big Data and Machine Learning via Google Earth Engine , 2019, Water Resources Research.

[48]  E. A. Garwood,et al.  Hydrological consequences of artificial drainage of grassland , 1991 .

[49]  Mahesh Pal,et al.  Random forest classifier for remote sensing classification , 2005 .

[50]  Annette Freibauer,et al.  High CO 2 fluxes from grassland on histic Gleysol along soil carbon and drainage gradients , 2014 .

[51]  Chi-Chih Chen,et al.  DETECTION OF BURIED AGRICULTURAL DRAINAGE PIPE WITH GEOPHYSICAL METHODS , 2004 .

[52]  Harry Vereecken,et al.  Aerial photograph-based delineation of artificially drained areas as a basis for water balance and phosphorus modelling in large river basins , 2009 .

[53]  L. F. Galvin.  THE DRAINAGE OF IMPERMEABLE SOILS IN HIGH RAINFALL AREAS , 1983 .

[54]  Laura C. Bowling,et al.  Detecting subsurface drainage systems and estimating drain spacing in intensively managed agricultural landscapes. , 2009 .

[55]  George Alan Blackburn,et al.  Discrimination of plant stress caused by oil pollution and waterlogging using hyperspectral and thermal remote sensing , 2013 .