Neutrons on Rails: Transregional Monitoring of Soil Moisture and Snow Water Equivalent

Abstract

Water stored in soils and snow controls the energy and water exchange between the terrestrial surface and the atmosphere (Vogel et al., 2018), impacts regional weather, and shapes the development of hydrometeorological extremes like heat waves, droughts, floods, or avalanches (e.g., Douville & Chauvin, 2000;Lehning et al., 1999;Liang & Yuan, 2021). Therefore, a solid estimation of land surface water at relevant spatiotemporal scales is of utmost importance.Satellite-based remote sensing platforms aim at global estimations of water in soils and snow at resolutions of several kilometers with the prospect of finer resolutions using new instrumentation and algorithms (Chan et al., 2016;Foucras et al., 2020;Mattia et al., 2018). However, major limitations are the shallow measurement depth (∼cm), long return frequencies (∼days), and low performance during complex weather conditions, under vegetation cover, and in complex terrain (Fang & Lakshmi, 2014;Lawford, 2014). Ground-based in situ sensors have been developed to measure water content in different soil depths with high spatiotemporal precision and extent from the point to smaller field scales (

Empirical analysis of machine learning-based moisture sensing platforms for agricultural applications: A statistical perspective

“…Based on this ARS value, it can be observed that ECT [24], PQCWC [25], MHPS [21], FoS [20], HPCM [6], PHS [17], HF RFID TFS [9], PMMA [15], FFCSM [16], HSAAA [32], PWM [10], FTO [35], CRNS [3] and CM [11] have better overall performance, thus can be used for efficient moisture sensing applications. LEWS LR [1] LR RBM [2] CRNS [3] MWMS [5] GPS [7] HF RFID TFS [9] CM [11] PLMR [12] MSR [14] FFCSM [16] SMAP [18] FoS [20] PRS [22] PQCWC [25] GOFCHS [27] SAR [29] CSMOS [31] SSMDI [33] FTO [35] PBG [38] MSOCCML [40] CRNS [42] Accuracy of moisture sensing models…”

“…While highly accurate sensing interfaces are costly, but can be used for high-speed moisture sensing applications. Specifically, DBN RBM, CRNS, SMAP LEWS LR [1] DBN RBM [2] LR RBM [2] BP RBM [2] CRNS [3] FBG [4] MWMS [5] HPCM [6] GPS [7] UHF RFID [8] HF RFID TFS [9] PWM [10] CM [11] PPMR [12] PLMR [12] RFID UHF [13] MSR [14] PMMA [15] FFCSM [16] PHS [17] SMAP [18] SMAP RF DN [19] FoS [20] MHPS [21] PRS [22] ECT [24] PQCWC [25] HDES [26] GOFCHS [27] TDR [28] SAR [29] SMI MODIS [30] CSMOS [31] HSAAA [32] SSMDI [33] P Band & L Band [34] FTO [35] eSMAP [36] PBG [38] MSNs [39] MSOCCML [40] SMAP TFC [41] CRNS [42] kCRNS [43] Computational RF DN, GOFCHS, TDR, and P Band & L Band models outperform other models; thus, they can be used for highly accurate moisture detection applications. While, HPCM, HF RFID TFS, PWM, PMMA, FFCSM, MHPS, ECT, PQCWC, ...…”

“…Here, PBG is used for high accuracy, laboratory applications, while MSNs are deployed for large-scale & accurate sensing of moisture levels, which assists in improving their applicability for a wide variety of deployment scenarios. Similar applications are discussed in [40,41,42], wherein researchers have deployed moisture sensors for 3D mapping of soil organic carbon content with machine learning (MSOCCML), detection of moisture for temperate forest canopies using SMAP (SMAP TFC), and Cosmic-ray neutron sensing (CRNS) for train materials. It was observed that MSOCCML is useful for general purpose applications, while SMAP & CRNS must be retuned and deployed for context-specific application scenarios.…”

Probabilistic prediction by means of the propagation of response variable uncertainty through a Monte Carlo approach in regression random forest: Application to soil moisture regionalization

“…The measured quantity is the counting number of neutrons per time unit (Zreda et al, 2012). This quantity has to be transformed by a function g into gravimetric soil moisture and associated uncertainties, while the measurement uncertainty of neutron counts has been already propagated through g (Jakobi et al, 2020;Schrön et al, 2021).…”

“…As a benchmark, we use prediction uncertainty achieved by QRRF, which ignores the quantified uncertainty of the input data. Our response variable is gravimetric soil moisture sparsely measured by mobile cosmic-ray neutron sensing (McJannet et al, 2017;Schrön et al, 2018;Jakobi et al, 2020;Schrön et al, 2021). The second goal of this study is to perform a comparative analysis of QRRF and MC approaches to understand the kind of uncertainty quantified by each method.…”