Brief description
The Soil Moisture Integration and Prediction System (SMIPS) produces national extent daily estimates of volumetric soil moisture at a resolution of approximately 1km or 0.01 decimal degrees. SMIPS also generates an index of between 0-1 which approximates how full the 90cm metre soil moisture store is at a particular location and time. The SMIPS model itself consists of two linked soil moisture stores, a shallow quick responding 10cm upper store and a deeper, slower responding 80cm store. SMIPS is parameterised using physical properties from the Soil and Landscape Grid of Australia and takes a data model fusion approach for model forcing. Version 1.0 of the SMIPS model uses precipitation and potential evapotranspiration data from the Bureau of Meteorology’s AWRA Model. In addition to version 1.0 of the model, an experimental version of the model is available for user testing. This version of the model uses precipitation data supplied by an experimental CSIRO daily rainfall surface generated using spatial data from the NASA Global Precipitation Mission as a base and enhanced using rainfall observations from the Bureau of Meteorology (BoM) rainfall gauge network, and various landscape covariates, processed using a machine learning approach.To help increase model accuracy, the internal SMIPS model states are adjusted or ‘bumped’ by daily observational data from the European Space Agency’s Soil Moisture and Ocean Salinity (SMOS) satellite mission.
Lineage
Progress Code: onGoingNotes
CreditWe at TERN acknowledge the Traditional Owners and Custodians throughout Australia, New Zealand and all nations. We honour their profound connections to land, water, biodiversity and culture and pay our respects to their Elders past, present and emerging.
The development of the SMIPS National Soil Moisture products has been made possible by the National Landcare Program, through the Department of Agriculture and Water Resources, and by TERN, an NCRIS-funded national research infrastructure project. SMIPS model updates and re-factorisation, together with creation of associated digital soil information model inputs performed by CSIRO. Modelling workflow design and model code implementation by CSIRO. The Rainfall forcing and Potential Evapotranspiration data are supplied by the Australian Bureau of Meteorology. The Experimental Machine Learning Derived Rainfall data are supplied by CSIRO. The core SMIPS model is adapted from N.S.Wimalathunge and T.F.A.Bishop https://www.sciencedirect.com/science/article/pii/S0016706118316434 Original SMIPS concept by Luigi Renzullo CSIRO Land and Water (now at Australian National University).
The SMIPS system provides daily estimates of volumetric soil moisture and a soil moisture index to assist with modelling such as crop growth and yield, runoff generation, deep drainage and can be used along with other relevant data for purposes such as fertiliser application planning, irrigation scheduling, fire risk assessments and as a component of a general landscape condition assessment.
Data Quality Assessment Scope
local :
dataset
The modelled daily volumetric soil moisture estimates have been validated against a network of 86 publicly available soil moisture probes across Australia, mostly in SE agricultural regions. 76 of these probes are capacitance type probes which typically measure a soil moisture value with a 5cm radius of the probes at 10cm interval down to 1 metre. Ten of the probes were cosmic ray neutron type probes (TERN CosmOz Network: https://cosmoz.csiro.au/) which use neutron counts to estimate soil moisture over an area of approximately 2Ha to a depth ranging from 10 - 40 cm. These Cosmos probes integrate soil moisture measurements at a similar scale to that of the model and thus are well suited to this validation method. The raw probe soil moisture values were converted to volumetric values using automated transformation routines and compared to the modelled values at the same locations to derive a statistical validation.
Data Quality Assessment Result
local :
Quality Result
There are numerous sources of errors in the observed soil moisture values and transformation routines used to estimate volumetric moisture contents, so the validations are at best indicative only. <br>
The model had an overall R2 of 0.51. The R2 is the proportion of the variation in the model that can be explained by the values of the soil moisture probe values or degree of correlation between 2 sets of values. It gives an indication of the overall model accuracy. <br>
The model had an overall Lin’s Concordance Correlation Coefficient (LCCC) of 0.48. The LCC is a measure of how well the model values compare to the observed data. LCC measures both precision and accuracy with values of -1 to 1, with perfect agreement at 1. LCCC. <br>
Created: 2021-08-25
Issued: 2021-10-20
Modified: 2024-10-22
Data time period: 2015-11-20
User Contributed Tags
Login to tag this record with meaningful keywords to make it easier to discover
Point-of-truth metadata URL- URI : geonetwork.tern.org.au/geonetwork/srv/eng/catalog.search#/metadata/d1995ee8-53f0-4a7d-91c2-ad5e4a23e5e0
- global : d1995ee8-53f0-4a7d-91c2-ad5e4a23e5e0
