Development of the higher tier MACRO 5 drainflow tool

This page provides background information on how the MACRO 5 tool was developed and highlights some key changes from the MACRO 4 version.

The MACRO 5 higher tier drainflow (HTDF) tool has been developed to perform HTDF assessments on the Windows 10 operating system.

Guidance on the general approach to HTDF assessments.

The original version of the Health and Safety Executive’s (HSE’s) MACRO HTDF tool utilised MACRO 4. MACRO 4 operates on a Windows 7 and earlier operating system. Where possible, the same parameters used in the MACRO 4 tool have been used in the MACRO 5 tool. However, due to fundamental differences between MACRO 4 and MACRO 5, some parameters were unable to be transcribed directly from MACRO 4 to MACRO 5.

HSE also updated certain parameters and approaches to better align the outputs of the MACRO 4 and MACRO 5 tools. This included an initial review of the weather data used to define the climate scenarios. Further information on MACRO 5 and descriptions of the parameters, are provided on the Swedish University of Agricultural Sciences website.

• Required MACRO 5 updated parameters
• Updated weather files
• Elected MACRO 5 updated parameters
• Future updates
• Resources

Required MACRO 5 updated parameters

The following table summarises the parameters updated in the MACRO 5 tool and the reasoning.

Updated weather files

When MACRO 5 was run with only the above parameters updated, HSE’s initial analysis identified good alignment in results between MACRO 5 and MACRO 4 for more immobile compounds but poor alignment for more mobile compounds. Since the more mobile compounds are typically those that result in greatest exposure, this poor alignment required further investigation. HSE considers the difference in results to be due to the fundamental differences between MACRO 4 and MACRO 5 and the equations they use to perform the simulations. As such, HSE explored a number of options, including reviewing the weather files used in the simulations in order to improve the alignment. The weather files used in the MACRO 4 simulations were obtained from SEISMIC and were generated using the WGEN weather generator model (Richardson & Wright, 1984). The files include 32 years (1957 to 1988) of daily precipitation, minimum temperature, maximum temperature and potential evapotranspiration (PET; calculated using the Linacre equation (Linacre, 1977) with an additional adjustment factor) data. The locations of the weather files are ‘Dry’ = Cambridge, England, ‘Medium’ = Mylnefield, Scotland and ‘Wet’ = Keele, England.

Updated daily weather data from 1989 to 2020 has been obtained from the MARS 25 (v3.1, JRC) dataset. Grid squares corresponding to the same locations used in the MACRO 4 simulations were selected. These were: ‘Dry’ = 109082, ‘Medium’ = 129077 and ‘Wet’ = 113077. HSE compared the mean annual precipitation values from the differing weather files and with the HSE definition of ‘Dry’, ‘Medium’ and ‘Wet’, summarised below.

As the MARS 25 ‘Dry’ mean annual precipitation was still within the climatic range definition, this weather data was selected for use in the MACRO 5 assessment. The ‘Medium’ MARS 25 annual mean precipitation was slightly higher than the maximum limit of the medium range definition. However, as it was only slightly above the maximum, it was considered sufficiently similar to be used in the MACRO 5 assessment as an interim measure, pending a full investigation into the weather data and whether the original climatic ranges are still representative.

The ‘Wet’ MARS 25 annual mean precipitation was significantly lower than the lower limit of the wet range. Therefore, another site (grid square: 111074) on the Herefordshire, Shropshire and Welsh border was selected based on its statistical similarity (summarised below) and relative geographical proximity to the MACRO 4 Keele site. This weather data was used in the MACRO 5 assessment to represent wet climates.

In addition to providing daily precipitation, temperature and PET data, the MARS 25 dataset also provides daily solar radiation, vapour pressure and wind speed data. These additional parameters can be used by MACRO to calculate the PET within the model during the simulation. HSE ran MACRO 5 simulations using pre-defined PET values from the MARS 25 dataset and PET values calculated within the MACRO model. Although both the MARS 25 and MACRO programs use the Penman-Monteith equation (Monteith, 1965) to calculate the PET, an improved drainage profile was observed using the values calculated within the MACRO model. The Penman-Monteith equation is the favoured method of calculation by the UN Food and Agriculture Organization.

Therefore, this approach was selected going forward with MACRO 5 simulations and differs from the approach taken in MACRO 4. Note, the MARS 25 solar radiation values were converted from KJ/m2 to W/m2 by multiplying the values by 0.0116, in line with the approach detailed in the FOCUS SW repair report (EFSA, 2020). The MARS 25 vapour pressure values have also been converted from hPa to kPa by dividing the values by 10.

The resulting MACRO 5 drainflow results showed an improved alignment for more mobile compounds than exhibited using the MACRO 4 weather files.

Elected MACRO 5 updated parameters

One of the consequences of improving the mobile compound drainflow alignment between MACRO versions (using updated weather data and using the MACRO model to calculate PET) was the more immobile compound results became less well aligned.

HSE performed additional testing and determined that doubling the fraction of sorption sites in the macropores (FRACMAC) for each soil from the values used in MACRO 4 had the effect of somewhat counteracting the impact of using updated weather data and model simulated PET values. Therefore, by increasing the fraction of sorption sites in the macropores, this has the effect of helping to better align the drainflow values of the more immobile compounds with minimal impact on the mobile compounds.

The original FRACMAC values parameterised in MACRO 4 were selected based on soil-type estimates. These values and the proposed new values are presented as follows:

The updated FRACMAC values were checked to ensure the values were still within model constraints, i.e. less than the ratio of macroporosity to total porosity, using the following equations:

The results are as follows:

As can be seen, the new FRACMAC values are still less than the fractions of macroporosity to total porosity for each soil/layer, suggesting they are still appropriate for use.

Other parameters HSE has elected to update to improve the alignment between MACRO versions are summarised as follows:

Future updates

HSE will conduct a full review into the weather data to determine whether the climatic range definitions (and subsequently the sites selected) are still appropriate due to, for example, the impact of climate change.

Additional changes may arise because of an ongoing review into our approaches to assessing the risks from drainflow.

The MACRO 5 tool may be updated as further evidence is gathered from regulatory submissions received during the transitional period using either the new MACRO 5 tool or using the MACRO 4 tool (see MACRO higher tier drainflow modelling for pesticide registration in Great Britain and Northern Ireland for further information on HSE’s approach to evaluating HTDF assessments). Any future changes in approach will be communicated via these pages.

Resources

MACRO 4 PECsw drainflow (higher tier MACRO) tool from Environmental fate models: Excel calculator tools.

MACRO 5 PECsw drainflow (higher tier MACRO) tool from  Environmental fate models: Excel calculator tools.

EFSA, 2020. Scientific report of EFSA on the ‘repair action’ of the FOCUS surface water scenarios. EFSA Journal 2020;18(6):6119.

Linacre, E.T., 1977. A simple formula for estimating evaporation rates in various climates, using temperature data alone. Agricultural meteorology, 18:409-424.

Monteith, J. L., 1965. Evaporation and environment. Symposia of the Society for Experimental Biology. 19: 205–234.

Richardson, C.W. & Wright, D.A., 1984. WGEN: a model for generating daily weather variables. ARS-8, US Department of Agriculture, Agricultural Research Service.