Management operations that control the plant growth cycle, the timing of fertilizer and pesticide and the removal of plant biomass are explained in this chapter. Water management and the simulation of urban areas are summarized in subsequent chapters.

A new operations input file was added that allows user to schedule management by Julian day and calendar year without considering cropping rotations and without using heat unit scheduling. Operations have been added for contouring, terracing, subsurface drains, filter strips, fire, and grass waterways.

Previous versions of SWAT+ only allowed the growth of one plant species at a time to be simulated. Algorithms from the ALMANAC model (Kiniry et al., 1992, Johnson et al., 2009) have been added to simulate multiple plant species growing and competing within a plant community. Plant communities that have been simulated include: crops and weeds, trees and grasses, different tree species in a boreal forest, and grasses and shrubs in rangeland communities. A data file is developed prior to simulation that describes the various plants within each community.

The kill operation stops plant growth in the HRU. All plant biomass is converted to residue.

The only information required by the kill operation is the timing of the operation (month and day or fraction of plant potential heat units).

Biological mixing is the redistribution of soil constituents as a result of the activity of biota in the soil (e.g. earthworms, etc.). Studies have shown that biological mixing can be significant in systems where the soil is only infrequently disturbed. In general, as a management system shifts from conventional tillage to conservation tillage to no-till there will be an increase in biological mixing. SWAT+ allows biological mixing to occur to a depth of 300 mm (or the bottom of the soil profile if it is shallower than 300 mm). The efficiency of biological mixing is defined by the user. The redistribution of nutrients by biological mixing is calculated using the same methodology as that used for a tillage operation.

Fertilization in an HRU may be scheduled by the user or automatically applied by SWAT+. When the user selects auto-application of fertilizer in an HRU, a nitrogen stress threshold must be specified. The nitrogen stress threshold is a fraction of potential plant growth. Anytime actual plant growth falls below this threshold fraction due to nitrogen stress, the model will automatically apply fertilizer to the HRU. The user specifies the type of fertilizer, the fraction of total fertilizer applied to the soil surface, the maximum amount of fertilizer that can be applied during the year, the maximum amount of fertilizer that can be applied in any one application, and the application efficiency.

To determine the amount of fertilizer applied, an estimate of the amount of nitrogen that will be removed in the yield is needed. For the first year of simulation, the model has no information about the amount of nitrogen removed from the soil by the plant. The nitrogen yield estimate is initially assigned a value using the following equations:

$yld_{est,N}=350*fr_{N,yld}*RUE$ if $HI_{opt}<1.0$ 6:1.8.1

$yld_{est,N}=1000*fr_{N,yld}*RUE$ if $HI_{opt}\ge 1.0$ 6:1.8.2

where $yld_{est,N}$ is the nitrogen yield estimate (kg N/ha), $fr_{N,yld}$ is the fraction of nitrogen in the yield, $RUE$ is the radiation-use efficiency of the plant (kg/ha⋅(MJ/m$^2$)$^{-1}$ or 10$^{-1}$g/MJ), and $HI_{opt}$ is the potential harvest index for the plant at maturity given ideal growing conditions. The nitrogen yield estimate is updated at the end of every simulation year using the equation:

$yld_{est,N}=\frac{yld_{est,Nprev}*yr_{sim}+yld_{yr,N}}{yr_{sim}+1}$ 6:1.8.3

where $yld_{est,N}$ is the nitrogen yield estimate update for the current year (kg N/ha), $yld_{est,Nprev}$ is the nitrogen yield estimate from the previous year (kg N/ha), $yr_{sim}$ is the year of simulation, $yld_{yr,N}$ is the nitrogen yield target for the current year (kg N/ha). The nitrogen yield target for the current year is calculated at the time of harvest using the equation:

$yld_{yr,N}=bio_{ag}*fr_N*fert_{eff}$ 6:1.8.4

where $yld_{yr,N}$ is the nitrogen yield target for the current year (kg N/ha), $bio_{ag}$ is the aboveground biomass on the day of harvest (kg ha$^{-1}$), $fr_N$ is the fraction of nitrogen in the plant biomass calculated with equation 5:2.3.1, and $fert_{eff}$ is the fertilizer application efficiency assigned by the user. The fertilizer application efficiency allows the user to modify the amount of fertilizer applied as a function of plant demand. If the user would like to apply additional fertilizer to adjust for loss in runoff, $fert_{eff}$ will be set to a value greater than 1. If the user would like to apply just enough fertilizer to meet the expected demand, $fert_{eff}$ will be set to 1. If the user would like to apply only a fraction of the demand, $fert_{eff}$ will be set to a value less than 1.

The optimal amount of mineral nitrogen to be applied is calculated:

$minN_{app}=yld_{est,N}-(NO3+NH4)-bio_N$ 6:1.8.5

where $minN_{app}$ is the amount of mineral nitrogen applied (kg N/ha), $yld_{est,N}$ is the nitrogen yield estimate (kg N/ha), $NO3$ is the nitrate content of the soil profile (kg NO$_3$-N/ha), NH4 is the ammonium content of the soil profile (kg NH$_4$-N/ha), and $bio_N$ is the actual mass of nitrogen stored in plant material (kg N/ha). If the amount of mineral nitrogen calculated with equation 6:1.8.5 exceeds the maximum amount allowed for any one application, $minN_{app}$ is reset to the maximum value ($minN_{app}=minN_{app,mx}$). The total amount of nitrogen applied during the year is also compared to the maximum amount allowed for the year.

Once the amount applied reaches the maximum amount allowed for the year ($minN_{app,mxyr}$), SWAT+ will not apply any additional fertilizer regardless of nitrogen stress. Once the amount of mineral nitrogen applied is determined, the total amount of fertilizer applied is calculated by dividing the mass of mineral nitrogen applied by the fraction of mineral nitrogen in the fertilizer:

$fert=\frac{minN_{app}}{fert_{minN}}$ 6:1.8.6

where $fert$ is the amount of fertilizer applied (kg/ha), $minN_{app}$ is the amount of mineral nitrogen applied (kg N/ha), and $fert_{minN}$ is the fraction of mineral nitrogen in the fertilizer.

The type of fertilizer applied in the HRU is specified by the user. In addition to mineral nitrogen, organic nitrogen and phosphorus and mineral phosphorus are applied to the HRU. If a manure is applied, bacteria loadings to the HRU are also determined. The amount of each type of nutrient and bacteria is calculated from the amount of fertilizer and fraction of the various nutrient types in the fertilizer as summarized in Section 6:1.7.

While the model does not allow fertilizer to be applied as a function of phosphorus stress, the model does monitor phosphorus stress in the auto-fertilization subroutine. If phosphorus stress causes plant growth to fall below 75% of potential growth, the model ignores the fraction of mineral phosphorus in the fertilizer and applies an amount of mineral phosphorus equal to ($\frac{1}{7}*minN_{app}$).

Table 6:1-8: SWAT+ input variables that pertain to auto-fertilization.

The fertilizer operation applies fertilizer or manure to the soil.

Information required in the fertilizer operation includes the timing of the operation (month and day or fraction of plant potential heat units), the type of fertilizer/manure applied, the amount of fertilizer/manure applied, and the depth distribution of fertilizer application.

SWAT+ assumes surface runoff interacts with the top 10 mm of soil. Nutrients contained in this surface layer are available for transport to the main channel in surface runoff. The fertilizer operation allows the user to specify the fraction of fertilizer that is applied to the top 10 mm. The remainder of the fertilizer is added to the first soil layer defined in the HRU .sol file.

In the fertilizer database, the weight fraction of different types of nutrients and bacteria are defined for the fertilizer. The amounts of nutrient added to the different pools in the soil are calculated:

$NO3_{fert}=fert_{minN}*(1-fert_{NH4})*fert$ 6:1.7.1

$NH4_{fert}=fert_{minN}*fert_{NH4}*fert$ 6:1.7.2

$orgN_{frsh,fert}=0.5*fert_{orgN}*fert$ 6:1.7.3

$orgN_{act,fert}=0.5*fert_{orgN}*fert$ 6:1.7.4

$P_{solution,fert}=fert_{minP}*fert$ 6:1.7.5

$orgP_{frsh,fert}=0.5*fert_{orgP}*fert$ 6:1.7.6

$orgP_{hum,fert}=0.5*fert_{orgP}*fert$ 6:1.7.7

where $NO3_{fert}$ is the amount of nitrate added to the soil in the fertilizer (kg N/ha), $NH4_{fert}$ is the amount of ammonium added to the soil in the fertilizer (kg N/ha), $orgN_{frsh,fert}$ is the amount of nitrogen in the fresh organic pool added to the soil in the fertilizer (kg N/ha), $orgN_{act,fert}$ is the amount of nitrogen in the active organic pool added to the soil in the fertilizer (kg N/ha), $P_{solution,fert}$ is the amount of phosphorus in the solution pool added to the soil in the fertilizer (kg P/ha), $orgP_{frsh,fert}$ is the amount of phosphorus in the fresh organic pool added to the soil in the fertilizer (kg P/ha), $orgP_{hum,fert}$ is the amount of phosphorus in the humus organic pool added to the soil in the fertilizer (kg P/ha), $fert_{minN}$ is the fraction of mineral N in the fertilizer, $fert_{NH4}$ is the fraction of mineral $N$ in the fertilizer that is ammonium, $fert_{orgN}$ is the fraction of organic $N$ in the fertilizer, $fert_{minP}$ is the fraction of mineral $P$ in the fertilizer, $fert_{orgP}$ is the fraction of organic $P$ in the fertilizer, and $fert$ is the amount of fertilizer applied to the soil (kg/ha).

If manure is applied, the bacteria in the manure may become attached to plant foliage or be incorporated into the soil surface layer during application. The amount of bacteria reaching the ground surface and the amount of bacteria adhering to the plant foliage is calculated as a function of ground cover. The ground cover provided by plants is:

$gc=\frac{1.99532-erfc[1.333*LAI-2]}{2.1}$ 6:1.7.8

where $gc$ is the fraction of the ground surface covered by plants, $erfc$ is the complementary error function, and $LAI$ is the leaf area index.

The complementary error function frequently occurs in solutions to advective-dispersive equations. Values for $erfc(\beta)$ and $erf(\beta)$ ($erf$ is the error function for $\beta$), where $\beta$ is the argument of the function, are graphed in Figure 6:1-1. The figure shows that $erf(\beta)$ ranges from –1 to +1 while $erfc(\beta)$ ranges from 0 to +2. The complementary error function takes on a value greater than 1 only for negative values of the argument.

Once the fraction of ground covered by plants is known, the amount of bacteria applied to the foliage is calculated:

$bact_{lp,fol}=\frac{gc*fr_{active}*fert_{lpbact}*fert}{10}$ 6:1.7.9

$bact_{p,fol}=\frac{gc*fr_{active}*fert_{pbact}*fert}{10}$ 6:1.7.10

and the amount of bacteria applied to the soil surface is

$bact_{lpsol,fert}=\frac{(1-gc)*fr_{active}*fert_{lpbact}*k_{bact}*fert}{10}$ 6:1.7.11

$bact_{lpsorb,fert}=\frac{(1-gc)*fr_{active}*fert_{lpbact}*(1-k_{bact})*fert}{10}$ 6:1.7.12

$bact_{psol,fert}=\frac{(1-gc)*fr_{active}*fert_{pbact}*k_{bact}*fert}{10}$ 6:1.7.13

$bact_{psorb,fert}=\frac{(1-gc)*fr_{active}*fert_{pbact}*(1-k_{bact})*fert}{10}$ 6:1.7.14

where $bact_{lp,fol}$ is the amount of less persistent bacteria attached to the foliage (# cfu/m$^2$), $bact_{p,fol}$ is the amount of persistent bacteria attached to the foliage (# cfu/m$^2$), $bact_{lpsol,fert}$ is the amount of less persistent bacteria in the solution pool added to the soil (# cfu/m$^2$), $bact_{lpsorb,fert}$ is the amount of less persistent bacteria in the sorbed pool added to the soil (# cfu/m$^2$), $bact_{psol,fert}$ is the amount of persistent bacteria in the solution pool added to the soil (# cfu/m$^2$), $bact_{psorb,fert}$ is the amount of persistent bacteria in the sorbed pool added to the soil (# cfu/m$^2$), $gc$ is the fraction of the ground surface covered by plants, $fr_{active}$ is the fraction of the manure containing active colony forming units, $fert_{lpbact}$ is the concentration of less persistent bacteria in the fertilizer (# cfu/g manure), $fert_{pbact}$ is the concentration of persistent bacteria in the fertilizer (# cfu/g manure), $k_{bact}$ is the bacterial partition coefficient, and $fert$ is the amount of fertilizer/manure applied to the soil (kg/ha).

Table 6:1-7: SWAT+ input variables that pertain to fertilizer application.

A primary mechanism of disposal for manure generated by intensive animal operations such as confined animal feedlots is the land application of waste. In this type of a land management system, waste is applied every few days to the fields. Using the continuous fertilization operation allows a user to specify the frequency and quantity of manure applied to an HRU without the need to insert a fertilizer operation in the management file for every single application.

The continuous fertilizer operation requires the user to specify the beginning date of the continuous fertilization period, the total length of the fertilization period, and the number of days between individual fertilizer/manure applications. The amount of fertilizer/manure applied in each application is specified as well as the type of fertilizer/manure.

Nutrients and bacteria in the fertilizer/manure are applied to the soil surface. Unlike the fertilization operation or auto-fertilization operation, the continuous fertilization operation does not allow the nutrient and bacteria loadings to be partitioned between the surface 10 mm and the part of the 1st soil layer underlying the top 10 mm. Everything is added to the top 10 mm, making it available for transport by surface runoff.

Nutrient and bacteria loadings to the HRU are calculated using the equations reviewed in Section 6:1.7.

Table 6:1-9: SWAT+ input variables that pertain to continuous fertilization.

The pesticide operation applies pesticide to the HRU.

Information required in the pesticide operation includes the timing of the operation (month and day or fraction of plant potential heat units), the type of pesticide applied, and the amount of pesticide applied.

Field studies have shown that even on days with little or no wind, a portion of pesticide applied to the field is lost. The fraction of pesticide that reaches the foliage or soil surface is defined by the pesticide’s application efficiency. The amount of pesticide that reaches the foliage or ground is:

6:1.10.1

where is the effective amount of pesticide applied (kg pst/ha), is the pesticide application efficiency, and pest is the actual amount of pesticide applied (kg pst/ha).

The amount of pesticide reaching the ground surface and the amount of pesticide added to the plant foliage is calculated as a function of ground cover. The ground cover provided by plants is:

6:1.10.2

where is the fraction of the ground surface covered by plants, is the complementary error function, and is the leaf area index.

The complementary error function frequently occurs in solutions to advective-dispersive equations. Values for and (erf is the error function for ), where is the argument of the function, are graphed in Figure 6:1-1. The figure shows that ranges from –1 to +1 while ranges from 0 to +2. The complementary error function takes on a value greater than 1 only for negative values of the argument.

Once the fraction of ground covered by plants is known, the amount of pesticide applied to the foliage is calculated:

6:1.10.3

and the amount of pesticide applied to the soil surface is

6:1.10.4

where is the amount of pesticide applied to foliage (kg pst/ha), is the amount of pesticide applied to the soil surface (kg pst/ha), is the fraction of the ground surface covered by plants, and is the effective amount of pesticide applied (kg pst/ha).

Table 6:1-10: SWAT+ input variables that pertain to pesticide application.

The grazing operation simulates plant biomass removal and manure deposition over a specified period of time. This operation is used to simulate pasture or range grazed by animals.

Information required in the grazing operation includes the time during the year at which grazing begins (month and day or fraction of plant potential heat units), the length of the grazing period, the amount of biomass removed daily, the amount of manure deposited daily, and the type of manure deposited. The amount of biomass trampled is an optional input.

Biomass removal in the grazing operation is similar to that in the harvest operation. However, instead of a fraction of biomass being specified, an absolute amount to be removed every day is given. In some conditions, this can result in a reduction of the plant biomass to a very low level that will result in increased erosion in the HRU. To prevent this, a minimum plant biomass for grazing may be specified (BIO_MIN). When the plant biomass falls below the amount specified for BIO_MIN, the model will not graze, trample, or apply manure in the HRU on that day.

If the user specifies an amount of biomass to be removed daily by trampling, this biomass is converted to residue.

Nutrient fractions and bacteria content of the manure applied during grazing must be stored in the fertilizer database. The manure nutrient and bacteria loadings are added to the topmost 10 mm of soil. This is the portion of the soil with which surface runoff interacts.

After biomass is removed by grazing and/or trampling, the plant’s leaf area index and accumulated heat units are set back by the fraction of biomass removed.

The harvest and kill operation stops plant growth in the HRU. The fraction of biomass specified in the land cover’s harvest index (in the plant growth database) is removed from the HRU as yield. The remaining fraction of plant biomass is converted to residue on the soil surface.

The only information required by the harvest and kill operation is the timing of the operation (month and day or fraction of plant potential heat units). The user also has the option of updating the moisture condition II curve number in this operation.

The tillage operation redistributes residue, nutrients, pesticides and bacteria in the soil profile. Information required in the tillage operation includes the timing of the operation (month and day or fraction of base zero potential heat units), and the type of tillage operation.

The user has the option of varying the curve number in the HRU throughout the year. New curve number values may be entered in a plant operation, tillage operation and harvest and kill operation. The curve number entered for these operations are for moisture condition II. SWAT+ adjusts the entered value daily to reflect change in water content.

The mixing efficiency of the tillage implement defines the fraction of a residue/nutrient/pesticide/bacteria pool in each soil layer that is redistributed through the depth of soil that is mixed by the implement. To illustrate the redistribution of constituents in the soil, assume a soil profile has the following distribution of nitrate.

If this soil is tilled with a field cultivator, the soil will be mixed to a depth of 100 mm with 30% efficiency. The change in the distribution of nitrate in the soil is:

Because the soil is mixed to a depth of 100 mm by the implement, only the nitrate in the surface layer and layer 1 is available for redistribution. To calculated redistribution, the depth of the layer is divided by the tillage mixing depth and multiplied by the total amount of nitrate mixed. To calculate the final nitrate content, the redistributed nitrate is added to the unmixed nitrate for the layer.

All nutrient/pesticide/bacteria/residue pools are treated in the same manner as the nitrate example above. Bacteria mixed into layers below the surface layer is assumed to die.

The harvest operation will remove plant biomass without killing the plant. This operation is most commonly used to cut hay or grass.

The only information required by the harvest operation is the date. However, a harvest index override and a harvest efficiency can be set.

When no harvest index override is specified, SWAT+ uses the plant harvest index from the plant growth database to set the fraction of biomass removed. The plant harvest index in the plant growth database is set to the fraction of the plant biomass partitioned into seed for agricultural crops and a typical fraction of biomass removed in a cutting for hay. If the user prefers a different fraction of biomass to be removed, the harvest index override should be set to the desired value.

A harvest efficiency may also be defined for the operation. This value specifies the fraction of harvested plant biomass removed from the HRU. The remaining fraction is converted to residue on the soil surface. If the harvest efficiency is left blank or set to zero, the model assumes this feature is not being used and removes 100% of the harvested biomass (no biomass is converted to residue).

After biomass is removed in a harvest operation, the plant’s leaf area index and accumulated heat units are set back by the fraction of biomass removed. Reducing the number of accumulated heat units shifts the plant’s development to an earlier period in which growth is usually occurring at a faster rate.

The plant operation initiates plant growth. This operation can be used to designate the time of planting for agricultural crops or the initiation of plant growth in the spring for a land cover that requires several years to reach maturity (forests, orchards, etc.).

The plant operation will be performed by SWAT+ only when no land cover is growing in an HRU. Before planting a new land cover, the previous land cover must be removed with a kill operation or a harvest and kill operation. If two plant operations are placed in the management file and the first land cover is not killed prior to the second plant operation, the second plant operation is ignored by the model.

Information required in the plant operation includes the timing of the operation (month and day or fraction of base zero potential heat units), the total number of heat units required for the land cover to reach maturity, and the specific land cover to be simulated in the HRU. If the land cover is being transplanted, the leaf area index and biomass for the land cover at the time of transplanting must be provided. Also, for transplanted land covers, the total number of heat units for the land cover to reach maturity should be from the period the land cover is transplanted to maturity (not from seed generation). Heat units are reviewed in Chapter 5:1.

The user has the option of varying the curve number in the HRU throughout the year. New curve number values may be entered in a plant operation, tillage operation and harvest and kill operation. The curve number entered for these operations are for moisture condition II. SWAT+ adjusts the entered value daily to reflect change in water content or plant evapotranspiration.

For simulations where a certain amount of crop yield and biomass is required, the user can force the model to meet this amount by setting a harvest index target and a biomass target. These targets are effective only if a harvest and kill operation is used to harvest the crop.