Agriculture Damage Computation Proceedures

When flooding occurs in agricultural areas, interruptions to the planting, growing, and harvesting of crops result in economic impacts. The computational procedures and data requirements for calculating economic losses to agriculture due to flooding is described below.

Geospatial Distribution of Crops

The NASS Cropland Data Layer is a product that represents the geographic distribution of the types of crops throughout the entire United States. NASS provides the Cropland Data Layer in GeoTiff format, where each cell represents a crop type for a 30 meter grid cell. This data can be leveraged directly from the NASS API to streamline the collection of the type and distribution of crops in the study area. For information about the NASS Cropland Data visit the following site https://www.nass.usda.gov/Research_and_Science/Cropland/SARS1a.php.

The image above is taken from an example NASS CDL tif. In this example each yellow pixel represents corn and each dark green pixel represents soybeans.

The distribution of each agricultural category in the study area is used in combination with the planting, harvesting, and crop loss relationships in the calculation of agricultural flood damage.

Hazard Information

Damage to crops depends on the location, timing of the event and the duration of the event. The timing determines if the crops are planted, and how far the crop is in the production cycle. The duration defines how much damage the crop will sustain. Arrival time is expressed as time of the arrival of water above the soil (Time T1 in the image blow). Duration is defined in decimal days how long the water is above the ground surface elevation (GSE) (time T1 to T3 below) . The input must be geospatial to be intersected with the NASS CDL data.

Crop Schedule

The Crop Schedule defines a window of crop planting time and the time required for the maturity of the crop. A crop schedule has three parameters, the start time for the planting window, the end time for the planting window, and the duration in days for the crop to reach maturity. A crop schedule can be used to compute the crop damage case which helps determine how to apply the crop loss methodology.

The arrival time of the flood is compared to the start planting window and the days to maturity to determine if the flood started outside of the flood season. If the event started outside of the flood season, then the duration of the flood event is used to determine if the planting window was impacted. If the duration of the event covers the entire planting window either the crop is not planted at all or a substitute crop is planted. If the duration of the event delays planting, then the crop damages are computed in terms of the delayed planting losses.

If the arrival time of the event happens after the start of the planting window but before the days to maturity the crop is determined to be impacted, and the duration is used to determine the significance of the impact.

Crop schedules can have planting windows in one calendar year and mature in the next calendar year, but the base implementation of the crop schedule requires that the days to maturity be less than 365.

Crop Loss Function

The crop loss function describes the loss of expenses that will not be recouped by the farmer. It is based on the loss of crop or the reduction in the harvest for the crop based upon the arrival time and duration of the event.

Damage to crops is dependent upon the value added by the farmer to the field at the time of flooding, and the vulnerability of the crop at that time to flooding. The driving damage parameters are duration and the timing of the event.

Crop Budget Data

In order to determine the exposed value across time the user inputs a Production Function. The Production Function (P) is expressed as the sum of the fixed costs (FC) such as rent or land taxes and monthly variable costs (MVC) and is time dependent. The Production Function, specifically the monthly variable costs, are crop dependent. The crop production function information typically comes from local Cooperative Extension System Offices, which can be found on the USDA's National Institute of Food and Agriculture (NIFA)'s website. The final components of the crop planting data required are the dates of the first possible and last possible plantings. The crop planting data is used to create a curve see the Seasonally Based Value image below, that represents the cumulative cost of the inputs in the field at any point during the year from the first plant date.

The cumulative cost is converted to a percentage of the maximum total cost of the inputs in the field to calculate the potential losses at any point during the year. The maximum total cost input into the field is not necessarily equivalent to the total cost experienced by the farmer for the crops pulled from the field. First, the harvest cost and shipping costs need to be added to all of the inputs to get the full cost to produce the crops. The computation uses a proration of the total cost input less the harvest costs as a proxy for an exposed value.

Variable

Name

Description

$\begin{array}{l}\displaystyle P_t(c)\end{array}$

Production Function

The cumulative fixed and variable costs.

$\begin{array}{l}\displaystyle c\end{array}$

Crop

example: Corn, Soybeans, etc...

$\begin{array}{l}\displaystyle MVC_t(c)\end{array}$

Monthly Variable Costs

Costs of growing the unique crops in the field. example: detasseling corn.

$\begin{array}{l}\displaystyle FC_t(c)\end{array}$

Fixed Costs

Rent and land taxes.

$\begin{array}{l}\displaystyle EV(P_t(c),at)\end{array}$

Exposed Value

The cumulative fixed and variable costs at the arrival time of the event

$\begin{array}{l}\displaystyle at\end{array}$

Arrival Time

The time inundation begins

Crop Value

Crop characteristics are necessary to compute the appropriate reduction in crop value due to the costs associated with harvesting. The required input characteristics are harvest date, harvest cost, yield, unit price, and percentage of total crop value lost due to late planting. Crop values should reflect prices with subsidies removed for federal benefit to cost analysis.

$\begin{array}{l}\displaystyle P_t(c)= MVC_t(c)+FC_t(c)\end{array}$

The Exposed Value (EV) is dependent upon the arrival time of the event and the production Function

$\begin{array}{l}\displaystyle EV(P_t(c),at) = \sum_{t=0}^{at} P_t(c)\end{array}$

To illustrate how the seasonally based value changes with time an example plot below is provided. In the figure below the seasonally based value is expressed as a ratio of seasonal value to total value.

As shown in the figure above, if a crop is planted later in the season, the crop has a different value curve (e.g., red curve in the figure above). For seasonally based value changes, it is assumed that if the flood does not interrupt the farmer, the crops will be planted by the first plant date. Thus, late planting will only occur if the flood start plus the duration ends before the late plant date but after the first plant date. This seasonally based value is intended to reflect that farmers may adopt different schedules for application of watering, fertilizer, and other processes in raising the crop; or, the farmer may choose to grow another crop altogether.

The Loss (L) is a function of the crop type the arrival time and the duration. The loss is computed by taking the exposed value and multiplying it by the damage percentage computed by the crop type, arrival time of the flood and the duration of the event. The duration defines which seasonal loss function to choose and the arrival time defines which percentage loss value to pick off of the duration loss function family.

$\begin{array}{l}\displaystyle L(c,at,d) = EV(P_t(c),at)*D(c,at,d)\end{array}$

Damage fluctuates based on duration and the maturity of the crop as shown in the example relationships in the figure below. Some crops can better withstand short duration floods if the crops are out of critical development stages.

As shown in the figure above, longer duration flooding would yield larger losses, but more mature plants would be more robust to damage. This input (planting, harvesting, and crop loss relationship inputs) is provided by the user, and must reflect their expectation of how plants will react to different durations of flooding throughout the growing season.

Computational Results

Results from the agricultural damage computations include: a point based shapefile specifying the crop type; arrival time; duration; location; and total damage for each damaged grid cell. These results can be aggregated by crop type, or by any polygon loaded as a boundary shapefile.