Introduction to SMAP
Soil moisture strongly influences plant growth and agricultural productivity, especially during water shortages and droughts. Crop conditions shift quickly due to changes in soil moisture, so high-frequency, high-resolution geospatial data—especially during growing seasons—are critical for food security, assessments of crop yields, and informed decisionmaking in agricultural production and commodity markets. However, no global in situ network currently exists for monitoring soil moisture. Predictions are model-based with relatively low resolution and large uncertainties.
The Soil Moisture Active Passive (SMAP) mission, launched by the National Aeronautics and Space Administration (NASA) in 2015, is an Earth-observing satellite that measures the amount of water in the top two inches of soil everywhere on Earth’s surface. The soil moisture information collected by SMAP will dramatically improve model predictions and help inform drought monitoring and crop management, among many other efforts.
What is the societal value of SMAP information?
SMAP cost $916 million to design, develop, launch, and operate. What can society expect in return for this investment? Research is still underway to develop improved model predictions using information collected by SMAP, but tools to quantify the societal value of SMAP information already exist. In this paper, Roger Cooke and Alexander Golub present a market-based method that can be used for this purpose.
In their study, Cooke and Golub link the societal value of SMAP information to its ability to reduce uncertainty in predicting soil moisture, and therefore, its ability to reduce uncertainty in weather-related components that influence corn and soybean yields. Their approach exploits the fact that uncertainty about weather (including soil moisture conditions) is reflected in agricultural risks that are actively traded in large markets. By examining prices in these markets, the authors are able to quantify the value of reducing uncertainty in crop yields, and, therefore, the value of improvements in soil moisture predictions afforded by SMAP data.
Estimating the value of reducing uncertainty by looking at futures markets
Markets have mechanisms for trading risks between those who will pay to remove a risk and those who will accept payment to assume that risk. An example is the corn futures market. Farmers need to purchase seed corn before planting. Because the future price of seed corn is uncertain, farmers are willing to pay a premium in September to guarantee that they can purchase a given quantity of seed in April of the following year at a given price. The farmer can obtain this guarantee by purchasing a futures option. If the actual price of seed corn in April is below the price that was paid for the futures option in September, the seller of the futures option makes money, as they can buy seed corn at the market and sell for a higher price to the farmer. If the actual price of seed corn is above the price that was paid for the futures option in September, the seller loses money. In any event, the farmer purchases the seed corn at the price anticipated in September.
How does the price of the futures option relate to uncertainty regarding the future price of seed corn? Intuitively, as the uncertainty of the April price increases, so does the price of the futures option. The farmer is willing to pay more in September to secure a price for seed corn, and the seller will need to be paid more in order to take on the additional risk associated with greater uncertainty in future seed corn prices.
Bottom line: Examining prices in futures markets can reveal people’s perception of uncertainty and tell us how much they are willing to pay to reduce that uncertainty.
Applying the framework to the corn and soybean futures markets
The total annual volume of corn produced in the United States is worth about $53 billion per year; for soybeans, about $42 billion per year. As with any other commodity, corn and soybeans are primarily traded on the futures market. Prices for corn and soybeans exhibit relatively high volatility, and low predictability of commodity prices is an important factor of risk for commodity consumers and producers.
Cooke and Golub follow a two-step process to quantify the value of reducing weather uncertainty using data on corn and soybean futures. First, they use pricing models that translate options prices into quantifications of uncertainty. Several such option pricing formulas are described in the literature; the authors use relatively simple models called the Polya approximation and Bachelier formula. Second, the authors account for the fact that prices for corn and soybeans depend on both weather and on the state of the economy. Using fluctuations in the Dow Jones Index to represent uncertainty surrounding the overall economy, Cooke and Golub isolate the contribution of weather uncertainty to uncertainty in prices for corn and soybeans.
Because scientists are still in the process of incorporating SMAP data in predictions of weather-related components that influence corn and soybean yields, it is unclear how much weather uncertainty will be reduced by SMAP. However, the framework presented by Cooke and Golub is flexible and can estimate the value of a wide range of possible reductions in weather uncertainty.
For example, suppose that the weather-related uncertainty was reduced by 30 percent thanks to SMAP data. This reduction in weather-related uncertainty will result in a reduction in the uncertainty for corn prices. Using the market-based method by Cooke and Golub, given annual US corn production of about 15 billion bushels, the total value of information associated with a 30 percent reduction in weather uncertainty is about $0.9 billion. The value of information for soybeans, of which about 4.2 billion bushels are produced annually in the United States, is about $0.54 billion. Thus, the total value of information for corn and for soybeans attributed to a 30 percent reduction in weather uncertainty is about $1.44 billion per year. If the contribution of SMAP data in reducing weather uncertainty turns out to be greater than or less than 30 percent, the Cooke and Golub framework can provide the value of information associated with this actual reduction in uncertainty as well.
Download the working paper to read more.