Method and System for Remote Sensing and Limited Field Sampling to Prescribe and Deliver Soil Amendments Targeted Discretely Across Cultivated Fields

Abstract

A method and system using remote sensing to estimate soil amendment prescriptions for portions of an agricultural field. The prescriptions are based upon limited soil sampling data that are extrapolated across the entire field according to the greenness of the crop. The prescriptions are delivered by mechanical device or irrigation system that is equipped with a global positioning system.

Description

RELATED APPLICATIONS

This application claims the benefit of U.S. Patent Application Ser. No. 61/490,499, filed May 26, 2011, and U.S. Patent Application No. 61/486,193 filed May 13, 2011. The provisional patent applications listed above are incorporated herein by reference in their entireties.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates generally to the field of precision application of soil amendments based on forecast crop requirements.

2. Background

The method and system of the present invention is related to the remote sensing science and irrigation system technology disclosed in my co-pending application Ser. No. ______. That application describes the use of novel evapotranspiration formulas and applications that tailor water usage to the particular crop needs at various times during the growing season and in changing environmental circumstances by forecasting and predicting weather conditions, soil water content, and other factors.

Enhancing agricultural efficiency is a means to help reduce world hunger, to improve the United States international balance of payments and to conserve precious natural resources. This efficiency calls for the use of soil amendments, long-known and applied in a variety of manner and methods. However, the prior art methods are not well coordinated with the sophisticated techniques of precision irrigation control. Nor do known techniques for prescribing and delivering soil amendments properly harness remote sensing methods, such as Earth Observatory Satellite (EOS) data including scene statistics and pixel analysis that are available in today's sophisticated resources for collecting and distributing data that may be useful in achieving precise soil amendment applications. In particular, the prior art methods fail to precisely deliver prescribed soil amendments to different crops in small regions or zones of a field.

Recent skyrocketing costs for fertilizer are following the trend for the general increase in prices of energy and minerals. This has placed a strong incentive for the agriculture community to conserve fertilizer and apply only what is needed by a specific field. Soils across most fields are variable, sometimes highly variable, and zones may exist where there are chemical and physical constraints for good crop growth. In contrast, other portions of the field may provide adequate or even optimal conditions. What is needed is a system to provide soil amendments targeted to individual portions or zones of fields so that each portion of the crop receives exactly the amendments that will enable it to reach its potential.

Fertilizer corrects deficiencies in soil nutrients and the health of a crop and its productivity are based upon the most limiting nutrient. Justus von Liebig's Law of the Minimum, taught in agronomy classes, states that a crop's yield is proportional to the amount of the most limiting nutrient. The science of soil amendments is the science for correcting such limiting nutrients. The term “amendment” is used here to mean both fertilizer, added mineral nutrients required by plants, and minerals that provide correction for some condition in the soil that may impede plant growth. An example of the latter would be the use of elemental sulfur that is added to soils to correct problems with sodicity or polymers added to soils to improve structure, porosity and aeration.

Determining the precise amendment needs for a field may require many separate selected sampling points since not all portions of a field require the same amendments. In prior art systems, sampling costs for determining precise amendment prescriptions are very high because sampling must be carried out at numerous locations, often in the form of grids of many sample points. Hence, costs of sampling compete directly with the savings that can accrue through conservation of the amendment within targeted applications. What is needed is a method whereby limited sampling data obtained at carefully selected points are used for prescribing amendments across an entire field so that each zone is targeted with nutrients in the amounts required. Thus, low sampling costs combined with cost savings through using less material amendments and enhanced productivity would serve to make this method of amendment prescription highly attractive.

Individual prescriptions for amendments in some cases are made based upon the known soil relationships within a region, thus providing a more generalized prescription, or specifically interpreted and targeted for fields by agronomy experts. However, the lack of precision or specificity will not enable the efficiency and advantages desired. The prescription can be transferred via internet and wireless technology to the irrigation controller.

There is therefore a need to prescribe and deliver the correct soil amendments through the combination of remote sensing methods, geographic information and soil sampling of the field. Together, these methods and techniques prescribe the amendments required to enable a crop to deliver its maximum potential yield. In order to properly and efficiently implement these methods and techniques for determining optimized amendment prescription are needed large areas of farmed land to enable such a service to be cost-effective. Such inputs are ideally made using EOS data delivered through the medium of Internet and wireless connectivity.

SUMMARY OF THE INVENTION

In one embodiment of the invention there is provided a method for applying soil amendments to an agricultural field or portions thereof through a delivery system which includes the steps of obtaining EOS data including scene statistics and pixels for the field containing an existing crop, calculating NDVI* per pixel utilizing satellite scene statistics, identifying and ranking NDVI* values for selected portions of the field, obtaining soil samples at the selected portions of the field based on the ranked NDVI* values, analyzing the soil samples and prescribing soil amendments based on interpolation of requirements at high and low NDVI* values at the selected portions of the field, preparing an amendment mixture based on the prescribed soil amendment, and delivering the prescribed soil amendment mixture to the selected portions of the field based on the NDVI.* sample analysis values.

In a variation of the embodiment of the above-described method for applying soil amendments to an agricultural field, the steps include obtaining EOS data, calculating greenness based on that data, determining the maximum and minimum greenness values for selected portions of the field, obtaining soil samples in at least two selected portions of the field, analyzing the soil samples and prescribing soil amendments based on interpolation of the greenness values for the selected portions of the field, preparing soil amendment mixture based on the sampling results that were guided by greenness values selected portions of the field, and delivering the prescribed soil amendment mixture for the selected portions of the field based on interpolated values of sample greenness.

An embodiment of the invention comprises a system for applying soil amendments to an agricultural field including a computer system for acquiring and storing EOS data including scene statistics and pixels for the field containing existing; crops, an algorithm for calculating NDVI* per pixel based on the scene statistic, identifying selected portions of the field by GPS, an algorithm for ranking the NDVI* values for the selected portions of the field, physically obtaining soil samples at the selected portions of the field based on the maximum and minimum NDVI* values, an algorithm for analyzing the soil samples and prescribing soil amendments based on interpolation of the requirements at the high and low NDVI* values at the selected portions of the field, a container for mixing an amendment based on the prescribed soil amendment, and a mechanical apparatus for delivering the prescribed amendment mixture to the selected portions of the field based on the NDVI* sample analysis values.

DESCRIPTION OF THE DRAWINGS

FIG. 1 presents a graph that illustrates how soil amendment prescriptions are made in accordance with one embodiment of the present invention;

FIG. 2 is a flowchart of one embodiment of the method of the present invention based on remote sensing data calculations and field sampling; and

FIG. 3 is a flowchart of the embodiment of the method of the present invention that delivers the mixed prescribed amendments to selected portions of the field according to NDVI* values and geoposition information provided by a Global Positioning System (GPS).

DETAILED DESCRIPTION

An exemplary embodiment of the method of this invention was developed for use with center pivot irrigation systems, however, the invention is applicable to mechanized delivery or other types of irrigation systems that may be actively controlled, whether as blocks of sprinklers or at the level of the individual sprinkler. Controlled delivery systems and methods may comprise multiple sprinklers, traveling tricklers (hoses that hang from a center pivot), drip irrigation, or by direct application including delivery in a dry or liquid form by a tractor with suitable amendment spreader.

A crop's health and productivity are based, directly, on its greenness that is dependent upon its mineral nutrition and long term water status. Water status are managed adequately by replacing the water that the plant uses, that is, always providing sufficient water to meet the plant's demand. The relationship for soil nutrition is far more complex due to soil chemical and physical properties. Also, there is a suite of many required minerals for maintaining proper crop health and/or correcting deficiencies in any minerals that are the key to enabling high crop yields.

A crop's greenness is determined by the amount of chlorophyll in the canopy that supports photosynthesis and yield. Crop greenness is accurately assessed using a modification of the Normalized Difference Vegetation Index (NDVI) that represents the range of plant greenness on a scale from zero to one. The index, called NDVI*, also corrects for the influence of soil background and atmospheric scatter and attenuation, factors that would otherwise dilute the accuracy and quality of the greenness estimate. This index exists in the prior art specifically to enable comparison among dates and years for the same location. NDVI* is adapted within this embodiment specifically as it applies to determination and prescription for amendment of some limiting soil property and targeting this amendment differentially across a field or many fields.

Sampling in only a few locations within an individual field, and at least within the poorest and the best locations for crop growth, allows development of a high and low amendment prescription, perhaps for a suite of plant nutrients, for example nitrogen, phosphorous, potassium and micronutrients. These poorest and best portions of a field will have their maximal expression in the canopy of the crop. The best portions will be more verdant and the poorest will be the least. These locations are identified spatially using remotely-sensed data such as that acquired by satellite and aircraft, and using methods of remote sensing analysis and Geographic Information System (GIS) technology.

Through the use of remote sensing analysis, results from as few as two sampling locations, the poorest (minimum) and the best (maximum) selected locations in the field, enable generation of an amendment prescription for the entire field through interpolation. Such poorest and best expressions of the crop condition, called endmembers, are be identified using NDVI* as a surrogate measure of greenness. Depending upon known relationships for regional soils, the mineral needs of all locations in the field is estimated by creating a curve to interpolate between the endmember points. This relationship defines mineral needs as a dependent variable of NDVI*, a remotely-sensed estimator of greenness (FIG. 1).

The curve shown in FIG. 1 is a straight line, that is, the simplest curve useable for interpolation. Given sufficient knowledge of the regional soils, other curves may be equally valid and are easily fitted, especially if several additional points are also sampled within locations where the crop is performing moderately, that is, intermediate the maximum and minimum end member points. Such locations are used to more precisely define the curve. Only two endmembers are shown in FIG. 1; more samples are obtained to ensure the validity of the curve. Also, only one hypothetical amendment is shown, when, in actuality many different amendments are potentially be determined and prescribed from the soil test results using this embodiment of the invention.

The amendment prescription for the entire field are be made based upon the interpolation method shown in FIG. 1, because greenness expressed as NDVI* is measured and known throughout the field from the remotely sensed dataset.

Crop amendments are often prescribed based upon weigh (or other) crop yield data gathered during harvest that is paired with geoposition information provided by a GPS. In contrast, rather than basing the prescription on data from the previous year this invention develops the prescription for the current year and so, can positively influence the outcome of this year's crop. This is an important distinction because the crop may have been switched from last year and the nutrient requirements may be very different from last year's crop that may not have been sensitive to the requirements of this year's crop. An example of this would be switching from a soybean crop that fixes and preferentially uses) nitrogen, to a corn crop that requires lots of nitrogen.

The NDVI*based spatial prescription of amendments are stored electronically and are used to guide the rate of amendment application given known geopositioning provided by a global positioning system. Such applications are made mechanically by tractor or provided within irrigation water, a method called chemigation or fertigation. Chemigation is an attractive alternative means of application because it enables the irrigation system to be the vehicle for moving the amendments directly to where they are needed within a field. As an example, a center pivot system with variable rate zones or sprinkler heads is easily adapted for chemigation as prescribed by the embodiments of the present invention.

In the above-referenced co-pending application, crop greenness, also determined using NDVI* is used to prescribe irrigation amounts to supply plants with the forecasted amount of water that they need. A center pivot system with variable rate sprinklers is an ideal delivery method for applying both the irrigation and chemigation (this) invention. For targeted irrigation delivery, water is parsed at higher rates for portions of the field with higher NDVI* because greenness is proportional to water use. For chemigation, however, this relationship may be inverted and the greenest, high NDVI*, portions of the field would need to receive less of the prescribed amendment because they are already operating at or near optimum while the least green portions of the field are would receive more amendments to correct for sonic deficiency that constrains production.

An irrigation system that is adapted for chemigation can have mixing tanks for dilution of the prescribed mixture. The mixture, for example with multiple plant nutrients, is formulated for the field portions, mixed in concentrated liquid form and conveniently transported to the field by tanker load. The concentrated form is then metered to mix with water to the required dilution level that is then applied to the field. The system adapted for delivery then meters the prescribed application to individual portions of the field based upon greenness.

Crop Greenness and NDVI

Crop greenness are portrayed by vegetation indices that combine red and near infrared light. Greenness, a term that is used here by convention, makes sense to our visual world, but paradoxically is most accurately determined using red light. In terms of what is actually measured by vegetation indices, the index would more properly be called “lack of redness”. Plants only appear green because chlorophyll strongly absorbs red light in the act of photosynthesis; green is simply what is not used and is therefore reflected and visible. Crop canopies reflect highly in the near infrared, as do many background surfaces, a common example being dry soils. However, the ratiometry of red versus near infrared create highly useful indices of plant activity that are inversely proportional to the red signal. The normalized difference vegetation index (NDVI; Equation 1) is the most commonly used among these indices.

$\begin{array}{cc}NDVI=\frac{\mathrm{NIR}-\mathrm{Red}}{\mathrm{NIR}+\mathrm{Red}}& \mathrm{Equation}1\end{array}$

• • Where NIR is the near infrared band and Red is the red band within the digital data commonly measured by sensors borne by either aircraft or EOS platforms.

In its role as an estimator of canopy greenness, NDVI is beset with problems with accuracy that are due to soil background and atmospheric aerosol effects of scatter and attenuation. These influences combine to dilute the vegetation signal it is designed to portray. The accuracy for NDVI to portray hydrologic responses is enhanced by conversion to NDVI* that calibrates the canopy from zero to one to represent the full range of vegetation greenness from none to saturated. This index has been shown to outperform all other available vegetation indices in use in the field of remote sensing. The calculation of NDVI* corrects for the error-inducing effects from soil background and atmospheric aerosols (Equation 2),

$\begin{array}{cc}NDVI^{*}=\frac{NDVI_i-NDVI_0}{NDVI_S-NDVI_0}& \mathrm{Equation}2\end{array}$

• • Where NDVI_i is the measured NDVI for the ith pixel, NDVI_S is the saturated value for NDVI, and NDVI₀ is the NDVI value representing bare soil.

NDVI* is calibrated using scene statistics and hence, requires no specific ground target or ground-based measurements. For calculation of NDVI*, there are times of the year where a maximally verdant target suitable for setting NDVI_S will be missing, for example, for periods outside of the main growing season, e.g., fall, winter and spring. Hence, the NDVI_S value is simply chosen as an empirical constant, since the peak value for non-cloudy scenes in an atmosphere relatively clear of non-gaseous aerosols occupies a known range that is determined empirically. The choice of the saturated NDVI value as a median for this range to represent NDVI_S, produces only insignificant influence upon the resulting NDVI* values. These relationships have been determined for use with Landsat™, however, calibrating other EOS sensors to be mathematically equivalent to Landsat™ is accurately accomplished using linear regression analysis of extracted pixel values.

For making amendment prescriptions, the relational information provided by virtually any vegetation index could be used, since it is the differences in production, expressed as greenness, that are of interest. NDVI* is specified here, specifically, because it is the index that will be used for guiding amendment application to meet crop needs and because it is calibrated in a such a manner that it permits comparison from year to year to cross check the efficacy of the treatments on any region of a field or fields. Thus, NDVI*, calculated in the manner described here, provides standardization.

For comparison of EOS data across seasons, the effect of variable solar angles and distance to the sun are corrected through the calculation of reflectance: the ratio of reflected light to the incident light. Reflectance is often calculated at the top of the atmosphere; such values are further corrected to at-ground reflectance by means known to those with ordinary skill in the art. NDVI* functions as an approximation of at-ground NDVI but also corrects for the influence of the soil background that generally raises NDVT for zero vegetation cover to some value greater than zero.

NDVI and NDVI* can also be calculated from the digital number (the uncorrected output) from the sensors, or from radiance which is the calculation of the relative light received by the sensor, irrespective of the incident light. Papers describing these calculations are available in the published literature. Some appropriate EOS data (having red and near infrared bands) are available at resolutions of several meters. However, EOS datasets covering large geographic areas (many tens of thousands of square miles), may only be available in moderate resolution (20-30 m/pixel). The spatial resolution of these data are enhanced by using kriging that can bring the spatial resolution down to several meters—a scale appropriate for controlling this application.

Spatial datasets in raster form permit the mathematical manipulation that is described here for prescribing amendments. The use of raster-based calculations enables serving the farming community across an immense geographical area.

The flowcharts, FIGS. 2 and 3, provide a means to track the embodiment through the various steps, from EOS data to the application of the amendment to the field. The conventions used in FIGS. 2 and 3 are

j_ith refers to the jth day. The jth day for EOS data is the day of the overpass.

i_ith refers to the ith pixel.

n_ith refers to the nth field.

The process begins at Start (S100) when satellite data for any desired jth day is obtained and stored at block S102. Initial development of the crop is preferably required for the most advantageous application of the embodiment of the invention and so the crop first needs to develop over about the first month, or so, in order for it to express the limitations imposed by the soil environment. Amendments are applied as many times per season as is necessary, though one application is generally sufficient.

From block S102 the workflow passes to block S104 that calculates the NDVI* for each ith pixel using scene statistics as described in the literature (for example Baugh, W. M. and Groeneveld, D. P., “Broadband vegetation index performance evaluated for a low-cover environment,” International Journal of Remote Sensing, 27:4715-4730 (2006)). From 5104, the workflow splits to block S106 that provides for the use of a geostatistical procedure that interpolates the NDVI* values (as pixel centroids) across cropped field n. Kriging will be used if the EOS data are of moderate resolution (20-30 m/pixel). The kriged values provide for metering amendments at a finer scale than the coarse blocks that would otherwise be defined by pixels. Fields of geo-referenced NDVI* values are sent to the controller at block 5108 that meters the amendment onto the field.

From the split at S104, the other path proceeds to S110 to provide input for field sampling by identifying high and low values in the geo-referenced NDVI* for field n. These values are used to select sampling locations in field n at S112 to determine the amendments that are needed to make crop growth achieve the desired level of productivity. Although the sampled soil beneath the high crop expression may represent the high for field n, the crop may not be operating at the desired level of production and so, this peak value for field n may also require additional amendments. This is shown in FIG. 1, where the hypothetical peak NDVI* is 0.9 (defined zero amendment requirement), however, the measured high for field n is less than that. To achieve the 0.9 NDVI*, FIG. 1 indicates that an addition of about 0.1 (application scaled zero to one) would be needed for the field n high value to become optimal. This optimum is optimal only with regard to production, without taking into account efficiency.

The FIG. 1 hypothetical presents an opportunity for the farmer to interact with the flow of information to choose the percent of optimality as a target and thereby influence the efficiency of the crop. The choice for percent of optimality is to either run close to optimal, in which case input costs for fertilization and water run high, or at some fraction below optimal that does not require such high input costs. These and other considerations are made at the sample analysis stage in block S114. The prescription is passed to the amendment mixing stage at S116.

FIG. 3 receives input from FIG. 2 that enables the prescribed amendment to be delivered. The data that was prepared to enable metering application at fields of geo-referenced values of NDVI* enters at S120. The prescription for mixing the amendments is received at block S140. The prescribed amendment is mixed in a tank or similar vessel and then delivered to field n at S142 by the methods previously described.

Both the portions of the fields of NDVI* determined soil conditions and the amendment mixture are regulated by the controller at S122 that receives input for the geoposition within field n according to GPS data at S150. The controller as described within the example embodiment controls the prescription and mixture of amendments and subsequent application to the field, in one embodiment, in a diluted liquid form through individually controlled sprinklers. Alternatively, delivery may be in the form of a tractor with sprayer, or if in solid form as a spreader that is guided by GPS to distribute the amendments according to the prescription tied to NDVI*. This process precisely applies amendment to correct nutrient problems for less-green portions of field n at step S124. Once the application is complete the process ends at S126. Data from each of these steps, including soil sampling, NDVI*, amendment prescriptions, etc. are stored electronically and referenced by geoposition for future use.

Various preferred and other embodiments of the invention have been described but it will be understood by those of ordinary skill in the art that modifications may be made without departing from the spirit and scope of the invention of the system and method. For example, a wide variety of soil amendments and nutrients other than those mentioned above may be employed depending upon the soil and crop in the field. Various delivery methods and mechanical systems may be employed for delivery of the prescribed amendments as determined by the variety of data from various sources as described above. Accordingly, it is to be understood that the invention is not to be limited by the specific illustrated and described embodiments but only by the scope of the appended claims.

Claims

1. A precision method for applying soils amendments to an agricultural field or portions thereof through a delivery system comprising the steps of:

obtaining EOS data including scene statistics and pixels for the field containing an existing crop;

calculating NDVI* per pixel satellite scene statistics;

identifying and ranking NDVI* values for selected portions of the field;

obtaining soil samples at the selected portions of the field based on the ranked NDVI* values;

analyzing the soil samples and prescribing soil amendments based on interpolation of requirements at high and low NDVI* values at the selected portions of the field;

preparing an amendment mixture based on the prescribed soil amendment; and

delivering the prescribed soil amendment mixture to the selected portions of the field based on the NDVI* sample analysis values.

2. The method of claim 1 wherein the delivery system comprises a water irrigation system and the soil amendments are water-soluble and distributed through sprinklers of the irrigation system.

3. The method of claim 2 wherein the water irrigation system is precisely controlled by a field located controller in communication with a central computer system that gathers the data and performs the calculations to prescribe the soil amendment mixture.

4. The method of claim 3 wherein the water irrigation system includes multiple sprinklers that are controllable individually or in blocks.

5. The method of claim 3 wherein the water irrigation system is a center-pivot mechanical irrigation system.

6. The method of claim 1 additionally including the step of geospatially interpolating the EOS data before calculating the NDVI* for a pixel.

7. The method of claim 1 wherein the EOS data is adjusted during the growing season to reflect the variable solar angle and the distance from the sun.

8. The method of claim 1 wherein GPS location is used to determine the portion of the field where the sampling takes place.

9. The method of claim 1 wherein the soil amendments are selected from the group comprising nitrogen phosphorous, potassium, and micronutrients, or chemical substances that change soil properties limiting to plant growth.

10. The method of claim 1 wherein the NDVI* corrects for the soil background and atmospheric scatter and attenuation.

11. A system for applying soil amendments to an agricultural field or portions comprising:

a. computer system for acquiring and storing EOS data including scene statistics and pixels for the field containing an existing crop;

an algorithm for calculating NDVI* per pixel based on said scene statistics;

identifying selected portions of the field by GPS;

an algorithm for ranking the NDVI* values for the selected portions of the field;

physically obtaining soil samples at the selected portions of the field based on the maximum and minimum NDVI* values;

an algorithm for analyzing the soil samples and prescribing soil amendments based on interpolation of the requirements at the high and low NDVI*values at the selected portions of the field;

a container for mixing an amendment based on the prescribed soil amendment; and

a mechanical apparatus for delivering the prescribed amendment mixture to the selected portions of the field based on the NDVI* sample analysis values.

12. The system for applying soil amendments to an agricultural field of claim 11 wherein the EOS data is adjusted during the growing season to reflect the variable solar angle and the distance from the sun.

13. The system for applying soil amendments to an agricultural field of claim 11 additionally including GPS location data to determine the portion of the field where the sample is taken.

14. A precision method for applying soils amendments to an agricultural field or portions thereof through a delivery system comprising the steps of:

obtaining EOS data including scene statistics and pixels for the field containing an existing crop;

calculating greenness utilizing the satellite scene statistics;

determining the maximum and minimum greenness values for selected portions of the field;

obtaining soil samples at the two selected portions of the field;

analyzing the soil samples and prescribing soil amendments based on interpolation of the greenness values for the selected portions of the field;

preparing soil amendment mixture based on the greenness values of the selected portions of the field;

delivering the prescribed soil amendment mixture to the selected portions of the field based on the sample greenness analysis.

15. The method of claim 14 wherein the delivery system is precisely controlled by a field located controller in communication with a central computer system that gathers the data and performs the calculations to prescribe the soil amendment mixture.

16. The method of claim 15 wherein the water irrigation system includes multiple sprinklers controllable individually or in blocks.

17. The method of claim 14 wherein the delivery system is a center-pivot mechanical irrigation system.

18. The method of claim 14 wherein the delivery system comprises travelling tricklers.

19. The method of claim 14 wherein the delivery system is a drip irrigation system.

20. The method of claim 18 wherein the delivery system comprises a mechanical tractor with a dry soil amendment spreader.