5. Model testing against independent data

and the LAI. Larger values of the extinction coefficient have more light intercepted at a given

Measurement of light interception by the plant canopy is described at the website: https:// www.ars.usda.gov/plains-area/temple-tx/grassland-soil-and-water-research-laboratory/docs/ 193226/. To derive leaf area, biomass, and the extinction coefficient for Beer's law, LAI measurements are derived every 2 weeks during the active growing season via light measurements taken above and below the canopy between 10 a.m. and 2 p.m. on a clear day. The Decagon ceptometer (or something similar) is used to measure light as photosynthetically active radiation, since those are the wavelengths critical for photosynthesis. A random sample area for the area of interest is chosen where the forage is growing. The stand in the area for taking light measurements should not be trampled. Areas adjacent to where previous samples were taken should be avoided and should be ungrazed. A quadrat 0.5 m wide by the length of our light

If there are any non-targeted plants in or overshadowing our quadrat, they should be removed, or the quadrat should be relocated. Only canopy cover from targeted specie should be measured. The time of day, average phenology, and the average plant height in centimeters

a. Select an area under direct sunlight near our plots, and level the external sensor on the tripod. (Note: Whenever moving the tripod, level the sensor and calibrate again.)

b. Calibrate the light bar with the external sensor. Take at least 10 measurements with the light bar under direct sunlight. (Note: Make sure measurements are taken facing the sun, thus avoiding shading the light bar or the external sensor.) Record the shown average of

c. When taking light measurements under the canopy using the ceptometer, take at least six evenly spaced measurements in each quadrat near ground level. Record the average. Always take care to avoid biasing the sample in favor of more plants or more bare ground.

d. Finally, harvest plants, removing all plant material in the quadrat directly above the site

Repeat these steps at least three more times for a targeted plant species. For each set of measurements, make sure to measure plants on the same soil or ecological site. When returning to the general area for future measurements, select the same species to measure but not the exact same

Process plant material as soon as possible after sampling to avoid desiccation effects on leaf

a. When weighing the entire sample from field, if the entire sample is greater than 100 g, take a representative subsample. This is between 10 and 30% of the entire sample but no less than 100 g. Weigh and record the subsample weight. Make sure to select a subsample with the same proportion of green leaves, dead material, stems, and reproductive structure as

should be recorded. Light interception readings using the ceptometer are taken as:

LAI.

46 Forage Groups

bar (0.8 m) is reasonable for the sampling area.

all 10 measurements on the datasheet.

plant/plot area as previously measured.

the entire sample.

area.

where light was measured and place in labeled bag.

Following successful calibration of the ALMANAC model with field measured parameters, the model was applied to simulate forage yields across years and diverse environments in the U.S. For model testing, we used published forage yield data from Natural Resources Conservation Service, United States Dept. of Agric. 2017. Web Soil Survey. Available online: http://websoilsurvey.sc.egov. usda.gov/App/WebSoilSurvey.aspx.

Many common native and introduced grasses or grass mixtures in the U.S. have annual productivity values reported as USDA-NRCS ecological site productivity (for native forages) or NRCS crop productivity (for improved grasses) for many representative areas. As discussed below, once plant parameters for a particular forage are derived, they are tested on different soils in contrasting U.S. counties. The counties simulated are selected because they have soils with quantified annual biomass yields for the forage of interest (NRCS Web Soil Survey) (http://websoilsurvey.nrcs.usda.gov/app/HomePage.htm).

Total annual production of forages reported by NRCS are derived from end-of-season sampling on sites with closed canopy stands of the species of interest over 3 years or more. The NRCS procedure involves measuring dry matter biomass production above a 5-cm cutting height in at least 10 randomly selected plots at each field site. The specific soils for a location of interest can be downloaded as described above. Mean simulated forage yield over 10 years of real weather data can be compared to the reported annual production (from USDA-NRCS Web Soil Survey) for a site. The NRCS value of Animal Unit Month (AUM) is converted to Mg ha<sup>1</sup> (0% moisture) with a conversion factor assuming 700 lbs. (318 kg) of air-dried biomass (90% moisture) per AUM. Values for key plant parameters for the plant species of interest are derived from the field measurements described above.

#### 5.1. Examples of testing ALMANAC's simulation of forage yields

We have several published examples of testing ALMANAC's simulation of forage yields. The first was for several Texas range sites with native warm-season grasses [32, 33]. Next, we simulated old world bluestems (Bothriochloa Kuntze, Capillipedium Stapf, and Dichanthium Willemet) and buffelgrass (Pennisetum ciliare (L.) Link) in Oklahoma, Texas, and Mexico [18]. To evaluate the ability of the model to simulate introduced or improved grasses, we tested coastal bermudagrass (Cynodon dactylon (L.) Pers.) and bahiagrass (Paspalum notatum Flügge var. saurae Parodi) at several sites in Texas [19]. Western grasses in low-rainfall sites in Montana were simulated using parameters derived for some common native grasses there [20]. The cool-season forage "tall fescue" was simulated at several sites in several states where this grass is commonly grown [14]. In addition, creosote bush (Larrea tridentata [DC.] Cov.) parameters were derived the and model testing for its ability to describe competition of this woody species with forages in arid sites in western Texas [34].

Finally, legacy effects due to previous years' weather conditions and previous years' nutrient cycling need to be investigated. This has been studied with switchgrass [35], but needs more

Forage Yield Estimation with a Process-Based Simulation Model

http://dx.doi.org/10.5772/intechopen.79987

49

In this chapter, we described the ALMANAC model, including the process simulated, how to derive plant parameters for additional forage species, and how to validate using measured field data. Because of its accurate simulation of plant production, the water balance, and the nutrient balance, the model is capable of simulating a wide variety of environmental and management impacts on forage production, soil health, and conservation concerns, including nutrient and sediment losses. The model will be a useful and valuable tool for forage manage-

This material is based upon work supported by the Natural Resources Conservation Service, U.S. Department of Agriculture, Conservation Effects Assessment Project for Grazing Lands, under interagency agreement number 67-3A75-13-129. This work was also supported in part by an appointment to Agricultural Research Service administered by Oak Ridge Institute for Science and Education through interagency agreement between U.S. Department of Energy (DOE) and U.S. Department of Agriculture (USDA), Agricultural Research Service Agreement #60-3098-5-002. The authors are grateful to many university, USDA-ARS, and USDA-NRCS

, M. Norman Meki<sup>3</sup> and Mari-Vaughn V. Johnson<sup>4</sup>

1 U.S. Department of Agriculture, Agricultural Research Service, Grassland, Soil and Water

4 U.S. Department of Agriculture, Natural Resource Conservation Service, Temple, TX, USA

collaborators that are listed as coauthors in the cited references by the senior author.

extensive studies with diverse representative forages.

ment in pastures and rangelands in a wide range of conditions.

Authors have declared that no competing interests exist.

\*Address all correspondence to: jim.kiniry@ars.usda.gov

3 Texas A&M AgriLife Research, Temple, TX, USA

2 Oak Ridge Institute for Science and Education, Oak Ridge, TN, USA

\*, Sumin Kim<sup>2</sup>

Research Laboratory, Temple, TX, USA

7. Conclusions

Acknowledgements

Conflict of interest

Author details

James R. Kiniry1

Overall, the ALMANAC model predicted forage yields with reasonable accuracy, and hence when fully calibrated, the model can be used as an effective management tool to evaluate management practices that maximize forage yields, optimize inputs, and minimize negative environmental outcomes.
