Evaluation of APEX modifications to simulate forage production for grazing management decision-support in the Western US Great Plains

doi:10.1016/j.agsy.2021.103139

Agricultural Systems

Volume 191, June 2021, 103139

https://doi.org/10.1016/j.agsy.2021.103139 Get rights and content

Highlights

•
Rangeland models can serve as decision-support tools after calibrating against experimental data.
•
CStudy evaluated APEX improvements related to grazing impacts on forage production impacts and a new rotational grazing modification.
•
APEX was able to simulate the responses of forage production to grazing management and soil type.
•
Current rotational sequences based on stakeholder criteria were optimal for grazing management in the semi-arid region.
•
APEX can assess grazing decisions on forage production and improve grazing management in semi-arid environments.

Abstract

Context

Understanding how grazing management decisions influence the productivity and composition of rangeland plant communities is essential for the development of effective strategies to sustainably produce multiple ecosystem goods and services. Informed with experimental measurements, simulation models can advance our understanding and stewardship of rangeland ecosystems.

Objective

Our main objective was to evaluate the APEX (Agricultural Policy/Environmental eXtender) plant growth modules and grazing animal selectivity in simulating forage production using experimental data collected from both traditional season-long grazing and adaptive rotational grazing management on western rangelands. Specifically, we evaluated APEX's capability to simulate forage productivity and its response to soil types and climate conditions under grazing management options.

Methods

Capitalizing on a comparative field study with 20 large pastures (> 123 ha each), APEX modifications were evaluated by comparing simulated forage production with experimental data. The field study evaluated traditional grazing (season long grazing on a single pasture) and an alternative grazing system that utilized collaborative adaptive rangeland management with stakeholders engaged in decision making (such as when and where to rotate a single herd). APEX was modified to include rotational grazing based on a user-defined sequence and automatic rotational grazing based on user-defined forage grazing limits and minimum/maximum grazing durations.

Results and conclusions

The APEX model was able to simulate the relative differences in forage production between grazing treatments, across years, and among soil types; however, APEX underestimated forage production in 2015 and 2017 due to overestimating drought stress for the warm season perennial grass functional group. Simulation of grazing management scenarios showed that the collaborative adaptive management decision criteria resulted in grazing durations that produced more forage than consistent 7- or 14-day rotation intervals.

Significance

These modifications were needed to capture the complexity of semiarid environments and thus enhance APEX to better assess grazing management decisions on forage production in regions such as the Western US Great Plains.

Graphical abstract

Introduction

Rangeland models can be used to evaluate risks and decision impacts of alternative management strategies under different circumstances (Derner et al., 2012; Ma et al., 2019). Previous studies showed that process-based rangeland models (e.g., SPUR, Stout et al., 1990; GPFARM-Range, Andales et al., 2005; Andales et al., 2006; Qi et al., 2012; Fang et al., 2014; GRAZPLAN, Moore and Ghahramani, 2013) can simulate elementary grazing impacts such as seasonal responses of plant growth (total biomass) and animal weight gain under varying stocking rates in a single pasture or rangeland type for a given season or year under continuous grazing. Recently, the Agricultural Policy/Environmental eXtender (APEX, v1605, Williams and Izaurralde, 2006) was enhanced with modifications to the plant growth module to simulate forage production (Zilverberg et al., 2017) and for selectivity by grazing animals for beef production in mixed grass prairie (Zilverberg et al., 2018).

Potential plant growth in APEX is driven by daily heat units and photosynthesis. Daily photosynthesis rate is based on radiation use efficiency and then modified for CO₂ concentration and vapor pressure deficit effects. Leaf area index (LAI), plant height, biomass partitioning to roots, and root growth are functions of heat units. Actual plant growth is modified by environmental factors, such as water, nitrogen, soil aeration, and temperature stresses. Actual daily root growth is also affected by soil bulk density (Williams, 1995; Williams and Izaurralde, 2006). Plant species compete for sunlight based on LAI and for water and nutrients based on plant demands and root distribution of each species (Kiniry et al., 1992; Williams and Izaurralde, 2006). Improvements to APEX by Zilverberg et al., 2017, Zilverberg et al., 2018, however, did not address predicting forage production under rotational grazing management, which is an urgent need for rangeland management decision support tools (Derner et al., 2012; Fust and Schlecht, 2018).

Our main objective was to evaluate the APEX plant growth modules and grazing animal selectivity in simulating forage production using experimental data collected from both traditional season-long grazing and adaptive rotational grazing management on western rangelands (Augustine et al., 2020). Specifically, we evaluated APEX's ability to simulate total forage production and production for plant functional groups and their response to different soil types and climate conditions under grazing management options.

Section snippets

Experimental data

Model evaluation utilized experimental data from the collaborative adaptive rangeland management (CARM) study conducted at the USDA-ARS Central Plains Experimental Range (40^o49’ N, 107^o47’ W), a Long-Term Agroecosystem Research (LTAR) network site (https://ltar.ars.usda.gov). Soils, rainfall, and vegetation at the site are representative of the extensive shortgrass steppe region in the western Great Plains (Burke and Lauenroth, 1993). Mean annual precipitation is 321 mm, with more than 80% of

Effects of annual precipitation

Total aboveground biomass measured inside grazing exclusion cages at the beginning of August varied considerably as influenced by precipitation (Fig. 1a, b). Total aboveground biomass was highest in 2015 (1699 ± 730 kg ha⁻¹), followed by 2014 (1461 ± 586 kg ha⁻¹), 2016 (1414 ± 639 kg ha⁻¹), 2017 (1369 ± 685 kg ha⁻¹), and 2018 (1186 ± 568 kg ha⁻¹). Simulated aboveground biomass was highest in 2014 (1798 kg ha⁻¹) and lowest in 2018 (865 kg ha⁻¹), with values of 1276 kg ha⁻¹, 1117 kg ha⁻¹, and

Conclusion

In this study, we demonstrated that APEX appropriately simulated relative differences in aboveground biomass under traditional and adaptive grazing management systems. The calibrated model also was able to accurately simulate the impacts of annual precipitation, soil texture, and alternative grazing management scenarios on total biomass production and production of plant functional groups. These results indicated that APEX was capable of assessing grazing management decisions on forage

Declaration of Competing Interest

I know of no conflict of interest that should be reported for “Evaluation of APEX modifications to simulate forage production for grazing management decision-support in the Western Great Plains” by G. Cheng, R. D. Harmel, L. Ma, J. D. Derner, D. J. Augustine, P. N. S. Bartling, Q. X. Fang, J. R. Williams, C. J. Zilverberg, R. B. Boone, D. Hoover, and Q. Yu for publication in Agricultural Systems.

Acknowledgements

This research was jointly supported by grants from the National Natural Science Foundation of China [41961124006, 41730645]; U.S. National Science Foundation [1903722]; and the program of China Scholarships Council [201806300098]. This research was a contribution from the Long-Term Agroecosystem Research (LTAR) network. LTAR is supported by the United States Department of Agriculture.

USDA is an equal opportunity employer and provider.

References (43)

A.A. Andales et al.
Evaluation of GPFARM for simulation of forage production and cow–calf weights
Rangel. Ecol. Manag.
(2005)
A.A. Andales et al.
Strategic and tactical prediction of forage production in northern mixed-grass prairie
Rangel. Ecol. Manag.
(2006)
D.J. Augustine et al.
Prescribed fire, soil inorganic nitrogen dynamics, and plant responses in a semiarid grassland
J. Arid Environ.
(2014)
D.J. Augustine et al.
Adaptive, multipaddock rotational grazing management: a ranch-scale assessment of effects on vegetation and livestock performance in semiarid rangeland
Rangel. Ecol. Manag.
(2020)
C. Bosi et al.
APSIM-tropical pasture: a model for simulating perennial tropical grass growth and its parameterisation for palisade grass (Brachiaria brizantha)
Agric. Syst.
(2020)
D.D. Briske et al.
Rotational grazing on rangelands: reconciliation of perception and experimental evidence
Rangel. Ecol. Manag.
(2008)
D.D. Briske et al.
Origin, persistence, and resolution of the rotational grazing debate: integrating human dimensions into rangeland research
Rangel. Ecol. Manag.
(2011)
I.C. Burke et al.
What do LTER results mean? Extrapolating from site to region and decade to century
Ecol. Model.
(1993)
J.D. Derner et al.
Heifer performance under two stocking rates on fourwing saltbush-dominated rangeland
Rangel. Ecol. Manag.
(2005)
J.D. Derner et al.
Grazing-induced modifications to peak standing crop in northern mixed-grass prairie
Rangel. Ecol. Manag.
(2007)

J.D. Derner et al.

Opportunities for increasing utility of models for rangeland management

Rangel. Ecol. Manag.

(2012)

J.D. Derner et al.

Can collaborative adaptive management improve cattle production in multipaddock grazing systems?

Rangel. Ecol. Manag.

(2021)

P. Fust et al.

Integrating spatio-temporal variation in resource availability and herbivore movements into rangeland management: RaMDry - an agent-based model on livestock feeding ecology in a dynamic, heterogeneous, semi-arid environment

Ecol. Model.

(2018)

L. Ma et al.

Application of grazing land models in ecosystem management: current status and next frontiers

Adv. Agron.

(2019)

Z. Qi et al.

Development and evaluation of the carbon–nitrogen cycle module for the GPFARM-range model

Comput. Electron. Agric.

(2012)

N.Q. Sima et al.

A modified F-test for evaluating model performance by including both experimental and simulation uncertainties

Environ. Model. Softw.

(2018)

W. Stout et al.

Evaluating SPUR model for predicting animal gains and biomass on eastern hill land pastures

Agric. Syst.

(1990)

W. Teague et al.

Patch dynamics under rotational and continuous grazing management in large, heterogeneous paddocks

J. Arid Environ.

(2003)

W. Teague et al.

Drought and grazing patch dynamics under different grazing management

J. Arid Environ.

(2004)

Z.S. Venter et al.

Rotational grazing management has little effect on remotely-sensed vegetation characteristics across farm fence-line contrasts

Agric. Ecosyst. Environ.

(2019)

T. Wang et al.

Evaluation of continuous and multipaddock grazing on vegetation and livestock performance - a modeling approach

Rangel. Ecol. Manag.

(2016)

Cited by (15)

Optimizing Economic Performance of Rangeland Livestock Grazing Under Price and Climate Stressors
2024, Rangeland Ecology and Management
Livestock grazing is the most economically important use of rangeland ecosystems in many parts of the world. An extensive body of literature has investigated livestock grazing plans that are economically optimal or ecologically sustainable. This paper's contribution to the literature is development and application of an empirical mathematical programming model for optimizing the economic performance of livestock grazing on rangeland ecosystems for a wide array of vegetation biomes, forage productivity levels, and economic conditions. The model is calibrated to replicate historically observed data for select counties, in which predictions of the income optimization model match available data on county-wide forage, animal performance, grazing pressure, stocking level, and net income. Results show how climate stress and economic conditions affect the economically optimized choice of stocking rates and net income for 18 observed conditions and 126 potential conditions for six counties in the Intermountain West and Mediterranean western US. While findings show optimized outcomes for a large set of conditions in that region, the methods developed here have potential application to rangeland ecosystems internationally wherever data required by the model can be secured. The modeling results provide insight and utility for ranchers, scientists, and policymakers who seek economically optimal rangeland ecosystem outcomes.
Modeling landscape wind erosion processes on rangelands using the APEX model
2022, Ecological Modelling
Citation Excerpt :
The APEX model simulates surface/subsurface hydrology and water quality, production, sustainable growth, and competition of a wide range of crops, grasses, trees, shrubs among several environmental variables (Choi et al., 2017; Gassman et al., 2010; Kamruzzaman et al., 2020; Mudgal et al., 2010; Saleh et al., 2004; Wang et al., 2006, 2008, 2014). Recent advances in APEX code in rangeland simulation make APEX a suitable model for evaluating land management effects on soil degradation, water quality, and plant communities in rangeland watersheds (Wang et al., 2014; Zilverberg et al., 2017, 2018; Cheng et al., 2021). The current version of APEX (version 1501) simulates wind erosion using the Wind Erosion Continuous Simulation (WECS Williams et al., 2008) model on smooth bare soils using daily wind speed distribution utilizing the Skidmore (1986) erosion equation.
Rangelands provide vital ecosystem services and comprise 36% of U.S. and 25% of world lands. Approximately 20 to 73% of these rangelands are already degraded. Soil erosion driven by winds is a significant cause of this degradation. Minimal work has been done to evaluate the influence of wind erosion on rangeland dominated landscapes or to assess the effect of landscape-scale soil and water conservation and management practices on reducing wind erosion due to its inherent complexity and nonpoint source nature. Our approach is to use an integrated and holistic eco-hydrologic model. The main objective of this study is to integrate process-based landscape wind erosion modeling schemes into the Agricultural Policy/Environmental eXtender (APEX) model (version 1905; APEX1905) for simulating horizontal aeolian sediment transport and vertical particles transport on rangelands. We demonstrate the performance and capability of the model using field data collected at the USDA ARS Jornada Experimental Range in New Mexico, U.S. A benchmark APEX1905 model that runs on high-resolution sub-daily wind speed data and daily average vegetation gap distribution derived from monthly field measurements. The benchmark APEX1905 model captured the variability in observed aeolian sediment transport during 2015–2017 reasonably well with an R² = 0.58 and a RMSE = 2.60. When the highly variable sub-daily measured wind speed and vegetation gap distribution data were substituted with the daily average wind speed and APEX1905 estimated daily vegetation gap, the APEX1905 reproduced well the benchmark model estimates with the performance statistic of R² = 0.77 and RMSE=0.54 for the horizontal aeolian sediment transport; and R² = 0.82 and RMSE = 0.85 for vertical particle transport. Overall, provided limited aeolian watershed data sets, the APEX1905 is demonstrated to be reliable in estimating wind erosion in arid, desert rangeland landscapes using daily average wind speed and simulated vegetation characteristics. Benefiting from the long-established algorithms of simulating plant growth dynamics and topsoil moisture content in APEX, APEX1905 offers a robust and process-based estimation of wind erosion in areas where wind and vegetation data are scarce.
Evaluation of the APEX cattle weight gain component for grazing decision-support in the Western Great Plains
2022, Rangeland Ecology and Management
Citation Excerpt :
Adaptive management under CARM might be improved by integrating simulation results, such as APEX autorotation criteria based on a minimum forage biomass threshold and movement to the pasture with the highest remaining biomass. This improved capability along with additional enhancements recently presented in Cheng et al. (2021) are critical advances because APEX is increasingly used for grazing decision-support and adaptive rangeland management (Wang et al. 2011). Considering the high spatial and temporal variabilities of soil type, topography, plant populations, and weather conditions in US rangelands, further enhancements in APEX-specific soil and plant databases are needed to extend its applicability across the nation.
Rotational grazing studies have produced mixed results related to animal performance (weight gain), which has contributed to producer uncertainty regarding grazing management decisions. To enhance decision-support for producers, we improved algorithms in the Agricultural Policy/Environmental eXtender (APEX) model to better represent cattle weight gain in real-world rangeland conditions under two grazing management strategies. Simulated weight gain and related forage effects were evaluated with experimental data from 2014 to 2018 under two grazing strategies. The traditional rangeland management strategy used continuous season-long grazing stocked at a moderate level. The collaborative adaptive rangeland management strategy employed grazing with one large herd rotated using a sequence developed by a stakeholder group with movement between pastures driven by predetermined decision triggers. For each grazing strategy, yearling steers grazed from mid-May to October on ten 130-ha pastures. With the APEX modifications, daily weight gain was adequately simulated for both continuous (traditional rangeland management) grazing and management intensive rotational (collaborative adaptive rangeland management) grazing. Dry matter intake, total digestible nutrients, and temporal distribution of dry matter intake were the primary influencers of cattle performance (weight gain). Once shown to be accurate, we used APEX to evaluate several management alternatives (i.e., stocking rate, rotation interval, and rotation decision criteria) to showcase its decision support capabilities. These important enhancements increase the utility of APEX in semiarid environments, such as the western Great Plains, in providing science-based rangeland decision support to ranchers, agency land managers, and policy makers.
Monitoring standing herbaceous biomass and thresholds in semiarid rangelands from harmonized Landsat 8 and Sentinel-2 imagery to support within-season adaptive management
2022, Remote Sensing of Environment
Adaptive management requires rangeland managers to respond to changing forage conditions (e.g., standing herbaceous biomass) within the grazing season at the scale of individual pastures. While within-season biomass can be measured or estimated in the field, it is often impractical to make field measurements in extensive rangeland systems with adequate frequency and spatial representation for responsive decision-making by rangeland managers. We sought to develop a single model to accurately predict daily herbaceous biomass across seasonally and annually varying conditions from the Harmonized Landsat-Sentinel satellite time series. We also sought to assess how information about plant community composition derived from a high-spatial resolution map would improve model performance. We used herbaceous biomass data from 1764 ground observations collected over 8 years in North American shortgrass steppe for training in a cross-validated model selection approach to evaluate (1) predictive performance of candidate models both spatially and temporally, (2) relative variable importance of individual spectral bands, vegetation indices, and recently developed broadband spectral angle indices, and (3) the degree to which including plant community composition improved model performance. All 11 candidate models identified in the model selection procedure contained a band angle index and an individual spectral band, and 6 contained one of each input feature type, demonstrating the benefit of combining spectral features for predicting herbaceous biomass across varying conditions. The spatial and temporal cross-validation and selection procedures produced the same top model with similar performance (mean absolute error = 151–182 kg ha⁻¹; R² = 0.75–0.79), suggesting that a single model performs well over space and time. Including plant community composition in the model reduced mean absolute error by 11–13%. Bootstrapping revealed that –six to seven years of training data were required to achieve consistent model performance across years with varying environmental conditions (e.g., wet, average, dry). The top model could accurately detect (70–87% accuracy) the week that biomass dropped below management-related thresholds as low as 450 kg ha⁻¹ in an independent dataset (n = 950) with modest commission error (10–26%). We discuss how maps showing the probability that herbaceous biomass is below a given threshold can support adaptive management in extensive semiarid rangelands across differing scenarios of risk perception and avoidance. In addition to producing maps to support precision rangeland management strategies, this study demonstrates the importance of combining complementary vegetation indices and acquiring long-term training datasets to achieve reliable predictions of herbaceous standing biomass in highly variable systems.
Evaluating the APEX model for alternative cow-calf grazing management strategies in Central Texas
2022, Agricultural Systems
Citation Excerpt :
The APEX grazing module was also improved for selective grazing of plant species and dietary-specific excretion of urine and feces under different stocking rates (Zilverberg et al., 2018). The model was also recently evaluated for simulating rotational grazing effects (grazing one pasture for a certain period and then grazing another one) on forage production between continuous and rotational grazing (Cheng et al., 2021). These APEX enhancements, however, did not consider multiple paddock grazing management and spatial differences in forage management (Wang et al., 2016).
Simulation tools are increasingly used to inform grazing management decisions by assessing livestock performance, as well as environmental and economic impacts. Ability to represent the grazing of multiple pastures (i.e., paddocks) that differ in soil, hydrology, vegetation, and management is critical for reliable grazing management decision support.
The main objectives of this study were to: 1) modify APEX (Agricultural Policy/Environmental eXtender) for sub-daily grazing, cow-calf weight gain, and supplemental hay, and 2) evaluate the APEX modifications in terms of simulating biomass, calf weight gain, economic impacts of alternative cow-calf grazing management strategies.
APEX was modified to enhance its ability to simulate alternative grazing management strategies by including sub-daily grazing among multiple paddocks, supplemental hay estimation, and optional simulation of cow-calf weight gain based either on energy or total digestible nutrients (TDN). Simulation results were evaluated against a 5-year experimental data set from Central Texas comparing multi-paddock rotational grazing and conventional continuous grazing.
The modified APEX model adequately simulated the responses of vegetation biomass (coefficient of determination, R² = 0.60–0.70), hay consumption (R² = 0.94–0.95), calf weaning weight (R² = 0.52–0.65), costs (R² = 0.98), and profits (R² = 0.89) to the two grazing treatments across years. Simulations with the energy-based weight gain algorithm showed more pronounced effects on above-ground biomass, whereas the TDN-based algorithm had a more pronounced weight gain response to forage intake and hay quality. No significant differences (p > 0.05) in biomass and calf weaning weight were observed between treatments across the years for measured data and for energy-based APEX simulation; however, the TDN-based algorithm simulated lower calf weaning weight in multi-paddock rotational grazing than in continuous grazing. Measured and simulated data also showed similar profits between the two grazing treatments, but total cost and gross return per ha were greater for continuous grazing.
The model enhancements for sub-daily grazing, cow-calf weight gain, and supplemental hay improved the potential utility of APEX for assessing environmental and economic impacts of alternative grazing strategies at ranch scale.
Developing a Framework for Sensitivity Analysis for Hydrological Models and Testing it for Farm-Scale Watersheds Under Grazing Operations
2024, SSRN

View all citing articles on Scopus

View full text

Published by Elsevier Ltd.

Short CommunicationEvaluation of APEX modifications to simulate forage production for grazing management decision-support in the Western US Great Plains

Highlights

Abstract

Context

Objective

Methods

Results and conclusions

Significance

Graphical abstract

Introduction

Section snippets

Experimental data

Effects of annual precipitation

Conclusion

Declaration of Competing Interest

Acknowledgements

Rangel. Ecol. Manag.

Rangel. Ecol. Manag.

J. Arid Environ.

Rangel. Ecol. Manag.

Agric. Syst.

Rangel. Ecol. Manag.

Rangel. Ecol. Manag.

Ecol. Model.

Rangel. Ecol. Manag.

Rangel. Ecol. Manag.

Rangel. Ecol. Manag.

Rangel. Ecol. Manag.

Ecol. Model.

Adv. Agron.

Comput. Electron. Agric.

Environ. Model. Softw.

Agric. Syst.

J. Arid Environ.

J. Arid Environ.

Agric. Ecosyst. Environ.

Rangel. Ecol. Manag.

Short Communication
Evaluation of APEX modifications to simulate forage production for grazing management decision-support in the Western US Great Plains