Management-intensive Grazing Affects Soil Health

Abstract. Management-intensive Grazing (MiG) on irrigated, perennial pastures has steadily increased in the western US due to pressure for reducing public lands grazing, overall declining land available for pasture, and decreasing commodity prices. However, there are still many unknowns regarding MiG and its environmental impact, especially with regards to soil health. Over a two-year period, we studied changes in soil health under a full-scale, 82 ha pivot-irrigated perennial pasture system grazed with ~ 230 animal units (AUs) using MiG. Soil analysis included 11 soil characteristics aggregated into the Soil Management Assessment Framework (SMAF), which outputs results for soil biological, physical, nutrient, chemical, and overall soil health indices (SHI). Positive impacts were observed in the biological SHI due to increases in microbial and enzymatic activities, even though soil organic C (SOC) remained relatively unchanged; however, positive biological SHI changes are likely precursors to future SOC increases. The nutrient SHI declined due to a reduction in plant-available soil P over time, potentially due to greater plant uptake. A negative impact was also observed in the physical SHI, driven primarily by increasing bulk density due to hoof pressure from cattle grazing. If managed correctly, results suggest that irrigated, MiG systems have the potential for success with regards to supporting grazing while promoting soil health for environmental and economic sustainability.


Over a two-year period, we studied changes in soil health under a full-scale, 82 ha pivot-irrigated 23 perennial pasture system grazed with ~ 230 animal units (AUs) using MiG. Soil analysis 24 included 11 soil characteristics aggregated into the Soil Management Assessment Framework 25 (SMAF), which outputs results for soil biological, physical, nutrient, chemical, and overall soil 26 health indices (SHI). Positive impacts were observed in the biological SHI due to increases in 27 microbial and enzymatic activities, even though soil organic C (SOC) remained relatively 28 unchanged; however, positive biological SHI changes are likely precursors to future SOC 29 increases. The nutrient SHI declined due to a reduction in plant-available soil P over time, 30 potentially due to greater plant uptake. A negative impact was also observed in the physical SHI, 31 driven primarily by increasing bulk density due to hoof pressure from cattle grazing. If managed 32 correctly, results suggest that irrigated, MiG systems have the potential for success with regards 33 to supporting grazing while promoting soil health for environmental and economic sustainability. 34

Introduction 35
Management-intensive Grazing (MiG) is often defined as "a flexible version of rotational 36 grazing that balances forage supply with animal demand" (Stout et al., 2000). Over the past 37 decade, interest in MiG on irrigated pastures has increased steadily due to the prospects of 38 reduced production costs, increased animal output, land use efficiency, and environmental 39 benefits. This system is being considered as an option by many farmers and ranchers in the 40 western US due to pressure to reduce public land grazing and the declining land available for Martinez et al. (2008) showed that microbial biomass C (MBC) was up to 6.6 times greater under 53 pasture compared to corresponding vegetable production sites. Intimately correlated to microbial 54 biomass C is the ability of microorganisms to degrade soil organic matter via enzymatic activity 55 (Turner et al., 2002). Enzymes are integral to all soil biological activity, including organic matter 56 decomposition. Specifically, the β-glucosidase (BG) enzyme plays an important role in the 57 residue correlated to higher penetrometer resistance, an indirect measure of soil bulk density. 104 Positive and negative grazing impacts in perennial pasture systems can make or break the 105 viability of a livestock enterprise, with the primary driver behind yields, forage health, and 106 profitability being the ability of soils to function properly, the basic premise behind soil health. 107 To date, we are unaware of any studies focused on quantifying soil health in irrigated, 108 MiG systems. The overall study objective was to quantify soil health changes caused by a land-109 use change from cropland to an irrigated, perennial pasture grazed by cattle using the Soil 110 Management Assessment Framework (SMAF), a soil health tool developed by Andrews et al. 111 (2004). Based on current literature, we hypothesized that converting irrigated cropland to 112 perennial, MiG pasture would cause 1) negative changes in the physical soil health index (SHI) 113 due to increases in bulk density exerted from hoof pressure; 2) the biological SHI to increase due 114 to microbial biomass C and enzymatic activity being stimulated from perennial grass roots and 115 lack of tillage; and 3) the nutrient SHI to increase due to greater P and K levels from manure and 116 urine deposition during the grazing season. 117 Prior to project establishment, the study area was managed for about a decade as a tilled 134 cropping system with crops including corn grain and silage (Zea mays L.), dry beans (Phaseolus 135 vulgaris L.), and alfalfa (Medicago sativa L.). In 2016, the area was converted to four forage 136 mixtures, one planted on each quarter (~ 20 ha) of the center pivot, including a simple grass-137 legume mix, complex grass-legume mix, simple grass mix, and complex grass mix (Table 1), 138 with all grasses being cool-season. Prior to planting, a deep ripper was used to alleviate a plow 139 pan that had formed due to the clayey soil texture that dominates the site, as well as previous 140 management. Following deep ripping, the field was moldboard plowed, disked twice, 141 cultipacked twice, and then rolled with a heavy steel roller to break up large soil aggregates and 142 firm the soil surface prior to seeding. Prior to the second cultipacking operation,  Table  148 1) failed due to winds that moved loose soil and damaged small seedlings. Oats were planted in 149 this quarter on March 23, 2017 to avoid further wind erosion. The planned mixture was re-150 planted in August 2017. For the mixtures containing legumes, most of the legumes that were 151 cross drilled in mid-March were killed due to a hard frost that occurred shortly after germination. 152 Establishment success of the legumes was approximated at less than 5%. Legumes were again 153 interseeded in early August of 2018. Thus, the forage mixtures with legumes contained few if 154 any legumes for the study duration. Although plant species mixes were an important facet of the 155 overall project, for this manuscript we focus on the use of plant material removal percentage 156 within the context of MiG (see section 2.3, Grazing Management below). 157

Project Design 158
Permanent infrastructure included an electrified perimeter fence with two concentric 159 inner fences constructed using high-tensile wire (Fig. 1). Ten permanent water blocks were 160 located around the pivot, eight within the outer concentric high-tensile fence line and two in the 161 center. The eight water blocks within the outer concentric fence had four sides with electrified 162 rope gates for controlling access to paddocks. Paddocks were generally created every 1-3 days using polywire and step-in posts. Fences that delineated paddocks were re-constructed in the 164 same locations throughout each grazing rotation using the GPS and paddock drawing tools on the 165 mobile application PastureMap ("PastureMap Grazing Management and Livestock Software," 166 2019). Further subdivisions to these paddocks were also made on an as-needed basis using 167 additional polywire and step-in posts to adjust for forage availability and animal numbers. These 168 fences were not placed in the same location each grazing rotation. 169 In 2017, approximately 171 cow-calf pairs were grazed from August 18 until October 24. 170 The grazing season was delayed due to fencing and water infrastructure being constructed. 171 Because of the delay, initial growth of forage in mixtures B, C, and D was harvested for hay. A 172 total of 309 Mg of above-ground biomass was removed as hay, which equated to 5 Mg ha -1 . 173 When grazing commenced on August 18, the cows were initially separated into two herds by 174 breed (Angus and Hereford) for breeding purposes and then combined on September 21 and 175 grazed as one herd until October 24. In 2018, approximately 136 cow-calf pairs, 49 replacement 176 heifers, and 5 steers were grazed from May 4 to October 7, with similar animal separation for 177 breeding purposes as in 2017. 178

Grazing Management 179
MiG systems require manipulating the length of time cattle graze and space allotted 180 based on available forage resources to achieve management goals and objectives (Shewmaker 181 and Bohle, 2010). For this project, cows were generally moved daily. In certain situations, 182 depending on forage availability and herd size, cows were moved every 2-4 days. This 183 management method allowed for making daily adjustments in order to maintain electric fencing, 184 monitor cattle health and soil conditions, and preserve plant health.
In terms of plant health and performance, the goal for forage removal was approximately 186 50% of available biomass during a given grazing period. Once returned to the laboratory, samples were stored in a refrigerator at 4ºC before 219 processing. Cores for moisture content and bulk density were weighed, then immediately dried at 220 105ºC for at least 24 hours, and then weighed again. The bulk soil samples were passed through 221 an 8-mm sieve, removing large pieces of organic material and rock. A representative sub-sample 222 of ~150 g of field-moist, 8-mm sieved soil was placed immediately in a plastic Ziplock bag, 223 labeled and stored at 4ºC for subsequent microbial biomass C analysis. Another sub-sample of 224 ~150 g of 8-mm sieved soil was passed through a 2-mm sieve and then air-dried, while the 225 remaining 8-mm sieved soil was allowed to air-dry for subsequent analyses. 226  (Table 2). 256

Soil Health and Laboratory Soil Analyses
This led to a significant decrease in the bulk density index score between years, from 0.80 to 257 0.37 and 0.59 to 0.34 in the 0-5 and 5-15 cm depths, respectively (Table 3). Soil surface bulk 258 density was likely lower post tillage due to soil mixing during ground preparation for planting 259 and would likely increase when cattle exert hoof pressure on the soil during grazing. The 260 minimum and maximum bulk densities measured in 2018 after the first grazing season were 0.77 261 and 1.89 g cm -3 , respectively. Previous studies have shown that when soils reach or exceed a 262 bulk density of 1.7 g cm -3 , root growth is impeded (Bruand and Gilkes, 2002). Although mean 263 bulk density levels did not reach 1.7 g cm -3 , future monitoring of this indicator will be important 264 from a soil health and forage productivity perspective. 265 Water stable aggregates (WSA) did not change significantly between pre-and post-266 grazing (Table 2). However, a significant difference between depths for WSA indicator values 267 and index scores existed (Tables 3 and 4, respectively). Greater aggregation was present at the 5-268 15 cm depth than the 0-5 cm depth. In 2018, mean WSA percentages were 45.7% at 0-5 cm and 269 59.6% at 5-15 cm. Factors that could have contributed to this difference include the lack of 270 tillage from 2017 to 2018, the addition of perennial grasses with fibrous root systems, and 271 microbial activity (e.g., increased MBC, discussed below) related to these management changes. 272 Soil aggregate formation relies heavily on microbial activity which is often greater in grazed, 273 improved pasture systems than in native or tilled systems, simply due to the level of production 274 present (Sparling, 1992;Warren et al., 1986). Less WSA in the 0-5 cm depth could be attributed 275 to the physical pressure of grazing. Warren et al. (1986) found that soil aggregate size was negatively correlated to trampling rate. Although perennial vegetation and microbial activity can 277 aid in aggregation, pressure on the soil surface from animals' hooves during grazing may have an 278 adverse effect on aggregation. 279 Soil bulk density and WSA data both contribute to the physical SHI value in SMAF. 280 Changes in bulk density were the primary factor that caused a significant decrease in the physical 281 SHI between 2017 and 2018 (Table 4). 282

Soil Biological Indicators and Biological Soil Health 283
β-glucosidase activity significantly increased from 2017 to 2018 (Table 2). This was 284 likely the result of a land-use change from a tilled cropping system to a perennial pasture system.  (Table 3). There was also a significant year by depth interaction due 297 to a greater increase at the 0-5 than 5-15 cm depth over time. Other studies have shown that 298 C (i.e., energy) source for microbes (Doran, 1987). Converting from a tilled, low residue system 300 to a perennial, grazed system likely provided an influx of C material responsible for the increase 301 in MBC. In addition, the grazing strategy employed in this study aimed to utilize only 50% of the 302 forage biomass present during a grazing period. According to multiple studies, partial plant 303 defoliation has been found to increase soluble exudates from plant roots (Bardgett et al., 1998;304 Holland et al., 1996). This rhizodeposition, in turn, would be expected to stimulate soil microbial 305 activity. Synthetic N was not added to this system once converted to a perennial pasture. 306 Reduced soil N has been found to stimulate the release of organic root exudates in grasses as 307 well as foster a greater microbial community (Bardgett et al., 1998;Hodge et al., 1996). Studies 308 have shown that cool-season, managed grasses, like those present in the current study, tend to 309 exude quickly decomposable C substrates which can stimulate microbial activity (Grayston et al., 310 1998). Easily decomposable C substrates were also likely present in this grazing system due to 311 cattle manure inputs and the fast-growing nature of modern cool-season grass varieties (Dawson 312 et al., 2000). 313 Potentially mineralizable nitrogen was significantly greater in 2018 than in 2017 (Table  314 2), causing a significant increase in the PMN index score from 2017 to 2018 over both soil 315 depths (Table 3). Precipitation (or irrigation in the current study) has been positively correlated 316 with PMN soil concentrations (Doran, 1987), with PMN serving as an indicator of a microbial 317 population's capacity to mineralize nitrogen from organic to plant-available forms. Thus, 318 managed irrigation could provide an advantage to MiG systems in terms of how quickly manure 319 N is mineralized; a further advantage would be that perennial pasture systems are not tilled. 320 Doran (1987) found that in a no-till system, soil microbial biomass and PMN distributions were 321 similar, with both being greatest in the top 7.5 cm of soil. In long-term grazing systems, manure  (Tables 2 and 3). It should be noted that changes in BG and MBC have 328 been detectable earlier than changes in SOC because of the rapid turnover rate, with BG and 329 MBC being early indicators of long-term soil C accumulation (Sparling, 1992;Turner et al., 330 2002). Given time, we would expect the system in the current study to significantly gain SOC, 331 which along with increased WSA under pasture settings, would lead to physical soil 332 improvements (Martens et al., 2004). Continued monitoring will be necessary to track possible 333 SOC changes over time and correlations with other indicators in this system. 334 The changes that occurred in three out of the four biological indicators caused an increase 335 in the biological soil health index score from 2017 to 2018 (Table 4). The land-use change from 336 a tilled, cropping system to a no-till perennial system has likely imparted positive changes on soil 337 biological activity and, thus, the biological soil health index. Veum et al. (2015) utilized the 338 SMAF to assess soil health for different annual and perennial cropping systems. They concluded 339 that biological and physical SHI categories were the most sensitive to changes in management. 340 Paudel et al. (2011) found that grazed pasture systems had greater β-glucosidase activity and soil 341 organic C. They concluded that because there is minimum disturbance, more organic matter can 342 accumulate resulting in ecological benefits to the system. Again, given time, we would expect 343 the current pasture system to gain soil organic C. 344

Soil Chemical Indicators and Chemical Soil Health
There was no significant change in pH or the pH index value from 2017 to 2018 (Tables  346   2 and 3). Due to percent calcium carbonate (7 to 15%) and CEC (10 to 28 cmolc kg -1 ) (CA Soil 347 Resource Lab, 2008), in addition to clay content (28 to 49%; as determined for the SMAF), soils 348 at this site likely have a high buffering capacity that resists change in pH. This could mean that 349 pH, even over the future long-term of this grazing project, may not significantly change.  (Table 4). 359

Soil Nutrient Indicators and Nutrient Soil Health 360
Extractable K concentrations significantly increased from 2017 to 2018, the 0-5 cm depth 361 contained greater extractable K than the 5-15 cm depth, and a significant year by depth 362 interaction existed ( Table 2). The extractable K index values were significantly different between 363 years and between depths with a higher indicator score for the 0-5 cm depth (Table 3) of dairy cattle using lysimeters, finding that ~20% of K that was applied remained within the top 370 0-5 cm of soil. 371 Olsen-extractable P significantly decreased from 2017 to 2018 (Table 2) leading to a 372 decrease in the extractable P index value between years (Table 3). Approximately 96% of P 373 intake by cattle is excreted in manure, 70% of which is primarily in inorganic forms (Eghball et 374 al., 2002). Because of this, extractable P concentrations were expected to increase due to cattle 375 manure deposition, yet this was not the case. The decrease in soil extractable P concentration led 376 to a significant reduction in the nutrient SHI (Table 4) and Lowery, 1973). Therefore, the likelihood of seeing an impact on K concentrations was three 384 times greater than the potential impact of manure deposition on P concentrations in the soil. 385

Combined Effects on Physical, Biological, Chemical, and Nutrient Soil Health on Overall 386
Soil Health 387 Physical and nutrient SHI's both decreased, while the biological SHI increased between 388 years. These changes essentially negated each other, leading to no significant change in the 389 overall soil health index from 2017 to 2018 (Table 4) Soil physical, biological, and nutrient SHI values responded significantly to management 397 changes from a tilled, irrigated cropping system to a no-till, irrigated perennial MiG system. 398 Positive soil health effects were observed in the biological SHI, in particular, increases in MBC 399 and BG enzymatic activity, both of which could be early indicators of future C sequestration. 400 These findings support our hypothesis that the biological SHI would increase under MiG. In the 401 future, the addition of microbial level physiological profiling (e.g., proportion of bacteria:fungi) 402 would increase our understanding of the implications of management-intensive grazing on soil 403 biological health. Soil organic C remained relatively unchanged but will be an important 404 indicator to monitor into the future, especially with regards to its link with MBC and PMN. 405 Negative impacts occurred to the physical SHI, driven primarily by increasing bulk 406 density. This finding supports our hypothesis that the physical SHI would decrease under MiG. 407 Furthermore, this result was likely caused by initial hoof compression of the soil surface during 408 grazing. Bulk density is an indicator that should be monitored closely in the future due to its 409 potential impacts on hydrology and root health. The nutrient SHI value declined due to the 410 observed reduction in extractable soil P, which did not support our hypothesis that this SHI 411 would increase under MiG. Cattle urine inputs likely contributed to a significant increase in 412 available K, which would also have been expected for P due to its high concentration in cattle  16.1 13.3 11.7 11.1 ** NS NS † ρb = bulk density, WSA = water-stable aggregates, BG = β-glucosidase activity (pnp = p-nitrophenol), 556 PMN = potentially mineralizable N, SOC = soil organic C, MBC = microbial biomass C; EC = electrical 557 conductivity, K = extractable K, and P = extractable phosphorus. 558 ⁋ NS = non-significant, * = p < 0.05, and ** = p < 0.01.