Cropping systems with perennial vegetation and livestock integration promote soil health

Soil health can differ across cropping systems because of variation in edaphic and management factors. We evaluated how biological indicators of soil health (soil organic matter [SOM], permanganate oxidizable carbon [POXC], mineralizable carbon [MinC], autoclaved‐citrate‐extractable [ACE] protein, and potentially mineralizable nitrogen [PMN]) compared across four common Wisconsin cropping systems: grazed cool‐season pastures, forage‐based rotations that included perennial legumes or grasses, annual rotations receiving manure, and annual rotations receiving synthetic fertilizers. Biological indicators of soil health were up to 195% greater in pastures than other cropping systems. MinC, POXC and PMN were 10%–90% greater in forage‐based rotations than annual cropping systems, but only MinC and POXC were greater in annual systems with manure compared to those without manure by 35% and 7%, respectively. Perennial vegetation and livestock integration offer the greatest potential to increase biological indicators of soil health in agricultural lands.


INTRODUCTION
Improving soil health in agricultural systems is a growing area of interest for farmers, agricultural and conservation professionals, researchers, and society. Soil health princi-principles increases from annual rotations to forage-based rotations, to pasture, through the inclusion of perennial vegetation and livestock. Soil health indicators related to labile carbon (C) and nitrogen (N) pools and C and N mineralization are sensitive to management and may serve as valuable measures of directional soil health change. Biological indicators of soil health, including permanganate oxidizable C (POXC), autoclaved-citrate-extractable (ACE) protein, mineralizable carbon (MinC), and anaerobic potentially mineralizable nitrogen (PMN) are promoted measures because they are logistically feasible, cost effective, responsive to management, and agronomically valuable (Culman et al., 2012;Drinkwater et al., 1996;Hurisso, Moebius-Clune, et al., 2018). While all measure some fraction of soil C and N pools, they each relate differently to biological soil properties. POXC, or active C, reflects a more processed pool of C and is indicative of greater soil C stabilization (Culman et al., 2012;Hurisso et al., 2016). MinC, or soil respiration, reflects the pool of C accessible to microbes and the potential microbial activity in the soil Hurisso et al., 2016). ACE is an estimate of soil protein, which constitutes a large pool of soil organic N and has a strong association with SOM and soil aggregation (Hurisso, Moebius-Clune, et al., 2018). Lastly, PMN represents plant available N as a function of anaerobic microbial activity (Drinkwater et al., 1996).
In Wisconsin, cash-grain and dairy operations are the dominant agricultural enterprises whose common cropping systems include annual crop rotations with and without manure, forage-based grain-alfalfa (Medicago sativa L.) rotations, and grazed cool-season pastures. Research at the Wisconsin Integrated Cropping Systems Trial (WICST) found that POXC, MinC, and PMN were greater in forage-based rotations and pasture than annual cropping systems, while these indicators did not differ within the annual grain or forage systems under a range of management (Diederich et al., 2019). While WICST is useful for evaluating soil health while controlling for inherent environmental and soil properties, on-farm assessments allow us to examine the generality of these findings in the context of environmental and management variability. We analyzed soils from 624 Wisconsin fields to compare biological indicators of soil health across a range of cropping systems.

MATERIALS AND METHODS
We compiled data from three Wisconsin on-farm soil health studies (Augarten, 2022;Malone, 2022;Richardson, 2018) and divided sites into four cropping system categories based on the last 5 years of management: annual rotations (n = 214), annual rotations with manure (n = 180), forage-based rotations of annual crops and perennial legumes or grasses (n = 138), and cool-season pasture grazed for at least 1 year

Core Ideas
• Wisconsin soil health assessment of 624 fields found that biological indicators were sensitive to cropping system. • Mineralizable carbon (MinC), permanganate oxidizable carbon (POXC), and potentially mineralizable nitrogen (PMN) differed among cropping systems more than soil organic matter (SOM) or autoclaved-citrate-extractable (ACE) protein.
• Pastures were associated with greater SOM and biological indicators than annual rotations. • Manure use in annual cropping systems was associated with greater POXC and MinC. • Combining all five soil health principles is the key to healthier soils.
(n = 92) ( Figure 1). Soils were predominantly silt loam (70%), which is common for the region. Across cropping system categories, the distribution of soil types was consistent: 14%-24% were coarse-textured (sand, loamy sand, fine loamy sand, sandy loam), 71%-82% were medium textured (silt loam, loam), and 4%-7% were fine textured (silty clay loam, clay loam, muck) soils. Sites were sampled once between late-April and mid-June from 2015 to 2021. Soils were sampled from 0 to 15 cm depth with a probe of 2.5-cm internal diameter, and then dried for 1 week at 32˚C in a forced air drier and ground through a 1or 2-mm sieve. Before drying, Richardson (2018) transferred soils to a freezer for less than 1 month, Augarten (2022) transferred soils to a refrigerator for 1-3 days, and Malone (2022) received soils by mail from participating farmers and crop advisors.
Analysis of variance and Fisher's least significant difference were used to evaluate cropping system effects (p < 0.05) on biological indicators of soil health (R Core Team, 2021).

RESULTS AND DISCUSSION
Biological indicators of soil health were greatest in grazed cool-season pastures, followed by forage-based rotations, F I G U R E 1 Site locations (n = 624) categorized by cropping system: annual (n = 214), annual receiving manure (n = 180), forage-based (n = 138), and grazed pasture (n = 92) annual rotations receiving manure, and annual rotations without manure ( Figure 2). Average SOM, MinC, ACE, and PMN were greater in pastures than forage and annual rotations by 40%-75% and 40%-195%, respectively. POXC did not differ between pastures and forage-based rotations but was 14%-22% greater in pastures compared to annual rotations with and without manure. MinC, POXC, and PMN were 10%-90% greater in forage-based rotations than either annual cropping system, but only MinC and POXC were greater in annual systems with manure compared to those without manure. These results align with findings from WICST (Diederich et al., 2019) and New York soil health assessments (Amsili et al., 2021) indicating pastures had greater biological soil health than forage cropping systems, followed by annual grain rotations. Similarly, cropping systems in Ohio with more crop diversity and perenniality were associated with greater biological soil health (Sprunger et al., 2020).

Biological indicators were greatest in pasture
Pastures had greater SOM and labile C and N than annual and forage cropping systems, aligning with previous studies (Becker et al., 2022;Guillaume et al., 2021;Sanford et al., 2012). As perennial systems, well-managed pastures continuously cover the soil surface, nearly eliminate soil disturbance, and provide greater C inputs through longer growing seasons, continual root production and turnover, and root exudation (Cates et al., 2016;Fran-zluebbers & Stuedemann, 2015;Jackson et al., 1997;Sanford et al., 2022;Sprunger et al., 2017;Teague & Kreuter, 2020). Further differentiating pastures from forage-based or annual rotations, well-managed grazing stimulates above-and belowground C inputs, recycles organic matter as manure, and maintains a diversity of plant species (including grasses, legumes, and forbs) (Franzluebbers & Stuedemann, 2015;Teague & Kreuter, 2020). Increasing plant diversity alone was shown to increase microbial necromass in warm-season restored prairies (Liang et al., 2016). Moreover, the grazed pasture at WICST, which had greater microbial necromass accumulation and carbon use efficiency, had more "stabilized" mineral-associated organic matter-C than annual and perennial forage systems .

Manure use in annual rotations
Annual systems receiving manure had 7% greater POXC and 35% greater MinC than those without manure. MinC tends to be more sensitive to manure use relative to POXC or SOM (Hurisso et al., 2016). Applying manure had no effect on POXC in some studies (Jokela et al., 2009;Wienhold, 2005), but others found greater POXC associated with manure use (Dungan et al., 2022;Min et al., 2003;Mirsky et al., 2008). On farms in North Carolina, both soil organic C and soil test biological activity were 50%-55% greater in fields receiving organic amendments (Franzluebbers, 2020a). While manure is a valuable nutrient source in Wisconsin and often is promoted in annual systems to improve soil health, there are inconsistent effects of manure use on SOM and C and N pools. Ten years of dairy manure applications had no effect on SOM in Wisconsin cash-grain systems (Rui et al., 2020), while a meta-analysis of Midwest farms revealed that manure increased surface soil organic carbon by 39% (Nunes, van Es, et al., 2020). Furthermore, manure applications are associated with greater soil health when used in conjunction with other C inputs, like cover crops or crop residue (Aguilera et al., 2013;van Es & Karlen, 2019). For example, Jokela et al. (2009) found that corn silage rotations with cover crops and liquid dairy manure had greater POXC than those just with manure. Combining soil health practices such as manure use, reduced soil disturbance, and C inputs may provide additive benefits to soil health (Aguilera et al., 2013;Franzluebbers & Stuedemann, 2015;.

Indicator response varied
Indicator response varied across cropping systems, reaffirming that these measures are capturing subtle differences in biological soil health. MinC, which differed among all cropping systems by 35%-195%, was the most sensitive indicator, as observed in other studies (Amsili et al., 2021;Diederich et al., 2019;Sprunger et al., 2020). This could be attributed to greater responsiveness to practices associated with SOM mineralization, such as manure use, integration of legumes, and perennial vegetation (Hurisso et al., 2016). MinC is also associated with greater analytical, temporal, and spatial variability compared to other indicators (Crookston et al., 2021;Wade et al., 2018), although longer incubations may reduce analytical variability (Franzluebbers, 2020b). In our analysis, average MinC values for each cropping system were 2.5 to 3 times greater in Richardson (2018) compared to Malone (2022) (data not shown). We speculate that mailing in field-moist samples without ice packs may have suppressed aerobic microbes in Malone (2022) (Moebius-Clune, 2016). Excluding Malone (2022) samples from our analysis, average MinC continued to be greatest in pastures (128 mg kg −1 ), followed by forage-based rotations (106 mg kg −1 ), and then annual systems, but did not differ between those with manure (92 mg kg −1 ) and those without manure (88 mg kg −1 ). POXC was the only measurement that did not differ between forage-based rotations and pastures, which aligns with previous soil health studies (Amsili et al., 2021;Diederich et al., 2019). POXC is associated with practices that promote SOM accumulation and stabilization, including reduced tillage and inputs of stable C (Hurisso et al., 2016;. In this study, forage-based rotations and pastures, which have reduced tillage and higher C inputs from crop biomass, were associated with greater POXC than annual rotations. Previous studies observed similar results for labile N measures: ACE did not differ between annual and forage-based rotations (Amsili et al., 2021) and PMN was more sensitive to cropping system or crop rotation than nutrient source (Culman et al., 2013;Diederich et al., 2019;Morrow et al., 2016).

CONCLUSIONS
This Wisconsin-wide cropping system comparison demonstrated that integrating a combination of soil health building practices was associated with greater biological indicators of soil health. Grazed cool-season pastures had the greatest values of biological indicators suggesting the importance of combing all five soil health principles: minimal soil disturbance, continuously living root systems, permanent soil cover, high plant diversity, and livestock integration. While adding manure to annual cropping systems may improve soil health, the response of biological indicators was weak compared to forage-based rotations or pasture, which both include perennials. Integrating all soil health building principles into annual cropping systems offers the greatest potential to improve soil health, yet doing so remains a great challenge and should be a focus of future research efforts.

AU T H O R C O N T R I B U T I O N S
Abigail J. Augarten: Conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, software, visualization, and writing-original draft. Shawn P. Conley: Funding acquisition, project administration, resources, and writing-review and editing. Eric T. Cooley: Funding acquisition, resources, and writing-review and editing. Randall D. Jackson: Funding acquisition and writing-review and editing. Lindsay Chamberlain Malone: Conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, and writing-review and editing. Gregory S. Richardson: Data curation, formal analysis, funding acquisition, investigation, methodology, and writing-review and editing. Amber M. Radatz: Conceptualization, funding acquisition, project administration, resources, and writing-review and editing. Matthew D. Ruark: Conceptualization, formal analysis, funding acquisition, methodology, project administration, resources, supervision, validation, and writing-original draft. Michel A. Wattiaux: Funding acquisition and writing-review and editing.

C O N F L I C T O F I N T E R E S T
The authors declare no conflict of interest.