3.1 Across the reference sites among and between MLRA
The rigor in selecting appropriate reference sites is foundational for achieving reliable comparative studies. Selection of reference within the proposed framework will ensure consistency across regional and national comparative studies. The procedural explanation in this manuscript was developed while conducting a study which established a comprehensive set of reference sites from both MLRA 106 (Eastern Nebraska) and MLRA 67A (Western Nebraska). The reference sites were meticulously chosen based on the ecosystem’s response to their distinctive pedogenetic and climatic signatures (Figure 2 and Table 1). Comparison across sites within several of the CREU were investigated to validate the framework. The discussion here will focus on soil organic matter (SOM) analyzed using the Loss on Ignition method (Hoogsteen et al., 2015). While SOM serves as an important metric, the framework accommodates the inclusion of other bio-physicochemical soil health indicators, contingent on research objectives. A key rationale for choosing SOM is its stability; unless there's a significant change in soil texture, climate, land-use or management, SOM variability remains minimal and within an acceptable range. The data were analyzed using analysis of variance (ANOVA), and significant effects were tested with the Tukey HSD test. All the test and analysis were done in R, using agricolae package (de Mendiburu and de Mendiburu, 2019).
Preliminary analyses comparing SOM values from reference sites in MLRA 106 versus MLRA 67A showed a significant difference (p<0.001) of 2.84% (Figure 2 & Figure 3A). This metric underscores the risk of drawing comparisons between reference sites and croplands from different MLRAs, as it could establish potentially unattainable benchmarks. By design, MLRAs demarcate areas that share homogeneity in soil, climate, physiography, geology, and hydrology, ensuring similar land use potentials and site comparability. Thus, MLRA should be the primary level of classification.
Table 1. Metadata pertaining to the reference sites and cropland spans both eastern and western Nebraska. Specifically, MLRA 106 is in eastern Nebraska, whereas MLRA 67A is situated in the western part of the Nebraska.
Sampling sites
|
MLRA Code
|
Ecological sites
|
Soil Series
|
Texture
|
Precipitation
|
Land groups
|
NM
|
106
|
Clayey Upland
|
Pawnee
|
Clay loam
|
30-33
|
reference
|
DP
|
106
|
Clayey Upland
|
Wymore
|
Clay loam
|
30-33
|
reference
|
DTP
|
106
|
Limy Upland
|
Steinauer
|
Clay loam
|
30-33
|
reference
|
WOLT
|
106
|
Clayey Upland
|
Mayberry
|
Clay loam
|
30-33
|
reference
|
DAR
|
67A
|
Loamy (Ly)
|
Tripp
|
Sandy loam
|
14-17
|
reference
|
CR-M
|
67A
|
Limy Upland (LiU)
|
Mitchell
|
Sandy loam
|
14-17
|
reference
|
CR-O
|
67A
|
Sandy (Sy)
|
Otero
|
Sandy loam with very fine sands
|
14-17
|
reference
|
SX
|
67A
|
Sands
|
Valent
|
Sandy
|
14-17
|
reference
|
CC-1
|
67A
|
Sands (Sa)
|
Valentine
|
Sandy loam
|
14-17
|
reference
|
CC-2
|
67A
|
Shallow (Sw)
|
Tassel
|
Fine sandy loam
|
14-17
|
reference
|
BM
|
67A
|
Limy Upland (LiU)
|
Mitchell
|
Sandy loam
|
14-17
|
reference
|
PD
|
106
|
Clayey Upland
|
Wymore
|
Clay loam
|
30 -33
|
cropland
|
KH-M (manure management)
|
67A
|
Loamy (Ly)
|
Tripp
|
Sandy loam
|
14-17
|
cropland
|
KH-C (control)
|
67A
|
Loamy (Ly)
|
Tripp
|
Sandy loam
|
14-17
|
cropland
|
In a comprehensive examination, where sites were categorized based on their MLRA affiliations and subsequently filtered by soil types, the homogeneity and difference among the reference sites was both rational and actionable (Figure 3B). From MLRA 106, the soil series of Wymore, Pawnee, & Steinauer—predominantly clay loam in texture—showcased minimal SOM variation (Figure 3B). Conversely, from MLRA 106, the Mayberry soil series displayed significant SOM variances compared with other clay loam textured soil series. From MLRA 67A, the Tripp, Mitchell, and Tassel soil series, characterized by their sandy loam texture, showed uniform SOM metrics, while from MLRA 67A, the soil series Valentine and Valent series being sandy in texture diverged significantly from Tripp and Mitchell series (Figure 2B). These findings aligns with many studies which reported the importance of texture class in determining the soil health outcomes (Amsili et al., 2021; Chahal et al., 2023)
Incorporating ecological sites as a factor in the ANOVA showed that the resulting SOM variances were statistically significant. However, these variations were primarily tied to soil textural differences rather than intrinsic ecological site attributes (Figure 4). For example, ecological sites with clay loam texture exhibited the highest SOM levels, while ecological sites with sands ranked the lowest (Figure 4). There was no notable difference between the ecological sites categorized as Loamy, Shallow, and Sandy (Figure 3). This highlights the notion that while ecological sites represent the differences in potential vegetation as influenced by the site's soil factors and hydrology (a top-down approach), they might not fully encapsulate the soil's heterogeneity and homogeneity (a bottom-up approach). Hence, while ecological sites and descriptions are crucial for pinpointing soil sampling reference points for specific soil series and locations or to determine the states of transition, the division of land for comparative soil analysis could be broader, detouring detailed ecological site theories. This is also supported by the study from Seybold et al., (1997), which explained a reference values determined for a group of soils that functions similarly or are genetically similar as defined by the National Cooperative Soil Survey will be more practical and appropriate for comparative studies.
In a focused comparison between reference sites, those within MLRA 106 (Eastern Nebraska) consistently demonstrated higher soil organic matter (SOM) relative to their counterparts in MLRA 67A (Western Nebraska). When data were further explained based on soil texture, distinct patterns were observed. Reference sites CR – M and BM (Table 1), with the Mitchell soil series, and DAR with Tripp soil series (Table 1), exhibited similar SOM profiles. Likewise, the CC-2 site, characterized by the Tassel soil series (Table 1), mirrored these SOM trends, with no statistically significant differences (Figure 5). It's pertinent to note that Tripp, Mitchell, and Tassel all classify as super-active calcareous mesic soil series. Such findings resonate with the perspective of Seybold et al. (1997) and CREU (Das and Maharjan, 2022), emphasizing the feasibility of formulating uniform reference points by clustering soils that exhibit similar pedogenetic attributes, thereby providing a pragmatic framework for guiding consistent management practices across a particular agricultural landscape, where the soils and precipitation pattern is uniform.
Variations in SOM can be predominantly traced back to the intrinsic characteristics of the soil series (p<0.001) and the regional precipitation patterns (p<0.001). The Valent soil series (site Sx), distinguished by its pronounced sandy texture (comprising 70-100% sand), showed lowest SOM when compared with other soil series (Figure 5). This underscores the strategic importance of melding the MLRA framework with the CREU concept. Such an integration facilitates a comprehensive landscape classification aiming to identify zones with uniform soil health potential. This method of stratification is poised to offer valuable insights for upcoming comparative soil health research endeavors. By capitalizing on the granularity inherent to the MLRA framework, in conjunction with data pertaining to soil series (groups) and climatic variables (precipitation), researchers can more adeptly delineate precise soil health benchmarks, optimized for specific land management interventions.
3.2 Soil Health Gap Assessment
The results from these comparative studies underscore the importance of judiciously selecting reference sites based on MLRA, soil, and precipitation gradients. The central role played by these reference sites, especially those that align with reference plant communities detailed in ecological site descriptions, is evident in generating accurate soil health gap assessments (Figure 6). The data revealed a significant difference between the primary reference site and its cropland, which were located within MLRA 67A, with precipitation in the range of 14-17 inches, and having Tripp soil series. It was observed that lands treated with manure showed minimal soil health gap (0.8%) than those subjected to conventional management practices (1.8%). This finding emphasizes the benefits of manure application in enhancing soil organic matter content and overall soil health. Similarly, a clear difference of 2.03% in organic matter was noted between the wymore reference site (DP) and corresponding cropland (PD) (Figure 6). While these differences are evident, they remain within a reasonable margin of improvement. Contrastingly, comparing a reference site in the Wymore soil of eastern Nebraska with cropland in western Nebraska, which exhibits distinct soil and climatic characteristics, yielded a significant 4% SOM gap. Such disparities pose considerable challenges for farmers in western Nebraska and reinforce the importance of prudent reference site selection based on native potential of each site within the proposed framework.