**1. Introduction**

The potential to reduce soil productivity and adversely affect other ecological functions of soil by trafficking soil with ground-based equipment has long been a concern in agriculture and forestry [1]. More than 150 papers published between 1970 and 1977 reported that soil compaction affected the growth of agricultural crops and forest trees [2]. Compaction of forest soils was the focus of 26 papers with only 2 papers failing to report compaction-reduced tree growth. These papers drew attention to the most obvious examples of agroforestry machines impacting soils and crop productivity. At the time, indiscriminate areal trafficking of soil and a lack of awareness of the problem were important issues. Concentrating the trafficking to a few skid trails reduced the impact [3], and other options to manage soil compaction were promoted [4]. New literature reviews continue to summarize the expanding range of impacts soil compaction causes and options for their managemen<sup>t</sup> [5–7], and evaluate a wider array of machines and operational practices to reduce soil impacts [8,9], but the potential for long-term term natural recovery of anthropogenically compacted soils remains speculative [10].

The ecological consequences of compaction on forest soils are complex. Soil compaction causes an increase in bulk density and a related decrease in air-filled porosity as

**Citation:** McNabb, D.H.; Startsev, A. Seven-Year Changes in Bulk Density Following Forest Harvesting and Machine Trafficking in Alberta, Canada. *Forests* **2022**, *13*, 553. https://doi.org/10.3390/f13040553

Academic Editors: Fuzhong Wu, Zhenfeng Xu and Wanqin Yang

Received: 1 March 2022 Accepted: 27 March 2022 Published: 31 March 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

a function of an increasing number of machine passes [11]. These changes in soil affect a wide range of other soil properties important for the growth of roots [12], including soil resistance to root penetration [13], soil aeration [14], soil water availability, and movement of water into and through soil [15,16]. The potential for soil compaction to affect multiple, interrelated soil properties has complicated our ability to relate changes in tree growth on compacted soil to a single-soil property. This complexity also makes it difficult to define specific values of bulk density that limit root and tree growth [4]. Values of natural and compacted bulk densities are also significantly correlated [17]. Particle size distribution and organic matter content are important soil properties responsible for these differences [18]. A newer alternative is to use relative density instead of bulk density as a measure of the severity of soil compaction. Relative density is the ratio of measured bulk density divided by the bulk density obtained using a standardized compaction or consolidation test [19,20]. Relative density has been related to differences that soil compaction has on tree growth for different sites and soils in British Columbia, Canada [21].

The effectiveness of natural decompaction processes to restore compacted soil to its original condition is uncertain because of the range of factors and interactions involved [22]. As a result, efforts to estimate soil recovery are based on studying a chronosequence of similarly treated sites. Unfortunately, the history of older sites may not be well documented, and the ability to visually identify compacted and adjacent control soils on older sites becomes less dependable [4]. Finally, the estimated time for compacted soil to recover in chronosequence studies is based on regression analyses of when the difference between the bulk density of samples selected to represent compacted and control soils becomes zero. Estimates of forest soil recovery based on this methodology have ranged from about two decades in northern climates [23] to 50 to 70 years in France [24] and more than 32 to 70 years in the western US [25,26]. Surface soils consistently recover faster than subsoil [25,26]. In a recent review of the natural recovery of skid trails, DeArmond et al. [10] identified 18 of 64 studies where the physical properties of the surface soil to a depth of 5 to 15 cm had recovered, but recovery was primarily limited to the less traveled skid trails.

More reliable measures of soil recovery should be obtained if bulk density is repeatedly measured on the same site over time, but short-term reports of soil recovery have been inconsistent. For example, Labelle and Jaeger [27] reported two loams in New Brunswick, Canada, which indicated no natural recovery after 5 years. On the other hand, Page-Dumroese et al. [28] reported a possible minor recovery of compacted, coarse-textured soil in a 0–10 cm layer after 5 years when a significance level of α = 0.10 was used. They did note that the number of samples required to estimate a 15% difference in bulk density on most of the sites was generally greater than the number of samples collected. For repeated measurement of bulk density over time to be dependable, many high-quality bulk density samples must be collected at each location and time. The standard agroforestry method of measuring bulk density by the core method recommends using large-volume rings with a diameter of at least 75 mm [29]. Validation of statistically significant differences in bulk density as a measure of soil recovery will be more dependable if the monitoring of the soil and climatic environment helps confirm that specific natural recovery processes were active [30].

Our objective was to quantify changes in bulk density over a 7-year period following forest harvesting and controlled skidding on boreal forest soils in west-central Alberta. We assumed that cold winters would make the freeze–thaw process an important factor contributing to the natural decompaction of these soils. Weather stations to measure air and soil temperatures and snow depth were installed at each installation to confirm when such events occurred.

#### **2. Materials and Methods**
