CONSERVATION MANAGEMENT PRACTICE IMPACTS ON RANGELANDS IN CALIFORNIA

Abstract

Rangelands hold potential for mitigating climate change through soil organic carbon (SOC) storage. SOC plays a critical role in plant growth, soil structure and water retention, yet significant degradation of the world's soils poses major risks to forage production and water quality. To address this, California has promoted the adoption of conservation practices to restore SOC storage. Given California’s diverse climatic zones, climate-specific conservation strategies are necessary, as climate influences the effectiveness of different practices. These practices not only affect overall SOC stocks but also influence how SOC is stabilized in the soil, particularly through the formation of SOC fractions - particulate organic carbon (POC) and mineral-associated organic carbon (MAOC). POC generally contributes to the short-term carbon pool due to its rapid turnover however, when microbial activity is limited, its decomposition slows, allowing it to persist in the long-term carbon pool. In contrast, MAOC is more inherently stable and primarily associated with long-term carbon storage. This thesis investigated the effects of three conservation management practices - riparian restoration, tree plantings, and perennial seeding - on SOC storage in California rangelands. We used a retrospective paired-site analysis, comparing ‘restored’ (i.e., locations where a conservation practice was adopted) and ‘unrestored’ sites (i.e., a nearby similar location but lacking adoption of a conservation practice). Restored sites varied by the time since conservation practices were adopted, providing a chronosequence approach to estimate SOC and SOC fractions (POC and MAOC) change over time. While overall SOC differences between restored and unrestored sites were inconclusive, clear trends between practice types emerged within the restored sites. In drier regions, perennial seeding had higher POC stock compared to riparian restoration and tree plantings. Climate significantly influenced apparent SOC accrual in tree plantings, with a rate of 3.1 Mg C ha-1 yr-1 observed in moist climates, while in drier climates, SOC stocks were lower in tree planting sites compared to the unrestored sites. However, soil under tree canopies had 9% higher SOC content compared with soil sampled between trees, outside the tree canopy. Canopy cover appeared to promote proportional contributions to both POC and MAOC, highlighting the potential of tree plantings to increase SOC stocks, in relatively cooler, wetter regions. These findings underscore the importance of climate-specific conservation strategies for maximizing carbon storage in rangelands, particularly given the challenges inherent in managing these dynamic ecosystems. The variability in the apparent response to conservation practice adoption from the retrospective paired-site analysis raised questions about potential confounding factors. While this approach offers an alternative to long-term experiments by leveraging existing conservation practices, it introduces inherent uncertainties, particularly concerning prior disturbances that may influence SOC storage. A key assumption of the paired analysis is that vegetation and soils were approximately the same on both sites within a pair before the adoption of conservation practice. However, even when controlling for factors such as soil type, topography, and current vegetation, differences in past land use - such as disturbance events occurring at one site but not the other - could have led to SOC stock differences prior to when conservation practices were implemented. These historical land-use differences may obscure or exaggerate the measurement inferred impacts of conservation practices, highlighting the need to account for site history when interpreting SOC dynamics in retrospective studies. To address this, we analyzed remote sensing imagery to evaluate site conditions, prior to conservation practice adoption, identifying disturbance events and assessing vegetation cover and soil exposure from historical observations dating back to 1984. Our analysis revealed that 12 out of 36 paired sites experienced a disturbance event, on only one of the sites within a pair, including mastication, tillage, and burn events, potentially confounding the assumption of similar SOC stocks prior to the time of conservation practice adoption. Additionally, restored sites with significant pre-treatment differences in vegetation cover and bare soil exposure often originated from more degraded conditions compared to the unrestored site in the pair. This suggests a potential selection bias toward implementing conservation practices on more degraded lands, emphasizing the need to account for pre-existing site conditions in retrospective studies. Integrating remote sensing into paired-site analyses enhances the accuracy of assessments of conservation practice effectiveness assessments on SOC dynamics. This study underscores the importance of both climate considerations in conservation management and the value of remote sensing tools for improving SOC research methodologies.