Resistance and resilience are commonly used terms in discussions about agriculture and future ecological conditions.

We provide a common understanding of these terms as they apply to the ecology of grazed systems.

We discuss the relationship between ecological resistance and resilience, disturbances, and ecological processes.

We present examples of how to make what is known about ecological resistance and resilience in grazed ecosystems accessible to planners and their clients.

Conservation policies promote sustainable grasslands management practices, like rotational or management-intensive (RMI) grazing, in the United States.

The self-reported adoption of RMI grazing declined in the United States between 2007 and 2017, and few studies have investigated this trend.

We use panel data to estimate county-level fixed effects regressions to assess if changes in the size of cattle operations are influencing this trend. We estimate a regression for the United States, as well as regressions by Climate Hub region.

We find nationally, as well as within six of the eight Climate Hub regions, an increase in cattle operations with <20 head leads to a smaller increase in RMI grazing relative to an increase in cattle operations with 20-199 head. However, this effect is reversed in the Northeast.

Nationally, we find similar effects on RMI grazing among cattle operations with 20-199 head and ≥200 head. Still, in two of the Climate Hub regions (Midwest and Southeast), cattle operations with 20-199 head lead to greater RMI grazing adoption than cattle operations with ≥200 head. Thus, declines in RMI grazing are most strongly associated with declines in small and medium-sized cattle operations.

Peak standing crop (PSC) and residual dry matter (RDM) are the primary measures of production and grazing intensity on California's annual rangelands.

One of the most common methods of monitoring forage metrics is to clip 1-ft² quadrats. The USDA Forest Service, Bureau of Land Management, universities, and other land managers have been using this methodology since the 1930s.

We used best linear unbiased predictors (BLUEs) to determine 95% confidence intervals for PSC and RDM. For both PSC and RDM, as the number of samples taken increased from 1 to 10, the predictive ability also significantly increased. We found no evidence of increased predictive power past 10 samples.