A mass^0.75 model shows 33% energy compensation for kangaroo rat exclusion on eight Portal plots in 2021–2025

We tested whether energy compensation for kangaroo rat exclusion returned to its 1997–2010 level at Portal, Arizona. We used four fixed control plots and four fixed exclusion plots from PortalData 6.117.0. Capture energy was mass raised to the power 0.75. Compensation was the summed increase in small-granivore energy on exclusions, relative to controls, divided by summed kangaroo rat energy on controls. In 44 complete surveys from June 2021 through December 2025, compensation was 32.5% (95% calendar-year bootstrap interval 22.9–44.2%). It was 51.5% in July 1997–December 2009 and 26.0% in January 2010–January 2020. The recent-minus-high contrast was −19.0 percentage points (−34.7 to −3.6). The recent-minus-previous contrast was +6.5 points (−5.1 to +20.0). Bailey's pocket mouse supplied 73% of the recent net gain. Different plot subsets, mass assumptions, and time averages kept the recent point estimate below the high-period estimate. An audit found repeated source rows. Their correction changed recent compensation by less than 0.1 point. Total energy ratios also include baseline small-granivore energy and target animals caught on exclusion plots. They must not be read as compensation. The results describe sampled surveys at one connected site. Energy is a proxy, detection is not estimated, and only five recent year groups are available. This is a reproducible update of earlier work, not a blind forecast test. Source tables, code, hashes, figures, and all results are preserved.

Portal Project · kangaroo rat exclusion · energy compensation · community dynamics · temporal replication · open data · data audit