Which counties and metropolitan areas had the largest increases in ICE arrests in 2025 compared with 2024?
Executive summary
A review of reporting and datasets compiled in 2025 shows the largest localized surges in ICE arrests occurred in specific counties and metropolitan areas rather than uniformly across the country: Denver County and several other Colorado counties led in raw counts (Denver 1,548) [1], Hawai‘i recorded roughly four times as many arrests through October 2025 as in all of 2024 [2], central Ohio’s 26-county area saw arrests triple in the first seven months of 2025 compared with the same period in 2024 [3], and Maryland’s total arrests nearly tripled—rising from 1,165 to 3,308 in the January–October window [4]. These local spikes align with national shifts in tactics and data that show broad increases in at-large and interior arrests across multiple field offices [5] [6].
1. The Colorado picture: Denver and front‑range counties led in volume
State and local reporting identifies Denver County as the single county with the highest number of ICE arrests in Colorado during the analyzed window—1,548 arrests—with El Paso, Arapahoe, Mesa, Adams and Pueblo following in descending order, marking Colorado as one of the states with both large absolute counts and notable year‑over‑year growth [1].
2. Hawai‘i’s disproportionate spike: quadrupling arrests through October
Hawai‘i’s available 2025 ICE operations data show roughly a fourfold increase in ICE arrests through October compared with all of 2024, with ICE booking 218 people into FDC Honolulu by Oct. 15 and monthly averages rising from about four arrests per month in 2024 to about 20 per month in 2025—illustrating a dramatic localized increase in enforcement activity [2].
3. Midwest hot spots: central Ohio’s triple increase and New York’s county surges
Central Ohio’s 26‑county region experienced a threefold rise in ICE arrests in the first seven months of 2025 versus the same period in 2024, a surge tied to high‑profile operations and local concern in Columbus and surrounding counties [3], while New York reporting highlights ballooning at‑large arrests in counties such as Monroe, Albany and Erie, with monthly totals in some periods reaching four times the prior year’s levels [7].
4. Maryland and North Carolina: near‑tripling and doubling in key metros
Maryland saw one of the largest proportional increases reported: 3,308 arrests from January through October 2025 compared with 1,165 in the same stretch of 2024—an increase of roughly 184% reported by regional outlets using Deportation Data Project figures [4]. In North Carolina, ICE arrests more than doubled in the first nine months of 2025, with about 770 arrests in the Triangle and adjacent counties and a notable rise in arrests of people with pending (rather than convicted) charges [8].
5. National context, tactics and data limits that shape the local picture
These county‑ and metro‑level spikes occurred against a national shift by ICE from reliance on local jails to tracking people down in communities—so‑called at‑large arrests—which helps explain sudden increases in interior metropolitan enforcement [5]. Analysts and data projects underpinning many local stories draw on ICE FOIA releases and the Deportation Data Project, but reporters note gaps and delays in official releases and that some local attributions rely on combining ICE office AORs with arrest location descriptions, creating limits to pinpointing every county precisely [9] [10].
6. What the available data do and do not show—caveats for interpretation
The datasets cited by local reporting and FOIA‑based projects provide strong evidence of large percentage increases in particular counties and metros through mid‑October and similar windows, but they are incomplete: ICE’s public dashboards cover through Dec. 31, 2024, while FOIA releases and the Deportation Data Project extend into 2025 with varying geographic detail, and several outlets warn that exact full‑year tallies and rigorous per‑capita comparisons depend on successive government releases and methodological choices [6] [9] [10].