Torrance County Detention Facility’s Troubling Role in Detaining Haitian Migrants During the 2021 Del Rio Incident
Torrance County Detention Facility Troubling Role in Detaining
Haitian Migrants During the 2021
Del Rio Incident
Published: October 24, 2024
Executive Summary
The Torrance County Detention Facility (Torrance) is one of approximately 200 facilities across the United States where Immigration and Customs Enforcement (ICE) detains immigrants with pending removal proceedings. Located in the rural New Mexico desert, this detention facility holds a notorious reputation for its inhumane living conditions and for the speed at which individuals detained there go through their removal proceedings, often without adequate legal counsel. Despite these noted abuses, in September 2021, ICE placed over 100 Haitian migrants into Torrance while they awaited removal proceedings. For months thereafter, the media continued to report on a variety of obstacles Haitians endured in detention including inadequate access to legal counsel.
Due to the increase in complaints from individuals being held at Torrance, the American Immigration Council (The Council) sought to ascertain whether particular barriers to due process exist for Haitian nationals, as well as to investigate the overall treatment of Haitian nationals at Torrance. The request sought data related to individuals detained at Torrance from January 1, 2021, including arrest/apprehension information, immigration status, biographic information, detention history, and release information. ICE responded by providing data between January 1, 2021 and November 17, 2022, and this is what we found when we analyzed it:
- ICE's Use of Racial Classifications Are Unreliable, Labeling Most Detained Individuals “White:” Individuals in detention at Torrance represented 54 different countries spanning five different continents. However, 86 percent of individuals detained at Torrance were categorized as racially “white.” The data suggests that ICE failed to systematically document the race of detained individuals.
- Africans Had the Highest Lengths of Detention at Torrance: Because ICE’s race categorizations proved unreliable, researchers grouped detained individuals by continent to measure the impact geographic location has on detention lengths. The data showed that African migrants had the highest lengths of detention.
- ICE Officers Continued to Populate Torrance Despite Multiple Warnings: During the reviewed time period, ICE had substantial warning signs that Torrance was not equipped to house detained migrants through failed inspections, COVID surges, staffing shortages, and even government oversight agency reports recommending shutting the facility down. Despite these warning signs, the data showed that ICE continued to detain migrants at Torrance, putting them at risk.
- Oversight Efforts Seemingly Reduced the Detained Population at Torrance—But Only Temporarily: The data shows that between August and November 2022, a period that included the suicide of Kesley Vial at Torrance and a government report calling for the closing of Torrance, the population of Torrance consistently decreased. However, in December 2022, ICE began repopulating the facility.
Background
Torrance County Detention Facility
Estancia, New Mexico is a small, rural town roughly 40 miles southeast of Albuquerque. Composed of merely 1,279 residents, Estancia is home to the Torrance County Detention Facility, one of New Mexico’s three operating centers detaining immigrants. Torrance has jailed individuals in the custody of ICE, U.S. Marshals Service, the State of New Mexico, and Torrance County. On May 15, 2019, ICE entered into an Intergovernmental Service Agreement (IGSA) with Torrance County, agreeing to maintain a minimum population of 714 at the detention facility. As a detention facility, Torrance has garnered notoriety for how quickly individuals detained there are passed through the removal process, which has raised concerns about violations of due process.
Haitian Migrants at Del Rio
Despite the Torrance facility’s problematic reputation, ICE has continually populated the facility with vulnerable populations. Among these were Haitian migrants apprehended in Del Rio, Texas in September 2021.
Around September 9, 2021, hundreds of migrants of different nationalities arrived at the Del Rio International Bridge in southwestern Texas. The majority were Haitian immigrants who had been living in South America, but the group also included other nationalities. By September 18, approximately 15,000 Haitian migrants had formed an encampment at the Bridge, hoping to seek asylum. The influx of migrants in Del Rio led DHS to announce a new strategy on September 18 surging agents and officers to the border, as well as increasing the pace of repatriation flights to more aggressively clear out the encampment.
On September 19, media outlets began to release shocking images demonstrating the aggressive application of the policy on Haitian migrants. Images appeared of Border Patrol agents on horseback seemingly using their reins to strike Haitian nationals as the migrants attempted to exit the dangerous Rio Grande, crossing into Del Rio. The next day, DHS launched an investigation into the incident. However, immigration enforcement agencies incarcerated hundreds of Haitian asylum seekers in detention centers across the country where they continued to face inhumane treatment.
The Council Began Its Efforts to Advocate for Haitians in Detention at Torrance
In November 2021, the American Immigration Council, in partnership with several organizations including the Haitian Bridge Alliance, Innovation Law Lab, The ACLU of New Mexico, and the National Immigration Project, began a campaign to allow legal access to the Haitians detained at Torrance, who were being sped through the deportation process without due process. There was a serious concern that they would be removed before they could give evidence in the Del Rio investigation. This also led to a letter from New Mexico’s congressional delegation to DHS asking for increased oversight and delineating multiple issues at Torrance.
The Council Filed a Freedom of Information Act (FOIA) Request
On December 21, 2021, the American Immigration Council filed a FOIA request for records related to ICE treatment of individuals held at Torrance. Given the general lack of information about the treatment of people in immigration proceedings broken down by nationality, the Council sought to ascertain whether particular barriers to due process existed for Haitian nationals, as well as to understand the overall treatment of Haitian nationals in detention facilities. The request sought data related to individuals detained at Torrance from January 1, 2021, including arrest/apprehension information, immigration status, biographic information, detention history at each facility, and release information.
In August 2022, the Council filed a lawsuit to compel the agency to respond to the organization’s FOIA request. ICE released data on detention stays at Torrance between January 1, 2021, and November 17, 2022. The dataset provided a complete overview of all individuals detained at Torrance within the designated timeline, as well as detention facilities they were placed into before and after entering Torrance. The data showed that between September and November 2021, over one hundred Haitian migrants ended up at the Torrance facility.
| Data | Description |
| Apprehension Information | Information about a detained individual’s apprehension program, entry status, and at what date and time they were apprehended. |
| Biographic Data | A detained individual’s background including race, ethnicity, gender, birth year, as well as birth and citizenship countries. |
| Stay Information | The total time an individual spent in detention, including all facilities and transfers. This section provides information on when a detained individual was placed into detention, when they ended their detention, and why they were released from detention. |
| Detention Information | Information about an individual’s stay in specific detention facilities, including the dates and times an individual was placed into a given facility, as well as the time and reason for transfer to a different facility. |
| Bond Information | Bond information shows the date a bond was set, and in some cases, how much the bond was set for. |
Methodology
Researchers implemented a mixed-methods approach to analyzing the data. On the quantitative side, researchers created a daily timeline documenting total ICE placements and removals into and out of Torrance, as well as the estimated population of Torrance per day. In addition, the research analyzed whether biographic data had an impact on average days individuals spent in detention. Qualitatively, researchers looked at emails and agency correspondence produced as a result of the FOIA request, as well as individual detention stories within the dataset. In particular, researchers used Innovation Law Lab (ILL)’s timeline of Torrance County Detention Facility to build the complete day-by-day timeline of events across this approximately two-year period. This timeline lists chronologically some of the events in the history of abuse at Torrance County Detention Facility, linking documents produced as a result of ILL’s own FOIA requests. Researchers merged the quantitative and qualitative aspects of the research to form a single timeline across the two-year period.
ICE provided the Council with an Excel file titled “ICE Detention Stays that Include Detention at Torrance County Detention Facility between 1/1/2021 and 11/17/2022.” The data included 5,336 total rows and 29 columns. ICE had redacted four of these 29 columns: birth date, detention facility code, person ID, and detention stay ID, under FOIA exemptions B(6), B(7)(E) and B(7)(C). Additionally, while there were three columns on custody information (action, action date, and decision), there was no data present. The data provided an overview of all detained individuals placed into Torrance within the designated timeline, as well as detention facilities they were placed into before and after entering Torrance.
Sample Data for Torrance Report.xlsxUnique ID Creation
Each row in the dataset included what time an individual was apprehended, when and where their detention began, when and where the individual was transferred, and when and where their detention ended. Researchers found there were multiple rows of data per detained individual, but because ICE had redacted detained individuals’ identification numbers, researchers created their own unique IDs. This process involved concatenating 11 variables: apprehension date, entry status, gender, race, ethnicity, birth country, citizenship country, birth year, stay book-in date time, stay book-out date time, stay release reason. In particular, apprehension date time, stay book-in/out date time, and birth year were highly specific, yielding more distinct IDs. Researchers concluded that individuals with rows with the same formulated ID indicated the same individual. Finally, researchers implemented four tests to verify the accuracy of the created IDs. These tests included checking the timings, detention and stay release reasons, and potential duplicates, as well as accounting for individuals placed into Torrance twice. The findings yielded 2,505 distinct IDs representing approximately 2,505 individuals detained at Torrance during this time period.
Timeline Formation
After filtering the column on “Detention Facility” to only include information on Torrance, researchers manipulated the “Book In Date Time” and “Book Out Date Time” categories to include only the date rather than both date and time. As such, the entry “5/14/2021 5:04 AM” became “5/14/2021.” Next, researchers created a pivot table with the count of each date. For example, there were 133 instances of 5/14/2021 indicating that ICE booked 133 people into Torrance on May 14, 2021. Finally, researchers created a day-by-day timeline during the two-year period adding the designated count in the database for each date. If a date did not have a count, researchers assumed there were no book-ins/book-outs on that day.
Estimated Population
“Estimated Population” refers to the estimated count of detained individuals at Torrance per day. Researchers calculated a dozen individuals in Torrance at the start of the timeline, with a steady decrease in the population until mid-May of 2021. This finding coincided with reports that fewer than a dozen individuals were in Torrance at the beginning of 2021. Using the estimated population on January 1, 2021, researchers then calculated the population on every other date by adding the total “Book Ins” and subtracting the total “Book Outs” on a designated day. For example, the estimated count at Torrance on August 4, 2021, was 469. On August 5, 2021, there were 24 book-ins and 19 book-outs, leading to the conclusion that the estimated population of Torrance on this day was 474. Researchers were able to verify some of the estimated count through TCDF ICE Reports Rep. Melanie Stansbury published between April 20, 2022 through November 4, 2022.
Organization
This web report serves four purposes. The first section analyzes the characteristics of detained individuals at Torrance between 2021 and 2022. The second section provides a macro-level overview of placements and removals from Torrance, as well as the estimated count of detained migrants at Torrance County Detention Facility throughout 2021. The third section focuses on Haitians apprehended in Del Rio and later placed into Torrance. The fourth section chronicles continuing abuses in 2022.
Analysis of Detained Individuals
Overview
The data received from ICE shows that between January 1, 2021, and November 17, 2022, there were an estimated 2,505 individuals detained at Torrance. Detained individuals represented 54 different nationalities spanning five different continents. On average, detained individuals spent slightly over two months in ICE detention at any facility (64.64 days), with an average of about six weeks (44.56 days) spent at Torrance in particular.
Locations and Transfers
Approximately one-third of the individuals detained at Torrance during this period spent the entirety of their detention stay at Torrance. The remaining two-thirds were transferred into and out of Torrance from other facilities. The data showed that ICE placed the detained individuals represented in this data into a total of 55 different locations spanning 44 cities and 15 distinct states during this period. A striking pattern at Torrance was that over two-thirds of detained individuals saw Torrance as their last or only stop, with the remaining one-third transferred into another detention facility prior to the conclusion of their detention. It is not uncommon for immigrants detained by ICE to experience inhumane living conditions. Government oversight agencies have repeatedly noted deficiencies in the care and living standards at numerous immigration detention centers throughout the country. Torrance, however, was the only detention facility that received a management alert from a government oversight agency that explicitly called for its closing due to violations of detention standards.
Race and Ethnicity Analysis
Race and Ethnicity
The data demonstrated that ICE categorizations of race and ethnicity were particularly unreliable. The ICE data reviewed on race and ethnicity failed to accurately measure detained individuals’ demographic information. As a result, the data inevitably skewed racial analysis of detained individuals. The findings also showed that ICE inadequately maintained data on race and ethnicity, which impacted researchers’ ability to draw firm conclusions from the available variables.
Honduras, India, Bangladesh, Peru, Senegal, El Salvador, Uzbekistan, Dominican Republic, Nepal, Russia, Sri Lanka, China, Peoples Republic of, Armenia, Bolivia, Guinea, Yemen, Argentina, Belize, Costa Rica, Georgia, Iran, Jamaica, Mali, Sudan, Ukraine, Afghanistan, Albania, British Virgin Islands, Cambodia, Cameroon, Chile, Gambia, Ghana, Kyrgyzstan, Mauritania, Netherlands, Nigeria, Pakistan, Panama, Rwanda, South Korea, Spain, St. Lucia, Syria
Issues Encountered with Ethnicity
Most federal agencies collect data on individuals’ race and ethnicity according to guidelines published by the Office of Management and Budget (OMB) in 1997. The 1997 standards recommend collecting this information through two separate questions: one for ethnicity and the other for race. In March 2024, the OMB announced changes to the 1997 standards, one of which would combine both questions by only asking about race and moving “Hispanic or Latino”—previously captured in the ethnicity question—to the race question to improve data collection. Because the OMB recommendations have yet to take effect, and the Torrance data is from 2021 and 2022, the ethnicity category in the dataset contains the labels Hispanic Origin, Not of Hispanic Origin, and Unknown.
The main problem with ICE’s ethnicity data, however, is that the agency reported ethnicity for only four percent of detained individuals. Due to the nearly complete lack of data, developing conclusions on ethnicity proved difficult. The data showed that there seemed to be a correlation between the agency affiliation of officers who had apprehended the individual and the likelihood that ethnicity would be recorded. Around 90 percent of individuals detained were apprehended by U.S. Border Patrol; however, not a single one of these records included ethnicity. Conversely, ethnicity was documented at a higher frequency when the data showed the apprehension was categorized under ICE’s Enforcement and Removal Operations (ERO) Criminal Alien Program, Fugitive Operations, or Homeland Security Investigation (HSI) Criminal Arrest.
Race and Nationality
The 1997 OMB standards on race data recognized five racial categories: White; Black or African American; American Indian or Alaska Native; Asian; and Native Hawaiian or Pacific Islander. In contrast, the Torrance data showed that ICE classified detained individuals into fewer categories: Asian or Pacific Islander (A/PI), Black, White, or Unknown. Significantly, under both the 1997 OMB standards and ICE categories, individuals from Arab-majority countries would often be categorized as “white.” One of the other changes announced by OMB to the 1997 standards is to add “Middle Eastern or North African” as a new minimum reporting racial category.
ICE officers documented a racial category for over 99 percent of detained individuals. To understand how ICE officers assigned racial categories to detained individuals, the researchers examined the intersection of race and nationality to identify any patterns in these designations. The following graphs illustrate the relationship between nationality and the racial categories assigned by ICE. The first graph displays nationalities with a single racial designation; for instance, all Albanians were classified as white, and all Cameroonians as Black. The second graph highlights nationalities with two racial designations; for example, some Haitians were categorized as white, while others were classified as Black, and similarly, some Armenians were identified as white, while others were categorized as Asian/Pacific Islander (A/PI). The final graph depicts nationalities with three racial categories; for example, Indians were categorized as white, Black, or A/PI, while Colombians were classified as white, Black, or Unknown. No nationality was categorized into more than three racial categories.
| One Race | Nationalities |
|---|---|
| White | Albania, Argentina, Belize, Chile, Costa Rica, Georgia, Iran, Mali, Mauritania, Netherlands, Panama, Peru, Russia, Spain, Syria, Turkey, Ukraine, Uzbekistan, Yemen |
| Black | British Virgin Islands, Cameroon, Gambia, Ghana, Guinea, Jamaica, Nigeria, Rwanda, St. Lucia, Sudan |
| A/PI | Cambodia, China, Kyrgyzstan, South Korea, Pakistan |
| Unknown | Afghanistan |
| No Data |
| Two Races | Nationalities |
|---|---|
| White or Black | Brazil, Dominican Republic, Haiti, Senegal, Sri Lanka |
| White or A/PI | Armenia, Bangladesh, Nepal |
| White or Unknown | Bolivia, Ecuador |
| White or No Data | El Salvador |
| Three Races | Nationalities |
|---|---|
| White, Black, or A/PI | India |
| White, Black, or Unknown | Colombia, Cuba, Nicaragua, Venezuela |
| White, Unknown, or No Data | Guatemala, Honduras, Mexico |
Problematic Use of “White” as a Racial Categorization
According to the data, the racial category "white" encompassed a diverse range of nationalities. All individuals from European countries such as Albania, Spain, Ukraine, and the Netherlands were classified as racially white. Additionally, all individuals from the transcontinental countries Georgia, Russia, and Turkey were also categorized as white. All individuals from several Latin American countries, including Argentina, Belize, Chile, Costa Rica, Panama, and Peru, were similarly classified. Of over 2,500 individuals at Torrance representing 54 distinct nationalities spanning five different continents, more than 2,100 individuals—86 percent—were categorized as racially “white.”
However, all individuals from many Middle Eastern countries, such as Iran, Syria, Turkey, and Yemen, were also categorized as white. Notably, all individuals from two multi-ethnic countries, Mali and Mauritania, were also classified as racially white. The lack of specificity in the race categories, particularly as it relates to the nationalities labeled as “white,” is problematic. Studies show that Middle Easterners and North Africans are not perceived, and often do not perceive themselves, to be white. This “White” categorization failed to capture the lived experiences of discriminatory treatment that people from Middle Eastern or North African countries face.
The same can be said about individuals from Latin American countries classified as “white.” In theory, a clearer picture could be gleaned for nationals of Latin American countries because ICE should have classified those individuals by also using the ethnicity categories. However, ICE failed to consistently note individuals’ ethnic background.
Accordingly, it is difficult to develop conclusive findings solely from the racial categorizations used by ICE. The analysis of the racial categories in the dataset revealed no clear pattern in how ICE officers made certain determinations. It is also unclear as to whether detained individuals self-reported race, or whether ICE officers themselves reported detained individuals’ races.
Disparities Between Racial Groups of a Given Nationality
While ICE’s categorizations of race and ethnicity seem inadequate to capture individuals’ lived experiences, researchers sought to determine whether disparities existed between detained individuals of the same nationality with different documented races. For example, researchers sought to examine whether detained individuals from Haiti labeled as racially white had lengths of detention lower or higher than detained individuals from Haiti labeled as racially Black. The data showed that in many instances, individuals marked as racially Black, A/PI or Unknown had higher lengths of detention at Torrance than their white counterparts. Armenians, Nepalese, and Bangladeshi nationals, for example, were marked as either racially A/PI or white. In examining lengths of detention between these nationalities and racial groups, the data demonstrated that Armenian and Nepalese nationals marked as A/PI had significantly higher lengths of detention than their counterparts marked as white.
There were five countries from which individuals were classified as either Black or white in ICE’s racial data: Brazil, Dominican Republic, Haiti, Senegal, and Sri Lanka. In four of these five instances, detained individuals marked as racially Black spent more time in detention at Torrance than their counterparts marked as white. The case of the Dominican Republic particularly demonstrated this disparity. The data showed that there were an estimated 13 individuals from the Dominican Republic, with approximately half categorized as Black (six people) and the other half categorized as white (seven people). However, individuals from the Dominican Republic categorized as racially Black had lengths of detention at Torrance 40 percent higher than their counterparts categorized as white.
Racial Analysis of Individuals Detained at Torrance
Impacts of Classifying a Broad Range of Individuals as “White”
Researchers sought to analyze the relationship between race and average days spent in detention. Overall, the data showed that average time in detention for those individuals who had been marked as Black was comparable to that of individuals marked as white. A closer look at the average lengths of detention in tandem with nationality tells a different story. Citizens of the non-European countries from which every individual was categorized as “white” had some of the highest lengths of detention. Yemenis, a nationality that should be classified as Middle Eastern under the 2024 OMB guidelines, had lengths of detention at Torrance 687 percent higher than the total average. Detained individuals from Latin American countries including Chileans, Belizeans, Costa Ricans, and Argentinians had lengths of detention at Torrance 289 percent, 115 percent, 82 percent, and 59 percent higher than the total average, respectively. Mauritanians, who were also racially categorized as white, had lengths of detention 10 percent higher than the total average found in the dataset.
Africans Had the Highest Lengths of Detention
The overwhelming categorization of detained individuals as racially white showed that ICE’s documentation of race and ethnicity is insufficient to demonstrate racial and ethnic disparities in ICE’s treatment of detained individuals. In order to create a more accurate category to measure biographic data, researchers analyzed whether grouping individuals by continent demonstrated any disparities in detention lengths. The purpose of using continents in this analysis was to replace the demonstrably unreliable ICE data on race with a more objective categorization to examine patterns in average lengths of detention. The rationale for grouping individuals by continent rather than at the country level was to allow for a sufficient sample size to draw conclusions from the data.
The data showed that African migrants were not only spending the highest amount of time in detention, but were also spending the majority of their stays specifically at Torrance. Africans spent approximately 44 percent more time in detention and spent 22 percent more of their time in detention specifically at Torrance, compared with the overall average.
Few Europeans Detained at Torrance, but Those at the Facility Spent the Least Time There
Of the over 2,500 estimated individuals at Torrance during the available time period, there were only about fifteen European nationals at the facility representing merely 0.5 percent of detained individuals, which is an extremely small sample size from which to draw effective conclusions. Europeans, in fact, made the smallest sample size of all geographic locations while North Americans (consisting of mostly Central American countries and the Caribbean) made the highest. An analysis of this small number demonstrated that compared to other geographic categorizations, individuals from European countries had longer average detention stays, but spent far less of their time at Torrance than members of other groups. On average, Europeans were spending only about 50 percent of their time in detention specifically at Torrance. This is the lowest rate of all continents. On the other hand, Asian migrants were spending an average of up to 77 percent of their time in detention at the facility. Further research based on greater access to race data, more accurate data, and a larger sample size is necessary to make more definitive conclusions about whether racial disparities exist.
Placements into Torrance in 2021
The analysis on detention trends at Torrance looks at ICE placement of migrants into Torrance.
May 2021: Detained Migrants Are Funneled into a Barely Populated Torrance
While the requested data begins on January 1, 2021, the timeline really began on May 14, 2021. Between January 1 to May 13, 2021, there were few detained individuals at Torrance. On January 1, there were an estimated 14 people at the facility and by May 13, there were an estimated five people at the facility. Between these five months, there was a steady decline in detained individuals at the facility.
However, on May 14, 2021, ICE began abruptly repopulating Torrance. On that day, ICE placed 133 migrants into the facility. Between May 14 and June 1, ICE placed close to 500 people into Torrance—sending between 90 and 133 individuals to the facility on each of four days during that period. The graphs show a variance in apprehensions, book-ins, and book-outs at Torrance. On one end, apprehensions of migrants later placed into Torrance are slower and more gradual, whereas book-ins are sharp and concentrated. Book-outs during the month are virtually non-existent, indicating a rapid repopulation of the facility.
By June 1, there were an estimated 484 people at the facility. This is the highest number of detained individuals in Torrance within the approximately two-year timeline encompassed by the data.
The graph showing a sudden repopulation of Torrance in May correlates with national statistics regarding the overall rise in migrants placed into ICE detention centers in 2021. The number of people in ICE detention at the start of 2021 declined to around 15,000, down from an all-time high exceeding 50,000 people in the summer of 2019. Figures show that there was a sharp increase in the number of people booked in to ICE detention centers between April to May of 2021, and placement of migrants into Torrance was consistent with this overall trend.
ICE Placed Migrants into Torrance Despite Warnings
The timeline of ICE placements demonstrates that ICE chose to continue placing detained migrants into Torrance despite warnings from watchdog entities of operational deficiencies and inadequate health conditions.
Timeline of Events in 2021
May 3 to May 7: ICE Noted Nine Operational Deficiencies During an Inspection
On May 3, the ICE Office of Inspections and Detention Oversight Division (ODO) conducted a compliance inspection of Torrance for the first time since 2020. At this point, there were only approximately five migrants housed at the facility. However, the inspection noted nine operational deficiencies related to security and care. Most notably, the ODO found that the facility’s detainee handbook had “no notification to detainees on how to access personal funds to pay for legal services,” a violation of ICE detention standards, and an example of the continued pattern at Torrance of hindered access to legal counsel. The aggressive repopulation of Torrance described in the prior section occurred exactly one week after these clear transgressions in operational compliance were documented.
May 13: Torrance Faced a Lawsuit for Negligence and Human Rights Abuses That Occurred In 2020
The American Civil Liberties Union (ACLU) of New Mexico filed a lawsuit against the facility for its negligence and human rights abuses that occurred in relation for a peaceful hunger strike that began at the facility on May 11, 2020. Approximately 20 detained individuals protested inadequate precautions against COVID-19, poor living conditions, and withholding of status updates on their immigration cases. In a disturbing video taken three days later, CoreCivic facility staff responded to the protest with extreme force, spraying protesters with chemical agents and holding them in an enclosed room for several minutes. The May 13, 2021 lawsuit alleged that CoreCivic staff improperly decontaminated protestors by using chemical agents that remained on individuals’ skin for days, and several individuals contracted COVID-19. According to the lawsuit, two protesters with a history of mental illness attempted suicide. While the ACLU-NM lawsuit marked the one-year anniversary of this incident, the data obtained through the FOIA request show that on this date, ICE officers began repopulating the facility with hundreds of migrants.
May 18 to June 9: COVID-19 Cases at Torrance Surged
ICE protocols claimed that the agency tested all incoming detained individuals for COVID-19 and separated new arrivals from the general population for fourteen days. However, each time ICE placed groups of migrants into Torrance between May 14 and June 1, COVID cases surged. On May 18, after ICE placed two groups at Torrance, there was a COVID outbreak infecting 110 of the approximately 270 individuals detained at the facility, as well as 16 staff members. Despite the COVID outbreak, ICE placed another 93 individuals into the facility on May 20. On May 26, after three mass placements had already occurred, New Mexico Health reported 157 COVID cases at the facility. Notwithstanding these reports, on June 1, ICE decided to place an additional 133 individuals into the facility. By June 22, the New Mexico Health Department reported that the number of COVID cases at Torrance rose to 204.
July 27 to July 29: Torrance Failed its Annual Inspection Due to Staffing Shortages
On July 27, the Nakamoto Group—a private entity contracted by ICE to certify compliance with detention standards—began its annual inspection of Torrance. On July 27, the first day of the inspection, ICE placed 196 people into Torrance—the largest mass book-in during the approximately two-year timeline of the data produced by ICE, bringing the population of Torrance from an estimated 230 to 420. Torrance’s failed Nakamoto Group inspection is notable because such inspections are notoriously lax. Unlike DHS OIG itself, which conducts unannounced inspections, the Nakamoto Group only conducts scheduled ones.
Nevertheless, on July 29, the Nakamoto Group determined that Torrance “Does Not Meet Standards.” This made Torrance one of three facilities out of 200 that failed its inspection. The Nakamoto Group identified 22 deficient components across eight standards, including health and safety, visitation rights, and self-harm prevention. Food safety and sanitation practices in particular were found to be failing ICE standards, and the Nakamoto Group found that Torrance was staffed at only 50 percent capacity of the authorized “correctional/security” positions, with staff working mandatory overtime shifts. And yet, well after the inspection concluded and inspectors informed ICE officers of the facility’s deficiencies, ICE placed another 88 migrants into the facility. Despite the extreme staffing shortages noted in the inspection, the data showed that there were about 474 individuals at Torrance on July 29, slightly over 50 percent of the maximum capacity.
November 16 to 18: ICE ODO Found 21 Operational Deficiencies at Torrance
On November 16, ICE ODO conducted its follow-up inspection of Torrance, six months after the May ODO inspection. The agency found 21 operational deficiencies in Torrance, an increase of 12 from its last inspection, and noted numerous instances of inadequate physical and mental health care.
Attorney Access Issues
In addition to warnings about the dismal conditions of confinement at Torrance, ICE had information about the difficulties individuals detained there encountered when trying to obtain legal representation. The American Immigration Council sent a letter on November 5, 2021, to ICE personnel requesting that Torrance grant access to attorneys because “conditions at Torrance ha[d] effectively prevented Haitian immigrants…from retaining or communicating with legal counsel.” The letter also noted that Torrance staff denied requests to conduct group presentations about legal rights, as well as other requirements laid out in the Performance Based National Detention Standards. The complaint filed by immigrant rights groups including the American Immigration Council with DHS Office for Civil Rights and Civil Liberties on November 23, 2021 also detailed obstacles encountered by attorneys when trying to reach individuals detained at Torrance, including delays to access because the staff member who scheduled the calls was “out sick” or “really busy.” In a report published by the ACLU on June 9, 2022, an attorney states that scheduling phone calls with clients at Torrance "really just depends on what is going on and the scheduler’s mood and how many people are detained.”
These advocacy efforts served as a basis for the December 16, 2021 letter members of Congress sent to DHS Secretary Alejandro Mayorkas and ICE Acting Director Tae Johnson asking DHS to conduct comprehensive oversight of Torrance to ensure attorney access to people detained there.
Access to attorneys for immigrants in detention is often precarious, leading to a lack of adequate counsel before court hearings. The complexity of immigration law coupled with the high evidentiary standards for presenting asylum claims makes self-representation extremely difficult. Data shows that detained migrants who have access to legal counsel are 10 times more likely to win relief, and almost seven times more likely to be released from custody, than those without counsel. However, detained migrants are often unable to access counsel in the first place. Detention facilities in remote locations, such as the Torrance facility, are often far from legal service providers. Consequently, many detained migrants lack adequate legal counsel well into their immigration proceedings.
Despite these known obstacles to securing legal representation at Torrance, the data showed that ICE placed over 100 Haitian nationals in detention at Torrance between September 21, 2021, to November 16, 2021.
Haitians at Torrance
In light of the serious allegations that dogged the Torrance County Detention Facility, it is important to analyze the timeline of when Haitian nationals ended up at Torrance.
ICE Was Not Initially Using Torrance to House Haitian Migrants Apprehended at Del Rio
On September 9, 2021, hundreds of Haitian migrants coming mainly by way of South America began forming an encampment under the Del Rio Ciudad Acuña International Bridge, while CBP officers processed the arriving migrants. Many of these migrants had fled dangerous conditions with the hopes of seeking asylum in the United States. As the number of migrants grew to around 10,000, on September 15, DHS ramped up its repatriation flights to Haiti to execute the expulsions and removals of Haitian nationals, which in many cases deprived migrants of their right to seek asylum. By September 18, immigration enforcement agencies estimated that numbers at the Del Rio encampment peaked at 14,921, prompting DHS to publish a written policy titled “Strategy to Address Increase in Migrants in Del Rio.”
The policy consisted of six key components including a surge in agents to the Del Rio sector and an increase in the pace and capacity of removal flights to Haiti. The main outcome of this policy was the forcible clearance of the encampment by immigration enforcement agencies. Between September 18 and September 25, DHS cleared the encampment of its nearly 15,000 inhabitants. While the majority of Haitian asylum seekers were either repatriated or pushed back into Mexico, hundreds from the encampment were sent to various jails, prisons, and detention centers across the country.
Documents and data obtained from the FOIA request show that between September 9 and September 20, detention centers such as El Paso SPC, Otero County DC, Karnes FSC, and Port Isabel DC were taking in Haitian migrants; the latter two facilities each taking 100-150 migrants daily. Other detention centers such as the South Texas Family Residential Center in Dilley, Texas were taking in families.
However, the data showed that ICE began placing Haitians into Torrance after the September 18 policy change. Between September 9 and September 20, there was not a single Haitian migrant at Torrance. Small numbers of Haitian nationals had been placed at Torrance in May, June, and July, but the last Haitian national had been removed from Torrance on August 28, 2021. The next placement of Haitian individuals at Torrance happened on September 21, when 58 Haitian nationals were detained there.
On September 17, a day before the policy change, immigration enforcement agencies documented an estimated 12,709 encounters of Haitian migrants at the encampment. The number of migrants at this port of entry prompted the City of Del Rio to declare a state of emergency. The data showed that no migrant apprehended on September 17 was ever placed into Torrance. However, a surge in agency correspondence on this day highlighted the high cadence of single adults at the Del Rio port of entry. According to the emails, Haitian migrants were arriving at Del Rio at a frequency so high that the agency was unable to process them in time to be placed on repatriation flights.
The emails specifically noted that these single adults who would not be processed in time for the next flight were “detention eligible.”
The following day, ICE agents began rapidly placing single adults from Haiti into detention centers across the country.
ICE's Use of Torrance to Detain Haitians Coincided with the DHS Policy Change
The dataset from the FOIA request demonstrated the intensity of the September 18 policy change. The DHS Strategy to Address Increase in Migrants in Del Rio led to the prompt deportation of thousands of Haitian migrants in Del Rio. Immigration enforcement agencies began pushing migrants back into Mexico or placing them on repatriation flights. On September 20, videos began emerging of the events between September 18 and 19 showing the aggressive nature of immigration enforcement tactics in Del Rio. Multiple news sources released images showing CBP using violent tactics on Haitian migrants between these two days. One of the most shocking images was of a CBP officer on horseback using his reins to forcibly stop a Haitian migrant.
In looking at apprehensions by date and time, the data showed that on September 18 at 9:12 PM, 37 migrants (all Haitian nationals) were apprehended and subsequently placed into Torrance in the following days. This is the largest apprehension surge at a particular date and time in the dataset, demonstrating the intensity of the effects of the policy change. At the same time the policy was introduced, Haitian nationals, who had not been detained at Torrance for around one month, began to be placed at Torrance. On September 21, 58 Haitians who had been apprehended in Del Rio, Texas on September 18 and September 19 were placed into Torrance.
Transfers into Torrance Followed a Pattern
A total of 106 Haitians apprehended from September 9 to September 24 were eventually detained at Torrance. Fifty-eight Haitians apprehended in Del Rio, Texas on September 18 and September 19 were placed into Torrance on September 21. After this date, there were no direct placements of Haitians into Torrance; however, detained Haitian migrants were transferred into Torrance from other facilities. The remaining 48 Haitians apprehended between September 9 and September 17, and between September 20 and September 24 in the dataset were transferred into Torrance between September 30 and November 16. The data showed that subsequent transfers occurred on September 30, October 16, October 30, and November 16.
Haitians Apprehended in Del Rio Remained in Detention for Longer Than Others at Torrance
According to the data, there were an estimated 177 Haitian nationals detained at Torrance between January 1, 2021 and November 17, 2022. This made Haitians the fifth-most common nationality in the dataset. Approximately 106 of these 177 Haitians were apprehended between September 9 and September 24 and subsequently placed into Torrance. The data indicated that overall, Haitian nationals spent around nine weeks in detention, approximately six weeks of which were spent specifically at Torrance. This is roughly the same amount of time as the average for non-Haitians. However, Haitians apprehended in Del Rio did not conform to this pattern.
Haitians apprehended before September 9 or after September 24 spent significantly less time in detention than Haitians apprehended during this time period: these two groups spent an average of 50.68 days in detention and were at Torrance for 26.05 of these days. On the other hand, Haitians apprehended between September 9 and September 24 spent approximately 66 days in detention, and 54.63 of those days were spent specifically at Torrance. When looking at Haitians specifically apprehended between September 18 and 19, the length of stay in detention was 62.78 days overall and 61.21 days at Torrance. The data showed that Haitians apprehended in Del Rio were not only spending more time in detention but were spending the majority of their time in detention specifically at Torrance.
Attorneys Representing Haitian Nationals Could Not Contact Their Clients
Given Torrance’s history of depriving detained migrants of access to immigration attorneys, lawyers representing Haitian nationals were unable to make contact with their clients and provide them adequate access to counsel before their immigration hearings. Emails produced as a result of the FOIA request demonstrated attorneys’ six-week-long struggle to contact individuals detained at Torrance. An attorney from El Paso Immigration Collaborative (EPIC)—a collection of multiple local and national legal organizations including the American Immigration Council through its Immigration Justice Campaign—who became the attorney for about 45 of the 58 Haitians placed into Torrance on September 21, stated they were not able to establish contact with their clients for months. A flurry of emails between the attorney and officials at Torrance shows the obstacles attorneys face when there are limited legal services available in remote locations. Despite continued pressure to increase attorney access, on December 7, the attorney was still having trouble contacting their clients. Emails showed the lack of communication between ICE officers and detained individuals’ attorneys.
The continuous advocacy of EPIC attorneys and the Immigration Justice Campaign on behalf of these individuals eventually led to attorney access for those detained, which ultimately resulted in the release of the vast majority of Haitian nationals detained at Torrance by the end of 2021. The data shows that by January 2022, ICE released all 106 Haitian nationals apprehended between September 9 to September 24; most were released on their own recognizance or under orders of supervision, while ICE released a few others on bond set either by ICE or by an immigration judge. Unfortunately, some of these individuals made appearances at preliminary hearings before immigration judges without an opportunity to speak to a lawyer, and a small group received orders of removal.
The American Immigration Council and other partner organizations filed a complaint with DHS’ Office for Civil Rights and Civil Liberties on November 23 regarding concerns about violations of due process encountered by individuals detained at Torrance. The complaint, which includes a statement from an individual detained at Torrance, provided grim details on the issues detained individuals faced, including lack of attorneys, failure to provide interpretation, and unusually swift court proceedings that culminated in removal orders.
Abuses At Torrance Continue into 2022
Despite mounting evidence presented to ICE by immigration advocates, attorneys, members of Congress, DHS oversight agencies, and detained individuals themselves that Torrance was ill-equipped to house migrants, the trends in 2022 show that ICE ignored all red flags and continued to place migrants into the facility. The following timeline is a continuation from the previous section of warning signs ICE received about inhumane conditions at Torrance coupled with a graph of the estimated population at Torrance and ICE placements into the facility.
Timeline of Events in 2022
February 1: DHS OIG Noted Inhumane Conditions at Torrance.
DHS OIG conducts unannounced inspections of detention facilities to assess compliance with ICE standards. On February 1, 2022, the OIG began its unannounced inspection of Torrance. At this point, there were approximately 157 people at the facility.
On February 4, an agency email uncovered by Innovation Law Lab revealed that the Chief of ICE’s Audit Liaison Unit knew that the OIG found “very poor conditions” at the facility, notably “unhealthy conditions, staff shortages, and detainees in segregation who had not been let out for a week.” The email says that the OIG would issue a management alert. The next day, ICE placed another 29 people into the facility.
By February 9, ICE shrank the maximum allowed capacity at Torrance by 30 percent, thereby reducing the required staffing levels of the facility. During this period, ICE began consistently removing individuals from the facility: between February 6 and March 15, ICE removed about 134 individuals from the facility, and only placed three new individuals in the facility.
March 16: The OIG Published the Management Alert Calling on ICE to Empty Torrance.
The alert noted that the inspectors found Torrance “so critically understaffed that the facility was unable to meet its contractual requirements to ensure detainees reside in a safe, secure, and humane environment.” Due to the rapid pace of book-outs from the facility after the February 9 population reduction order, there were about 34 individuals remaining at Torrance when this alert was issued. The next day, however, ICE transferred 26 individuals into Torrance. On March 18, four members of the New Mexico congressional delegation released a joint statement condemning Torrance’s inhumane and perilous environment.
March 23 to 31: Through the Month of March, ICE Seemingly Acknowledged the Problems at Torrance.
On March 23, ICE’s Office of Acquisition Management (OAQ) modified its Intergovernmental Support Agreement (IGSA), stating Torrance continually breached federal detention center standards leading to a deduction in monthly invoices “until the Government is otherwise satisfied that these on-going 2011 PBDNS [Performance-Based National Detention Standards] and contractual requirement violations have been satisfactorily addressed.” In describing the discrepancy or problem, the report noted that the “critically short staffing plans are directly responsible for the breakdown in the overall operational capabilities of the TCDF.” On March 31, the Nakamoto Group conducted a follow-up to its July 2021 inspection. The follow-up concluded that Torrance’s staffing issues had not yet been resolved, stating “clearly, [Torrance’s] recruitment and retention efforts are unsatisfactory.” During this time period, the population of Torrance consistently decreased.
April 14: ICE Began Repopulating Torrance Again
From March 18 to April 14, as evidence mounted that the detention facility was unable to humanely house individuals, the population of Torrance consistently decreased. By April 14, approximately one month after the OIG report came out calling for Torrance to be fully depopulated, the facility held about 24 individuals. However, after that date, ICE reversed course, rapidly increasing the pace of placements in the facility. On April 15, the facility’s population more than tripled to an estimated 83 individuals—the second largest increase in population since May 14, 2021. Between April 15 and April 19, 2022, 118 individuals were placed into Torrance, 38 of whom were transferred into Torrance from another detention facility. The number of detained individuals peaked at 140 on April 19 before declining again.
May 3 to June 26: Individuals Detained, Advocates, Members of Congress and DHS Oversight Agencies Raised Red Flags.
Another inspection conducted by ICE ODO between May 3 and May 5 found six operational deficiencies at Torrance. While this was a significant decrease from the last inspection in November finding 21 operational deficiencies, the ODO inspection found that Torrance had been out of compliance with at least six operational procedures since May 2021. Further, on May 15, 2022, DHS issued a report from its Prison Rape Elimination Act (PREA) Audit of DHS’ immigration detention facilities, which found that Torrance was out of compliance with 11 standards. However, in June 2022, an average of 173 individuals were detained at Torrance—a 226 percent increase from March—with a peak of 201 individuals detained between June 24 and 26.
Kesley Vial's Suicide Death in Torrance
On August 17, 2022, a 23-year-old Brazilian asylum seeker died by suicide at Torrance. Kesley Vial was initially placed into Torrance on April 29, 2022, but by early June, an immigration judge determined that Vial did not establish credible fear of persecution entitling him to an asylum claim, and the judge ordered him removed. After Vial had spent two and a half months at Torrance, ICE transferred him to the El Paso SPC on July 28 to be removed. However, Vial’s removal did not occur as planned. Between July 19 and July 27, ICE transferred Vial four times into three different detention facilities across New Mexico, Texas, and Arizona. On July 27, he was eventually retransferred into Torrance, where an officer informed him that his deportation had been postponed for another two weeks until August 17.
When August 17 came, an Enforcement and Removal Operations (ERO) deportation officer informed Vial that they had rescheduled his flight to September 1. Throughout the day, multiple individuals recorded that Vial was “crying his eyes out.” According to an officer, “Vial was shaking, crying, and very upset, stating he would never get out of TCDF. He expressed frustration over his ongoing detention and said he wanted to go home.” At a certain point, Vial even approached the deportation officer, informing him that “he wanted the death penalty if not allowed to leave.” Later that day, Vial died after a suicide attempt in his cell. While the average time spent in Torrance was around six weeks, Vial spent almost fourteen weeks specifically at Torrance—over twice the average length in detention of all individuals at Torrance included in the data reviewed.
Torrance Temporarily Changed During Fall 2022
Between August 17, 2022, and the end of the available data on November 17, 2022, Torrance started rapidly depopulating: 93 detained individuals either ended their detention at Torrance or were transferred into different detention facilities. Additionally, ICE shifted from using Torrance as a final destination for most migrants to await the resolution of their asylum claims, to using it as a holding space during transfers between other facilities. During this time period, ICE placed only sixteen new migrants into Torrance. The nature of ICE placements of migrants into Torrance shifted from concentrated spikes to very spread out and minimal placements across several weeks, with no more than two individuals sent there on any given day. The sixteen individuals who were placed into the facility during this time period consistently spent an average of eighteen hours there, compared to about six weeks among those transferred to Torrance prior to August 17.
By the end of 2022, it looked like Torrance’s pattern of abuses might finally end. The last days of available data produced by ICE in the FOIA request showed there were only an estimated fifteen detained individuals at the facility—approximately the same number as the start of the timeline. But just as it had in the past, rather than ending immigration detention at the facility completely, ICE began rapidly repopulating the facility in December 2022, funneling hundreds of asylum seekers there over the next few months. Between 2022 and 2024, some individuals attempted suicide and experienced physical abuse by guards, and complaints of severe medical neglect increased. Despite numerous calls to close the facility over the years, on September 25, 2024, Torrance County voted to extend its IGSA with ICE until December 31, 2024.
Conclusion
The data covering the events of 2021 shows that despite substantial knowledge that Torrance was poorly equipped to house detained migrants, ICE ignored significant warning signs and continued to place vulnerable populations in the facility. Despite COVID surges, failed inspections, and reports of attorney access issues, regular spikes in ICE placements of migrants into Torrance can be seen in the data following these events. However, the beginning of 2022 showed a new trend. As pressure mounted from advocates and policymakers to increase transparency into the facility, ICE seemingly acknowledged the issues at Torrance, and temporarily reduced the population of the facility.
Nevertheless, in a demonstrated pattern, before the population of Torrance could drop to zero, ICE rapidly repopulated the facility and continued the same patterns of abuse. As the DHS OIG report came out on March 16, 2022, the data showed a continual drop in the facility’s population and a reduction in placements at Torrance. The population of Torrance fell to very low numbers signaling a potential complete depopulation of the facility. However, in mid-April, as scrutiny over the facility decreased, ICE began rapidly repopulating the facility. The suicide of Kesley Vial on August 17, 2022, marked another turning point in the facility as the population at Torrance began rapidly depopulating, and the average length spent in detention at the facility decreased from approximately six weeks to merely eighteen hours. The population again dropped to low numbers indicating another potential depopulation. But once again, as scrutiny waned towards the end of the year, ICE began repopulating the facility in December.
The data obtained by the Council’s FOIA request also shows concerning patterns regarding the equitable treatment of vulnerable populations. ICE’s demographic data on the race and ethnicity of people in detention is unreliable due to inconsistent practices. However, a deeper dig into the available numbers shows disparities in the length of detention faced by certain migrants at Torrance during this time period. This evidence raises concerns that ICE’s inconsistent labeling practices are obscuring persistent disparities among detained individuals of different races.
Recommendations
As ICE continues to detain migrants at Torrance, the Council has developed the following recommendations to improve transparency and access to information from Torrance.
- Shut Down Torrance: The data presented, covering most of 2021 and 2022, shows that ICE continually disregarded the warning signs that Torrance was not properly equipped to house detained migrants. The trends in 2022 show continued efforts to depopulate the facility after warning signs were made public, but a repopulation within days after. As abuses continued into 2024, the Council continues to call on elected officials and DHS leadership to demand a full depopulation and closure of the facility.
- Standardize Practices on Race and Ethnicity: The research demonstrated that ICE categorizations of race and ethnicity were particularly unreliable. The government’s race and ethnicity categories failed to consider more specific definitions of race and ethnicity that may elucidate whether certain vulnerable populations were subject to disparate treatment. At a minimum, ICE must adopt the new OMB guidelines on race, which include a classification for Middle Eastern and Northern African natives, as soon as possible. Even though the research could not make definitive conclusions about the race data, analysis of different variables signals that people of different backgrounds may be subject to disparate treatment at Torrance. Further, officers’ failure to record individuals’ ethnicity is problematic. The Council recommends that ICE standardize its practices on race and ethnicity to provide greater insight on how different racial and ethnic groups are treated in detention.
- Publicize Data on Race and Ethnicity: The data provided to the Council showed that ICE does in fact keep records of detained individuals’ race and ethnicity. The Council recommends that ICE publicize its data on race and ethnicity for researchers, lawmakers, organizers, and others to readily conduct research on ICE practices.