Guest Post: Why does Angola report such a high level of landmine contamination?

Angola has perplexed the mine action community for many years. At the Thirteenth Meeting of States Parties (13MSP) in December Angola caused a kerfuffle when it reported there was 1,560km² of contaminated area remaining in the country.  Only two years ago they had reported 793km². Where did this much higher number come from? Is it correct?

Angola has always reported high levels of contamination. As early as 1993 Human Rights Watch and Physicians for Human Rights reported the head of the British Military Mission in Angola estimated that one-third of the territory was mine affected. The British Mission based the estimate on the fact that Angola had 20 million landmines in stock and if used properly would cover one third of Angola, although the Mission also said only four million of the mines were actually in the ground. Despite the rough calculation the “one-third” estimate (425,000km²) was commonly reported as the de-facto contaminated area until the completion of the Landmine Impact Survey (LIS) in 2007 when Angola declared it had 3,293 suspected hazardous areas (SHAs) covering 982km². But the LIS results, too, were controversial.  The LIS Final Report noted the contaminated area would be closer to approximately 175-200km² if the methodology used to measure SHAs by HALO Trust as a pilot test for the LIS had been consistently applied during the survey by the other operators. HALO conducted Technical Surveys rather than estimates usually employed by the LIS and used laser range finders, which at the time were relatively new to mine action. However, this lower level of contamination was never officially recognized and remains a largely unknown finding of the LIS. In the absence of other data 3,293 SHAs covering 982km² became the Baseline for Angola.

Four years later in its 2011 Article 5 Extension Request Angola reported it had 793km² of contaminated area remaining but acknowledged this was only an estimate and would fluctuate as further surveying and general data management issues were addressed. Commentary from the Analysis Group on the Extension Request did not challenge the level of contamination because the primary activity during the 5-year extension period would be a national Non-Technical Survey (NTS) to determine a new Baseline.  In its next report, the Article 7 report covering 2012, Angola omitted any references to the overall contamination area and only reported that of the 1,493 impacted communities in the LIS there were still 1,395 impacted as of 31 December 2013.

The Landmine Monitor  has been reporting on the various problems with the national database, since 2008. Among the major problems has been that the reports the National Intersectorial Commission for Demining and Humanitarian Assistance (CNIDAH), the national authority, produces do not reflect the information the NGOs send to CNIDAH. CNIDAH does not dispute the NGOs’ claim. The database problems are more related to poorly managing the flow of completion reports that the NGOs. Literally, the reports would pile up at CNIDAH in Luanda. Various errors regularly occurred at the data entry level. In addition, a subsequent review with the NGOs discovered CNIDAH had also lost completion reports. (See Landmine Monitor 2010 and 2011.)

At the Intersessional Standing Committee meetings in May 2013 Angola reported the contaminated area had increased to 1,560km² from 793km² without providing a reason for the increase. This increase was particularly odd because the NGOs were reporting large amounts of cancellation from the LIS.  The trend of the NTS is cancellation of SHAs and smaller Confirmed Hazardous Areas (CHAs). However, in May no one paid much attention to the reported increase. The attention was more focused on the ongoing NTS and progress in fixing the database problem.

Six months later in December at the 13MSP Angola reported the same 1,560km² of contamination and the figure was included in the draft progress report of the 13MSP.  When Angola gave this figure in its oral presentation it caused a minor uproar among the participants who were paying attention. Why was Angola reporting such a high number? However, the report Angola submitted to the Implementation Support Unit (ISU) at the meeting and posted online a week later includes data by province and it does not support 1,560km².  Or does it? On page four of the statement, in the format of a PowerPoint presentation, it says, “The current “Baseline” with 1,153 minefields with a size of 156.633.863.000m² (156.634km²).” Does 156.634km² mean 156,634km²? But if there is so such an unprecedented level of contamination in Angola why have there been so few casualties in recent years?

Earlier this month CNIDAH confirmed in an email that 1,560km² is the current level of contamination and that it includes results from the NTS. Still, there is enough evidence and common sense to doubt this figure.

In 2012 the Landmine Monitor reported one of the issues with the data from the Executive Commission for Demining (Comissão Executiva de Desminagem, CED) was that CED operators used both the metric and the imperial system of measurement (inches and feet) in its reporting. How did Angola calculate 1,560km²? It’s conceivable they just got confused in recording square meters and then converting them to square kilometers. Could it be a rounding error of some kind? Was the size of the mined area in completion reports incorrectly entered into the database?

There is wide agreement that the LIS in Angola overstated the level of contamination. So why is Angola reporting almost a 40% increase seven years after the LIS and 12 years since the last mine was laid?

A review of the table included in the report Angola submitted to the 13MSP supports the idea. For example, there is only one mined area reported for Luanda and it covers 818.800m². This can only mean 818,800m² or .818km² not anything else. If the same application of decimal points is used for the numbers for each province then 156km² is the correct figure, not 1,560km² and certainly not 156,000km².

The Landmine Monitor provides additional support for the more realistic 156km² when it reported that Norwegian Peoples Aid (NPA) and HALO Trust had identified 99km² of contaminated area in the 10 provinces they operate after completing non-technical and technical surveys. (Angola reported at the 13MSP that the contamination in the same 10 provinces was 120km², not far from 99km².)

The document Angola officially submitted at the 13MSP also included the estimated levels of contamination in the remaining eight provinces of Bengo, Cabinda, Cunene, Luanda, Lunda Norte, Lunda Sul, Moxico and Namibe based on preliminary and incomplete survey results. The eight provinces totaled only 37km², which pretty much limits what the final Baseline can be, and eliminates the possibility that contamination can be over 1,000km². While the overall contamination of these eight provinces will increase by the time survey is completed best estimates are that the total contamination in Angola and the new Baseline will likely be close to 200-250km². The reasons for the estimate are not complex.

Most of the remaining eight provinces are lightly contaminated compared to the 10 provinces where NPA and HALO have completed surveys, with the exception of Moxico, the most mine affected province in the country. In Cabinda province, which is the closest province to the off shore oil fields, fighting was at a much lower level than in other provinces. The same with the capital of Luanda, where there is one confirmed mined area on the outskirts of the city near an army base. Namibe in the south is mostly a desert where the LIS identified only 11 suspected hazardous areas. Presumably the re-survey will cancel several of them. Angola reported in December that Cunene had 16 minefields covering 880,000m² of contaminated area. CNIDAH had reported 160 suspected hazardous areas covering 122km² for Cunene in 2012. It is unknown how much more there is to survey in Cunene. That leaves Moxico, where it is estimated 20% of all mined area are located.  The Mines Advisory Group (MAG) has been there since 1996 and is currently re-surveying the province. How much more the new Baseline will be than 99km² largely depends on the survey results in Moxico.

Additionally, the European Commission (EC) in Angola, which is the major international donor of the ongoing Non-Technical Survey (NTS), cites between 200km² and 1,000km² in its funding guidelines for mine action in Angola released in August 2013, a range likely provided by Angola but when Angola finally announces the new Baseline the EC can be confident they will be basically right in their estimate.

The other questionable data point Angola reported is the amount of territory that is mined. As noted above the original estimate of contamination in Angola was one-third of the landmass, which apparently had some semblance of credibility in 1993 although it has none today.  At the 13MSP Angola reported that 12% of Angola’s territory is mined, though much less than 33%, it is still not credible. Globally, in all mine affected countries, with the exception of Croatia and Bosnia-Herzegovina, the total area of suspicious hazardous areas always measures less than one percent of the territory and confirmed mined areas is even less.  The only way 12% of the territory could be mined is if there were 156,000km² of contaminated area, which cannot be the level of contamination in Angola. Afghanistan, a country many consider the most mine affected in the world, has never reached 1,000km² or exceeded one percent of landmass. (Note: reporting contamination by landmass is a No Win situation. A high number is not credible and a low percentage of landmass could easily result in the dismissal of the idea that a serious landmine problem exists. Comparing mine contamination to landmass can create the perception that there is no mine problem if total contaminated landmass is the equivalent of one small dot representing all contamination on a map.)

The States Parties are expecting Angola to provide a progress report at the Review Conference in Maputo. The ICBL has issued a challenge to all States Parties to commit to finishing the job by 2024. What can Angola do in response?

Without question Angola has the money and the assets to meet the challenge. In its Article 5 Extension Request Angola reported that the government had allocated over $1 billion to clearing mines from 2003-2011. In its Extension Request Angola estimated the Executive Commission for Demining (Comissão Executiva de Desminagem, CED) needed $260 million to meet its development demining target of verifying 315km² by 2017 and CNIDAH needed approximately $75 million from the government to coordinate the clearance of 22km² also until 2017.  In the decision that approved Angola’s Extension Request at the 12MSP the States Parties requested Angola to clarify these large numbers at the 13MSP, which they failed to do. The States Parties also requested Angola to provide an update at the Review Conference in Maputo.

There are five international NGOs in Angola and the CED has over 4,000 personnel and in February 2014 reported they had 57 demining brigades. Assuming Angola will have 200-250km² remaining at the end of the survey they will need to release approximately 20-25km² per year to meet the ICBL challenge by 2024. This annual target seems within reach. The NGOs averaged 4km²from 2004-2011 CED operators average approximately 20km² over the same period.

However, the real question is whether the two institutions can cooperate and clear the estimated 200-250km² by 2024? With such a high level of national funding and CED’s demining assets it would seem that if CED plans to clear and verify 315km² (or is it 315,000km²?) over five years they could also allocate sufficient resources to include 50% of the estimated 200-250km².  Understandably, this may be controversial and politically sensitive as it may disrupt national development plans and cause tension within political alliances. Nevertheless, the issue needs to be raised with CED, CNIDAH, Ministry of Social Assistance and Reintegration (MINARS), and, if necessary, the president of Angola to see what can be done. It’s not clear if CED and CNIDAH can work this out themselves.  States Parties, the ISU and the ICBL also need to ask Angola why CED cannot do more to tackle the remaining 200-250km².

Angola can do more. They should also:

• Provide data that supports the claim there is 1,560km² of contaminated area or else revise its estimates.
• Adopt the mine free district (comunas) and province reporting from Mozambique.
• Provide more detail on clearance conducted by CED including number of mines areas cleared and how this impacted on the number of remaining mined areas. Until now CED data is in a vacuum and impossible to determine what they are doing to help Angola meet is Article 5 obligations.
• Assign more assets to clearing the estimated 200-250km² by either funding the NGOs through the Angola national budget or moving CED demining teams to help clear what CNIDAH describes as the contaminated area.
• Conduct quality assurance on reports before they are released for public consumption.

Additionally, the ISU and the ICBL should translate as many documents into Portuguese including at a minimum ICBL statements on Angola, Landmine Monitor reports on Angola and ISU documents that pertain to Angola to facilitate a wider readership within the Angolan government.

It may be remotely possible the contaminated area is more than the 200-250km² that is rationalized in this blog but it is not very probable. Angola needs to immediately improve its reporting including the presentation of data and the status of the NTS and explain what 1,560km² represents. Monitoring, evaluation and the use of data is the theme of the upcoming meeting of Mine Action Program Directors and UN Advisors. The agenda suggests some countries will be examined as examples of various uses of data in monitoring and evaluation. I recommend Angola as one country to look at.

Mike Kendellen has worked in mine action since 1999. From 2008-2013 he researched, reported and edited on more than 25 countries including Angola on Mine Ban Treaty Article 5 issues for the Landmine Monitor. He lives in Washington, D.C. Email: mike.kendellen@yahoo.com