Two conflicting versions of what occurred in Sudan during those years are in direct contention. On one side, witnesses and prosecutors testify that thousands were displaced when government forces and militias cleared areas for oil extraction. On the other, the company contends that it wasn\u2019t oil driving the conflict, that few people lived in those areas, and that the fires were due to traditional slash-and-burn practices, not conflict. <\/p>\n\n\n\n
Earth observation satellites<\/h2>\n\n\n\n
Expert witness Prins, called by the prosecutor explained that the Landsat satellites \u2013 jointly operated by NASA and the U.S. Geological Survey \u2013 have collected data on the planet\u2019s land surface since 1972. This makes Landsat the longest continuous satellite-based archive of Earth observation.<\/p>\n\n\n\n
The images from Sudan between 1997 and 2003 are sporadic \u2013 but in theory, the satellites passed over Block 5A every eight days. \u201cThe white areas in the satellite images,\u201d Prins explained, \u201care the result of cattle grazing and trampling down the grass. From pasture management we know that cattle expose bare, sandy soil, and satellites pick this up. It\u2019s a stronger marker than cropland, because crops may have been harvested.\u201d<\/p>\n\n\n\n
Prins recalled being approached by the European Coalition on Oil in Sudan (ECOS) in 2006, who asked if he could shed light on \u201crumours\u201d surrounding Block 5A<\/a>. \u201cI was known as someone who had expertise in environmental issues and could assess how the oil industry affected areas. ECOS\u2019s request was simple: to examine land-use patterns and how they changed over time.\u201d<\/p>\n\n\n\n \u201cOne image is worth more than a thousand words. I want to show the images to show the truth.\u201d In his presentation to the court, which Prins also published on his website<\/a>, it was clear: between 1999 and 2003, large changes were visible in South Sudan \u2013 mass displacement of people and villages destroyed.<\/p>\n\n\n\n His analysis showed that people settlements disappeared from the \u201coil road\u201d and areas near oil extraction and he argued that the images offered \u201cobjective evidence,\u201d which also matched the testimonies of refugees.<\/p>\n\n\n\n He also argued that the Nuer people\u2019s land use was particularly visible in satellite imagery, since they both grazed cattle and cultivated fields around their homesteads. \u201cThe roofs of huts can easily be seen in satellite images, as can the areas of cultivation,\u201d he said, describing how \u201caggregated data from spectral channels\u201d could reveal farmland.<\/p>\n\n\n\n He compared images taken in November over several years, since by then the rains had ended and fields had been harvested \u2013 and slash-and-burn fires were less frequent.<\/p>\n\n\n\n Satellite imagery has become an important source of information and evidence in international law. But according to a recent UN and Asser Institute study<\/a> their use is hindered by a lack of technical competence among legal experts on one hand, and satellite specialists\u2019 unfamiliarity with the legal field on the other.<\/p>\n\n\n\n From images in 2000, the court could see the \u201coil road\u201d between Thar Yath<\/a> and Leer<\/a> for the first time, along with a grid of seismic lines etched into the landscape. \u201cWe also see garrisons along the oil road in images from 2004,\u201d Prins told the court. From 1996 to 2002, his comparisons showed fire damage outside villages at first. By 2002, many had moved toward Bentiu \u2013 the cattle leaving visible tracks around the town.<\/p>\n\n\n\n In a summary slide, Prins compared the evolution between 1999 to 2002. Green markers showed that activity dropped by 25 percent. \u201cIn the years in between, you can see burning inside villages, which could be a sign of conflict. If bushfires enter settlements, it may indicate fighting.\u201d<\/p>\n\n\n\n Another slide highlighted in red the fields cultivated in 1999-2000, which later disappeared.<\/p>\n\n\n\n He also compared satellite data to Lundin\u2019s own security reports on incidents, showing patterns in three areas: the northern part of the all-weather road, Nhialdiu, with its major attack in February 2002, and other villages where displaced people later concentrated in Wicok and Chotchar.<\/p>\n\n\n\n One of his slides carried the heading: \u201cSatellites detect \u2013 ground reports confirm.\u201d Prins pointed out how the testimony of another trial witness, Diane de Guzman, who had drawn maps while walking through the region, matched what satellites revealed from above.<\/p>\n\n\n\n Finally, he presented a map comparing images from 1999 and 2002: in red, villages that were \u201cgone,\u201d and in green, new ones. The data show that fires in Nhialdiu peaked in 2001\u20132002.<\/p>\n\n\n\n Today, he added, anyone can search the entire Landsat archive via Climateengine.org<\/a>, covering more than 40 years. \u201cI was criticized for using too few images, but the entire archive is now searchable. The conclusions I presented back in 2013 [to ECOS]\u00a0have only been strengthened,\u201d he added.<\/p>\n\n\n\n He concluded that his findings rested on three layers of robustness \u2013 \u201cbelt and suspenders\u201d: the scientific method itself, corroborating witness testimony, and newly available data.<\/p>\n\n\n\n According to the defence, these satellite images were central to the decision to open a preliminary investigation: \u201cIt was a way to give the investigation more material and \u2018something tangible\u201d said the lawyers in court, during last spring<\/a>.<\/p>\n\n\n\n The defence\u2019s experts from Hatfield and Company reviewed the images and produced two reports, the first titled Remote Sensing Review<\/em>. The experts argued that the prosecution\u2019s use of the images is \u201cflawed\u201d and cannot be given any evidentiary value.<\/p>\n\n\n\n According to the experts, no conclusions can be drawn about human activity from the images: \u201cIt is difficult to identify roads or villages because of the lack of human footprints.\u201d \u201cThe number of error sources is too many, depending on season, cloud cover, the lack of fixed reference points, as well as the waterlogged ground and drainage along the road. The conclusion is that there is no scientific basis for the conclusions drawn in the Prins report.\u201d<\/p>\n\n\n\n The defence also argues that one source of error is that what prosecutors see as burned villages are in fact grass fires caused by slash-and-burn practices. \u201cFrom the drilling rig at Thar Yath, staff often saw villagers burning the grass at the end of the dry season.\u201d<\/p>\n\n\n\n The defence also pointed to the book\u00a0The Nuer<\/em>\u00a0from 1940 \u2013 which they had earlier used to argue that ethnic conflicts in the region were age-old \u2013 noting that it too described how grass was set on fire as soon as it dried, before the rains came and new shoots could grow. They also cited a global fire database showing that over half of South Sudan\u2019s land burns each year \u2013 a pattern consistent with traditional farming, not forced displacement.<\/p>\n\n\n\n Their conclusion was that \u201cburnt grass is more a sign of human presence than absence.\u201d<\/p>\n\n\n\t Prins was not the only \u2018satellite witness\u2019. During the investigation, prosecutors had engaged their own expert: engineer Torbj\u00f6rn Rost.<\/p>\n\n\n\n Rost tells the court, on June 25, he specialized in \u201cobject-based methods,\u201d used in studies of refugee camps in Somalia and elsewhere. For over a decade he worked almost exclusively for UNHCR, the UN Refugee Agency, using satellite images to estimate populations in inaccessible areas, including disaster zones.<\/p>\n\n\n\n When asked by police in 2013 to examine Block 5A, he began reviewing available images from 1997 to 2003. Like Prins, Rost focused on the available data. But he made it clear: his task was not to compare the same place across years, but simply to document what could be seen at each point in time. The key difference, he explained, was that Prins conducted \u201cchange analyses.\u201d On police instructions, they had not collaborated.<\/p>\n\n\n\n Prosecutor Annika Wennerstr\u00f6m showed the court a 1997 digital image with Block 5A outlined. \u201cThis is the maximum detail you can get. Only in Hollywood do you get closer. I assess this as human activity because the vegetation is stripped. But that is an assessment, an interpretation. To verify, I compared with high-resolution images,\u201d she said.<\/p>\n\n\n\n Switching to a 1995 image, she asked why he placed a \u201cdot\u201d on it. Rost repeated that the reddish soil had been altered by human activity. \u201cThe police wanted to see huts, and I told them: \u2018That\u2019s not possible.\u2019 I can\u2019t say it was a settlement, but I can say there was human activity.\u201d Throughout, he spoke of \u201chuman activity\u201d \u2013 ground worn by people or livestock.<\/p>\n\n\n\n His answers were technical and cautious. Others would have to draw conclusions.<\/p>\n\n\n\n He described the dark areas inside villages as \u201cburned grass,\u201d sometimes showing fire at the edges. Unlike Prins, he placed no broader significance on this.<\/p>\n\n\n\n On the court\u2019s four large screens, satellite images scrolled by: riverbeds, clouds, the White Nile. One image showed the road construction south from Bentiu. By November 29, 2000, the all-weather road had passed Duar en route to the rig at Thar Yath.<\/p>\n\n\n\n Asked whether military vehicles could traverse the area before the road, he said: \u201cAll I can say is that before the road \u2013 there was no road.\u201d<\/p>\n\n\n\n From March 8, 2002, images showed \u201csome burning\u201d along the new road toward Nhialdiu \u2013 slash-and-burn farming. \u201cLarge areas are burning, and you can see human activity,\u201d Rost said. On April 9, 2002, the road reached Nhialdiu. By June 12, 2002, it cut through the town.<\/p>\n\n\n\n Nhialdiu is central to the indictment: prosecutors argue the government required Lundin to fund an all-weather road there to secure seismic surveys near the town, which was also a base for militia commander Peter Gadet<\/a>.<\/p>\n\n\n\n After Prins\u2019s testimony, the defence had no questions. After Rost\u2019s, they had one.<\/p>\n\n\n\n They produced an earlier police interview in which Rost said that between 1995 and 2003 there were \u201cno significant differences in human activity.\u201d \u201cThat\u2019s correct,\u201d he confirmed, noting he had conducted spot checks to be sure. His conclusion remained the same: between 1995 and 2003, there were no observable changes in villages and settlements.<\/p>\n\n\n\n \u201cWe have no further questions,\u201d the defence concluded.<\/p>\n\n\nMass displacement of people<\/h2>\n\n\n\n
\u201cOil road\u201d appears in 2000<\/h2>\n\n\n\n
![\"Map](\"https:\/\/www.justiceinfo.net\/wp-content\/uploads\/Sudan_Lundin-trial-Landsat-satellite-map-evidence_@Erik-Prins.jpg\")
\u201cSatellites detect \u2013 ground reports confirm\u201d<\/strong><\/h2>\n\n\n\n
\u201cThe number of error sources is too many\u201d<\/h2>\n\n\n\n
\u201cBefore the road \u2013 there was no road\u201d<\/strong><\/h2>\n\n\n\n
Burning, and human activity<\/h2>\n\n\n\n
![\"Alexandre](\"https:\/\/www.justiceinfo.net\/wp-content\/uploads\/South-Sudan_Alexandre-Schneiter-Lundin-Oil-interview_@Bloomberg-540x360.jpg\")
<\/div><\/a>\r\n\t\t\t