The authors explained the values of reliability and accuracy in GIS modelling. They highlighted the visual impact of terrain. A second chapter focused on battlefields and detailed the essential processes from fieldwork to desktop mapping.18 Another important book confirmed the peculiarities of working with both historical data and maps.19A visit to Białowieźa was necessary to log important data into the database: the positions of old photographs, specific map references, and geo-reference points. The processing of the Luftwaffe map sections was a laborious and time-consuming task that involved adjusting different maps to a single useable map. Bettina began an advanced GIS course, and the department allowed the process of digitisation and the GIS mapping to be tested under university conditions. A high degree of expertise emerged that the university endorsed with a letter of commendation.20 Bettina began to incorporate more advanced methods of historical GIS.21To better understand the full breadth of Historical GIS and the basic operating principles, there were online seminars available to beginners. In May 2013, I joined an historical GIS course hosted by the Institute of Historical Research, London University. This involved working through an ArcGIS project in the classroom. This course reinforced the importance of managing several issues, including, digi-maps, geo-referencing, vector data and coordinates, symbolising data, pixelization, spatiality, data parcels, cartograms, and copyright.22 By the completion of the first stage and the preparatory work, we had produced a map (Map 3 below) in a digital, georeferenced form that would make possible a variety of analyses.
The second stage involved identifying data from the documents or qualitative content to form into specific layers. In a sense, this was akin to unpicking the spaghetti of data and trying to isolate common data sets. The significance of GIS is the integration of seemingly unrelated data and its reordering into meaningful information. Layers were identified from different sources. Infrastructure like roads, swamps, railways, roads, bridges, farms, and estates were digitized from the original Polish maps. The Luftwaffe had drawn information on their maps, such as the position of strongpoints, companies, and Jagdkommando. This data was integrated with the Polish maps and digitized. The next task involved data mining from the surviving diaries of the Luftwaffe. There were two defined periods with different commanders, tactics, and dogma. This was a very time-consuming process because of the form of handwriting. Sütterlin is not taught in German schools today, but was widely utilised during the war. Once the barrier of the handwriting was overcome, page by page, line by line, (about 120 pages) we were able to present the results for review by a German veteran, who explained more nuances about that writing form under combat conditions. The overall outcome was a wealth of details and data. This led to multiple complex layering and we began to compare colour pixelization against the black-white map format. We opted for the latter. There were so many map options we decided to compile a series of test maps. Copies of these maps were sent to the late Dr. Joe White and his team, at the US Holocaust Memorial and Museum in Washington DC in 2013 for an evaluation. We also began to examine the nature of time and its impact on the events. A partial experiment, involving multiple modelling was used to test the visualisation of progressive troop movement and patrols through segments of the forest. The outcome came down to tracking platoon, squad, and Jagdkommando movements, by minutes, hours, or days, depending upon the detail of supporting data.
Stage three involved formulating forensic analyses based upon the findings from the GIS layering tests. The key forensic mappings was classified under: the orders of battle or deployment of companies; the Bandenbekämpfung actions; population engineering; Judenjagd or Jew hunts; and larger operations. A 3-D model was drawn from the isolines of the maps. The outcomes of the analyses confirmed the working value of Historical GIS in a forensic dimension. How we presented and organized this evidence became crucial because the format selected would profile the narrative. The choices were: a military history format, the judgemental form of a war crime investigation, or a socio-cultural study of violence. In discussions about Bandenbekämpfung with retired US Army General Richard Trefry, comparisons emerged from Vietnam and Iraq.23 He recommended the US Army’s official report on My Lai as a structural model. The report incorporated a full schedule of maps and movement diagrams, which enabled comparisons with the logic behind the GIS mapping.24 Comparing post-war atrocities, such as My Lai, with Bandenbekämpfung was not the intention, but following the structure of the report of integrating maps in the narrative did seem appropriate. In 2013, a test of the mapping and narrative was made of the area where Siegfried Adams was killed in combat in June 1943 (see epilogue). The geo-data, geo-references, and qualitative content proved complete for Adams. The results were spectacularly successful.
A final test was to compare the findings to the content from Geographies of the Holocaust. This book had set the benchmark for applying historical GIS to the Holocaust. The book revealed the potential for a multidisciplinary approach to the Holocaust, but also the limitations when applied to military matters, and both are due to issues of spatiality. In a chapter about hunting Jews, the application of GIS was focused on time and space. The base data required a large body of data statistics including names, homes, mass deportations, and camps. In another chapter devoted to the killing grounds, the authors attempted to reconstruct the specialities of hunting Jews in Belarus. The chapter highlighted the formulation of ‘locational models of killing’, with GIS applied to map the movement patterns of killing units. There was graphical presentation of killings, a diagrammatic schedule of killings by time, and functional images of soldiery duties, which culminated in the summary—testimony, technology, and terrain. In both chapters there was an absence of integration and limited forensic outcomes. If the same methods had been applied to Białowieźa, they would have produced only minimal results.25 This did not reflect against the authors, but rather explained how different documentary evidence requires different historical GIS methods. We concluded that Historical GIS had unlocked the Luftwaffe’s mission and methods in Białowieźa. Historical GIS served three purposes: firstly, it had highlighted the prominent geographical features of the forest. Secondly, it recreated all military movements by timelines, operations, and outcomes, such as individual crimes. A deeper forensic analysis was achieved from mapping and visualising the effects of discipline, the routine, and orderliness of the killings. Thirdly, GIS exposed how a national political frontier divided responsibilities, but also explained the internal rivalries. In conclusion, GIS had exposed a grandiose Nazi scheme, Göring’s ambitions, and the soldiers’ behaviour—probably for the first time since 1945.
III. The GIS maps
This book contains more than twenty maps, the majority are from the Historical GIS analyses. The map below (uncaptioned) is an example of our first results. This is a composite map and although highly detailed it’s also a picture of information overload caused by the density of layering and grayscale. The ‘buffers’, or roundels, represent patrol distances within company areas and also radio signals ranges for small wireless devices. Arrow lines give general directions of patrols and deportations. The Germans incorporated the Jagen system, which represented a square kilometre ground—reflected in the grid pattern. This was an old Tsarist form of measurment used in Białowieźa for forestry management. The squares were mapped into the German maps and renamed Jagen. All of these factors have remained constant. However, in an effort to reduce the sense of clutter, we experimented with single layer maps and then with specific theme maps.
The solution we finally decided to accept were specific to the general findings from the research and forensic in design. The maps were finalised after series of experiments with colour, black and white,
