{
  "id": "t02_desert_null",
  "order": 2,
  "tag": "null",
  "eyebrow": "02 \u00b7 Null control",
  "title": "Empty desert produces a similar score distribution",
  "question": "Is the output any different when there is no pyramid at all?",
  "finding": "The mean best score is 0.306 on the pyramid footprint and 0.306 in the desert-only scene. Their empirical distribution distance is 0.0096. The absence of buried objects does not prevent structured output.",
  "limitations": "One realization per scene; the 32-pixel patches overlap on a 4-pixel grid, so the samples are dependent. The saved conventional KS p-value is descriptive only here. Similar distributions do not establish statistical equivalence or validate a detector.",
  "method": "Identical pipeline and grid as T1 on a scene containing only clutter and sparse rocks; scores compared pixel by pixel with T1's pyramid footprint.",
  "figures": [
    {
      "file": "results/t02_desert_null/figs/t02_transect_az.png",
      "caption": "<b>Desert only.</b> Flat ground with speckle and a few rocks, same transect, same pipeline. The 'tomogram' is as busy as the pyramid's."
    },
    {
      "file": "results/t02_desert_null/figs/t02_histograms.png",
      "caption": "<b>Overlapping distributions.</b> Best window-fit scores and their winning model depths for the pyramid and desert controls. Compare their effect sizes; the many overlapping patches are not independent experimental replicates."
    },
    {
      "file": "results/t02_desert_null/figs/t02_maps.png",
      "caption": "<b>Same scale, two scenes.</b> Best-fit score maps for a pyramid scene and a desert-only scene. Both inputs contain no buried objects. This comparison tests this score's specificity, rather than proving the two scenes identical."
    }
  ],
  "metrics": {
    "mean_best_adjR2_pyramid": 0.30645185708999634,
    "mean_best_adjR2_desert_scene": 0.30577531456947327,
    "frac_gt_0p5_pyramid": 0.013728078446162549,
    "frac_gt_0p5_desert_scene": 0.013092592592592593,
    "ks_statistic": 0.009612881597418688,
    "ks_pvalue": 0.074280704560997,
    "runtime_s": 262.3717818260193,
    "trajectory_rms_px_desert": 0.006491091789105842,
    "trajectory_rms_px_pyramid": 0.007339921294636644
  },
  "date": "2026-09-19",
  "provenance": {
    "python": "3.11.3",
    "numpy": "2.4.6",
    "source_sha256": "86578ebf9962547f59c204c7008397d14646103766f7abd70cc79a3339d177f4"
  }
}