data trove natural satellites moons

Overview

Source: /home/coolhand/html/datavis/data_trove/data/quirky/moons.json

Saturn profiled 6 rows across 18 columns. The stats below are deterministic and machine-readable; the prose is a language-model interpretation of those stats (opt-in, added after the fact, never sees raw rows).

name categorical label

This column contains names of well-known natural satellites (moons) in the solar system — Earth's Moon, Jupiter's Galilean moons (Io, Europa, Ganymede, Callisto), and Saturn's Titan. With only 6 rows, 6 unique values, zero nulls, and entropy_ratio of 1.0, every entry is distinct and the column is perfectly uniform. The 'long_tail' alert is a statistical artifact of the tiny dataset size rather than a genuine distributional concern.

Treatment: Use as a human-readable label or index key; no encoding needed given the tiny cardinality, but one-hot encoding is feasible if used as a categorical feature.

parent_planet categorical label

This column records the parent planet of entries (likely moons or satellites), with only 3 distinct values across 6 rows: Jupiter, Earth, and Saturn. Jupiter dominates heavily, accounting for 4 of 6 rows (66.7%), which triggers the long-tail alert despite the tiny dataset. The dataset itself is extremely small (n=6), so statistical patterns are barely meaningful.

Treatment: One-hot encode for modelling; note dataset has only 6 rows so any model output will be unreliable.

classification categorical label

This column classifies moons into one of two astronomical categories: 'Galilean Moon' (4 of 6 rows, 66.7%) and 'Major Moon' (2 of 6 rows, 33.3%). With only 6 rows and 2 unique values, this is almost certainly a small reference table — likely Jupiter's four Galilean moons plus two other notable moons. The entropy ratio of 0.918 indicates a moderately uneven but not extreme split between the two classes.

Treatment: Use as a categorical grouping variable or binary label; encode as 0/1 if needed for modelling given the tiny dataset size.

diameter_km numeric feature

This column records diameters in kilometres, almost certainly of astronomical bodies (planets, moons, or similar large objects). With only 6 rows, all unique and ranging from 3,122 km to 5,262 km, the dataset appears to be a small reference table of mid-to-large-sized bodies — all larger than Earth's Moon (~3,474 km) but smaller than Earth (~12,742 km). The distribution is remarkably symmetric (skew ≈ −0.03) with platykurtic spread (kurtosis ≈ −1.78), meaning values are spread fairly evenly across the range with no outliers and no clustering at either extreme.

Treatment: Use as-is for analysis; consider log-transform only if joined with a wider size range dataset.

mass_earth numeric feature

This column represents planetary or object mass expressed in Earth-mass units, with values ranging from 0.008 to 0.025 — a very narrow band corresponding to sub-Earth bodies (roughly Moon-to-Mars scale). The dataset is tiny (n=6, all unique), so distributional statistics are indicative at best; the near-zero skew (−0.034) and platykurtic kurtosis (−1.24) suggest values are spread fairly uniformly across the range rather than clustering. No nulls, no outliers, and no zeros are present. Analysts should note the extremely small sample size limits any modelling utility without augmentation.

Treatment: Use as-is for small-sample analysis; consider log-transform if combined with datasets spanning a wider mass range.

semi_major_axis_au numeric feature

This column contains orbital semi-major axis measurements in astronomical units (AU) for a very small sample of 6 objects, all with unique values and no nulls. Notably, every value is extremely small (max 0.01259 AU, mean 0.0063 AU), placing all objects well inside Mercury's orbit (~0.39 AU) — suggesting these are likely inner solar system bodies, spacecraft trajectories, or sub-orbital objects. The distribution is mildly right-skewed (skew 0.61) with a wide IQR (0.00468) relative to the mean, but with only 6 rows any statistical inference is severely limited.

Treatment: Use as-is or log-transform to compress the right tail before regression, but collect more samples before drawing distributional conclusions.

eccentricity numeric feature

This column represents orbital eccentricity values, likely for planets or moons in a solar system dataset — all values fall in the range 0.0013 to 0.0549, consistent with nearly circular orbits. With only 6 rows and 6 unique values, the dataset itself is tiny. One outlier is flagged (16.7% outlier rate), corresponding to the maximum value of 0.0549, which is notably higher than the median of 0.0084 and drives a right skew of 1.11. The std of 0.0207 is larger than the mean of 0.01765, signalling high relative dispersion for such a small sample.

Treatment: Use as-is or apply log-transform to reduce right skew before modelling, and investigate the single outlier at 0.0549.

inclination_deg numeric feature

This column represents orbital or geometric inclination measured in degrees, almost certainly for a small set of planetary bodies, satellites, or similar astronomical objects (n=6, all unique). The distribution is heavily right-skewed (skew=1.76) with a median of 0.375° but a mean pulled to 1.11° by one flagged outlier at 5.145°—a value roughly 10× the interquartile range above the median. With only 6 rows, any downstream statistic is highly sensitive to that single extreme value.

Treatment: Investigate the 5.145° outlier for data-entry error before modelling; with n=6, consider robust (median-based) scaling rather than z-score normalization.

ascending_node_deg numeric feature

This column represents the longitude of the ascending node (in degrees) for a set of 6 orbital bodies or satellites, a standard Keplerian orbital element ranging from 0° to 360°. With only 6 rows and 6 unique values, there is no duplication. The wide spread (min 28.06°, max 298.848°, IQR ~147°) and negative kurtosis (−1.15) indicate values are broadly distributed across the valid angular range rather than clustered, and the mild positive skew (0.616) suggests a slight lean toward lower values.

Treatment: Use as-is or apply circular/trigonometric encoding (sin/cos) if used in ML models to respect angular periodicity.

argument_perihelion_deg numeric feature

This column represents the argument of perihelion in degrees — an orbital mechanics parameter defining the angle between an orbit's ascending node and its perihelion point, constrained to [0°, 360°). With only 6 rows (all unique, no nulls), the dataset is extremely small, covering a range of 52.643° to 318.15°, which spans most of the possible angular range. The moderate positive skew (0.69) and wide IQR of 104.11° suggest the objects are orbitally diverse with no clustering around a preferred perihelion orientation, though the tiny n makes any distributional inference unreliable.

Treatment: Use as-is or apply circular/angular encoding (e.g., sin/cos transformation) before modelling to respect the periodic [0°, 360°) nature of the angle.

mean_anomaly_deg numeric feature

This column represents mean anomaly in degrees, an orbital mechanics parameter describing where an object is in its orbit (0–360°). With only 6 rows, all unique, the values span 135.27° to 342.021°, suggesting a small catalogue of celestial bodies or orbital observations. The platykurtic distribution (kurtosis –1.42) and wide IQR of 118.35° indicate values are spread fairly uniformly across roughly three-quarters of the full angular range, with no clustering near 0° or 360°—the upper portion of the angular circle is notably absent.

Treatment: Use as-is or apply circular/angular encoding (sin/cos decomposition) if used in a model sensitive to the wrap-around at 0°/360°.

orbital_period_years numeric feature

This column represents orbital period measurements in years, likely for a small set of celestial bodies (n=6) such as inner solar system planets or short-period exoplanets. All 6 values are unique with no nulls, ranging from 0.00485 to 0.0748 years — equivalent to roughly 1.8 to 27.3 days — strongly suggesting these are Mercury-class or hot-Jupiter-class short-period orbits. The distribution is mildly right-skewed (skew=0.45) with near-platykurtic spread (kurtosis≈-1.06), meaning values are spread relatively evenly rather than clustered, which is plausible for a curated planetary sample. With only 6 rows, statistical conclusions are severely limited and any modelling should treat this as a reference lookup rather than a training feature.

Treatment: Use as-is for lookup or join; if modelling, log-transform to compress range, but note n=6 is too small for robust inference.

rotation_period_days numeric feature

This column records the rotation period in days for a small set of 6 astronomical bodies (likely planets or moons), with all values unique and no nulls. The range spans 1.769 to 27.322 days, with a mean of ~12.07 and a median of 11.55, indicating only mild right skew (0.447). The near-flat kurtosis of -1.062 and a large IQR of 12.051 relative to the mean suggest the values are broadly spread with no tight central clustering — consistent with the naturally wide variation in rotation periods across different planetary bodies.

Treatment: Use as-is for modelling given small n and mild skew; consider log-transform if combined with datasets spanning faster-rotating bodies.

perihelion_distance_au numeric feature

This column represents perihelion distance in astronomical units (AU) — the closest orbital approach distance to the Sun for a set of solar system objects (likely comets or sun-grazing bodies). With only 6 rows and all values between 0.00243 AU and 0.0125 AU, every object passes extremely close to the Sun (Mercury's perihelion is ~0.31 AU), suggesting these are sun-grazing or near-sun objects. The distribution is mildly right-skewed (skew = 0.62) with no outliers and a tight IQR of 0.004515 AU, meaning the objects cluster closely in orbital characteristics.

Treatment: Use as-is or apply log-transform given right skew before any regression or clustering, given the small sample size (n=6) limit modelling utility.

aphelion_distance_au numeric feature

This column represents the aphelion distance of orbiting bodies measured in astronomical units (AU). All 6 values are extremely small (min 0.00271 AU, max 0.01268 AU), placing every object well within Mercury's orbit (~0.47 AU) — suggesting these may be near-Earth or inner solar system micro-objects, spacecraft trajectories, or possibly a unit mismatch (e.g., values intended in a different unit). The distribution is mildly right-skewed (skew 0.603) with no outliers and no nulls across the tiny 6-row dataset, making statistical conclusions very limited.

Treatment: Verify unit consistency (values are suspiciously small for AU); if confirmed correct, use as-is in regression, otherwise rescale before modelling.

texture_url categorical metadata

This column contains external image URLs pointing to planetary/moon texture files used for 3D rendering (sourced from three.js CDN and Wikimedia Commons), representing a small reference table of 6 celestial bodies (Moon, Io, Europa, Ganymede, Callisto, Titan). Every value is unique (cardinality 6, entropy_ratio 1.0, top_rate 0.167), meaning there are no repeated textures — each row maps to exactly one celestial body. The 'long_tail' alert is technically triggered but misleading given n=6 with perfectly uniform distribution. The dataset appears to be a tiny lookup/config table rather than an analytical dataset.

Treatment: Store as-is for reference; use as a foreign-key join target or pass the URL directly to image-loading logic — do not encode or embed.

data_source categorical metadata

This column identifies the data source for each record, with every single row (all 6) carrying the identical value 'NASA JPL Horizons'. With cardinality of 1, entropy of 0.0, and top_rate of 1.0, this column carries zero information variance across the dataset. The imbalance alert is technically correct but understates the situation — this is a fully constant column, not merely skewed.

Treatment: Drop before modelling — zero-variance constant column adds no predictive signal.

fetch_date categorical metadata

This column is a fetch or extraction timestamp recording when data was pulled, stored as a categorical string. With only 6 rows and a single unique value ('2026-01-19' appearing in all 6 records), the entire dataset appears to represent a single snapshot taken on that date. The zero entropy and top_rate of 1.0 confirm complete uniformity — this column carries no discriminatory signal whatsoever.

Treatment: Drop before modelling; retain only as provenance metadata for lineage tracking.