feat: add SRVI and SRWI indices to Sentinel-2 collection

sentinel-2/srvi/README.md

@@ -0,0 +1,29 @@
+---
+title: Symbolic Regression Vegetation Index
+parent: Sentinel-2
+grand_parent: Sentinel
+layout: script
+permalink: /sentinel-2/srvi/
+nav_exclude: true
+examples:
+  - zoom: 11
+    lat: 34.7
+    lng: 33.0
+    datasetId: S2L2A
+    fromTime: '2026-03-01'
+    toTime: '2026-03-15'
+    platform: CDSE
+    evalscripturl: https://raw.githubusercontent.com/c-chrysostomou/custom-scripts/master/sentinel-2/srvi/script.js
+---
+
+# Symbolic Regression Vegetation Index (SRVI)
+
+## General description of the script
+The Symbolic Regression Vegetation Index (SRVI) is a 4-band index optimized for global vegetation mapping. It was discovered using a data-driven symbolic regression framework designed to improve vegetation separability across diverse biomes and preserve sensitivity in high-biomass, saturated regions.
+
+The index formula utilizes Green (B03), Red (B04), NIR (B08), and SWIR1 (B11) bands:
+
+$$SRVI = \frac{2.0 \cdot N - 3.0 \cdot R}{N + R + 0.5 \cdot (G + S1)}$$
+
+## References
+[1] Chrysostomou, C., Neophytides, S.P., Mavrovouniotis, M. et al. Optimized spectral indices for global vegetation and water mapping using Sentinel-2. Sci Rep 16, 4491 (2026). https://doi.org/10.1038/s41598-025-34720-x

sentinel-2/srvi/script.js

@@ -0,0 +1,28 @@
+// //info
+// VERSION: 3
+// Name: Symbolic Regression Vegetation Index (SRVI)
+// Description: Optimized spectral index for global vegetation mapping discovered via symbolic regression.
+// Reference: Chrysostomou, C., Neophytides, S.P., Mavrovouniotis, M. et al. Optimized spectral indices for global vegetation and water mapping using Sentinel-2. Sci Rep 16, 4491 (2026).
+// DOI: https://doi.org/10.1038/s41598-025-34720-x
+
+function setup() {
+  return {
+    input: ["B03", "B04", "B08", "B11", "dataMask"],
+    output: { bands: 4 }
+  };
+}
+
+function evaluatePixel(samples) {
+  let G = samples.B03;  // Green
+  let R = samples.B04;  // Red
+  let N = samples.B08;  // NIR
+  let S1 = samples.B11; // SWIR1
+
+  let denominator = N + R + 0.5 * (G + S1);
+  let srvi = (denominator !== 0) ? (2.0 * N - 3.0 * R) / denominator : 0;
+
+  let visual = (srvi + 1.5) / 3.5; 
+  visual = Math.max(0, Math.min(1, visual));
+
+  return [0, visual, 0, samples.dataMask];
+}

sentinel-2/srwi/README.md

@@ -0,0 +1,29 @@
+---
+title: Symbolic Regression Water Index
+parent: Sentinel-2
+grand_parent: Sentinel
+layout: script
+permalink: /sentinel-2/srwi/
+nav_exclude: true
+examples:
+  - zoom: 11
+    lat: 34.7
+    lng: 33.0
+    datasetId: S2L2A
+    fromTime: '2026-03-01'
+    toTime: '2026-03-15'
+    platform: CDSE
+    evalscripturl: https://raw.githubusercontent.com/c-chrysostomou/custom-scripts/master/sentinel-2/srwi/script.js
+---
+
+# Symbolic Regression Water Index (SRWI)
+
+## General description of the script
+The Symbolic Regression Water Index (SRWI) is a 4-band spectral index optimized for accurate and consistent surface water mapping. Discovered via symbolic regression, it effectively addresses common classification challenges, reducing false positives caused by mountain shadows and complex urban surfaces.
+
+The index formula utilizes Blue (B02), Green (B03), NIR (B08), and SWIR1 (B11) bands:
+
+$$SRWI = \frac{(G + B) - (N + S1)}{(G + B) + (N + S1)}$$
+
+## References
+[1] Chrysostomou, C., Neophytides, S.P., Mavrovouniotis, M. et al. Optimized spectral indices for global vegetation and water mapping using Sentinel-2. Sci Rep 16, 4491 (2026). https://doi.org/10.1038/s41598-025-34720-x

sentinel-2/srwi/script.js

@@ -0,0 +1,29 @@
+// //info
+// VERSION: 3
+// Name: Symbolic Regression Water Index (SRWI)
+// Description: Optimized spectral index for consistent surface water delineation discovered via symbolic regression.
+// Reference: Chrysostomou, C., Neophytides, S.P., Mavrovouniotis, M. et al. Optimized spectral indices for global vegetation and water mapping using Sentinel-2. Sci Rep 16, 4491 (2026).
+// DOI: https://doi.org/10.1038/s41598-025-34720-x
+
+function setup() {
+  return {
+    input: ["B02", "B03", "B08", "B11", "dataMask"],
+    output: { bands: 4 }
+  };
+}
+
+function evaluatePixel(samples) {
+  let B = samples.B02;  // Blue
+  let G = samples.B03;  // Green
+  let N = samples.B08;  // NIR
+  let S1 = samples.B11; // SWIR1
+
+  let numerator = (G + B) - (N + S1);
+  let denominator = (G + B) + (N + S1);
+  let srwi = (denominator !== 0) ? numerator / denominator : 0;
+
+  let visual = (srwi + 1.0) / 2.0;
+  visual = Math.max(0, Math.min(1, visual));
+
+  return [0, 0, visual, samples.dataMask];
+}