diff --git a/src/main/java/scratch/kevin/nshm23/figures/WrapperCompareToExternal.java b/src/main/java/scratch/kevin/nshm23/figures/WrapperCompareToExternal.java
index 6759fa5e..595dd198 100644
--- a/src/main/java/scratch/kevin/nshm23/figures/WrapperCompareToExternal.java
+++ b/src/main/java/scratch/kevin/nshm23/figures/WrapperCompareToExternal.java
@@ -94,6 +94,8 @@ public static void main(String[] args) throws IOException {
 					}
 				}
 				System.out.println("Matched "+rp+" with CSV column "+extCSV.get(0, closestCol));
+				Preconditions.checkState(minDiff < 5d, "RP mimatch! |ours (%s = %s) - theirs (%s)| = %s",
+						rp, rp.returnPeriod, extCSV.get(0, closestCol), minDiff);
 				
 				int numSkipped = 0;
 				int numMatched = 0;
diff --git a/src/main/java/scratch/kevin/nshm23/wrapper/RewriteExternalHazard.java b/src/main/java/scratch/kevin/nshm23/wrapper/RewriteExternalHazard.java
new file mode 100644
index 00000000..549a7b2e
--- /dev/null
+++ b/src/main/java/scratch/kevin/nshm23/wrapper/RewriteExternalHazard.java
@@ -0,0 +1,116 @@
+package scratch.kevin.nshm23.wrapper;
+
+import java.io.BufferedOutputStream;
+import java.io.File;
+import java.io.FileOutputStream;
+import java.io.IOException;
+import java.io.OutputStreamWriter;
+import java.util.zip.ZipEntry;
+import java.util.zip.ZipOutputStream;
+
+import org.opensha.commons.data.CSVFile;
+import org.opensha.commons.data.xyz.GriddedGeoDataSet;
+import org.opensha.commons.geo.GriddedRegion;
+import org.opensha.commons.geo.Location;
+import org.opensha.commons.geo.json.Feature;
+import org.opensha.sha.earthquake.faultSysSolution.util.SolHazardMapCalc.ReturnPeriods;
+import org.opensha.sha.earthquake.rupForecastImpl.nshm23.util.NSHM23_RegionLoader;
+
+import com.google.common.base.Preconditions;
+
+public class RewriteExternalHazard {
+
+	public static void main(String[] args) throws IOException {
+		// utility to convert nshmp-haz runs to our hazard results zip file format
+		
+		File invsDir = new File("/data/kevin/nshm23/batch_inversions");
+		
+		File extCalcDir = new File("/home/kevin/OpenSHA/nshm23/nshmp-haz-models/ext_hazard_calcs/"
+//				+ "conus-6b4-nshmp-haz-grid_smooth_OFF-0p1-vs760-20240213-bbc02463b2e1ff/vs30-760");
+//		File outputDir = new File(invsDir, "2024_02_13-nshmp-haz-external-conus-6b4-ask2014-vs760-grid_smooth_OFF");
+				+ "conus-6b4-nshmp-haz-grid_smooth_optimize_OFF-0p1-vs760-20240213-195bddb0d73730/vs30-760");
+		File outputDir = new File(invsDir, "2024_02_15-nshmp-haz-external-conus-6b4-ask2014-vs760-grid_smooth_optimize_OFF");
+//				+ "conus-6b4-nshmp-haz-0p1-vs760-20240208-afaff93cd918f5/vs30-760");
+//		File outputDir = new File(invsDir, "2024_02_08-nshmp-haz-external-conus-6b4-ask2014-vs760");
+		
+		GriddedRegion gridReg = new GriddedRegion(
+				NSHM23_RegionLoader.loadFullConterminousUS(),
+				0.1, GriddedRegion.ANCHOR_0_0);
+		
+		double[] periods = {0d, 1d};
+		ReturnPeriods[] rps = {ReturnPeriods.TWO_IN_50, ReturnPeriods.TEN_IN_50};
+		
+		Preconditions.checkState(outputDir.exists() || outputDir.mkdir());
+		ZipOutputStream zout = new ZipOutputStream(new BufferedOutputStream(new FileOutputStream(
+				new File(outputDir, "results_hazard.zip"))));
+		
+		for (double period : periods) {
+			String extDirName;
+			String mapPrefix;
+			if (period == 0d) {
+				extDirName = "PGA";
+				mapPrefix = "map_pga";
+			} else {
+				extDirName = "SA"+((float)period+"").replace('.', 'P');
+				mapPrefix = "map_"+(float)period+"s";
+			}
+			File extSubDir = new File(extCalcDir, extDirName);
+			Preconditions.checkState(extSubDir.exists(), "Ext period dir doesn't exist: %s", extSubDir.getAbsolutePath());
+			
+			CSVFile<String> extCSV = CSVFile.readFile(new File(extSubDir, "map.csv"), true);
+			
+			for (ReturnPeriods rp : rps) {
+				double years = rp.returnPeriod;
+				double minDiff = Double.POSITIVE_INFINITY;
+				int closestCol = -1;
+				for (int col=2; col<extCSV.getNumCols(); col++) {
+					double csvYears = extCSV.getDouble(0, col);
+					double diff = Math.abs(csvYears - years);
+					if (diff < minDiff) {
+						closestCol = col;
+						minDiff = diff;
+					}
+				}
+				System.out.println("Matched "+rp+" with CSV column "+extCSV.get(0, closestCol));
+				Preconditions.checkState(minDiff < 5d, "RP mimatch! |ours (%s = %s) - theirs (%s)| = %s",
+						rp, rp.returnPeriod, extCSV.get(0, closestCol), minDiff);
+				
+				int numSkipped = 0;
+				int numMatched = 0;
+				GriddedGeoDataSet extXYZ = new GriddedGeoDataSet(gridReg);
+				for (int i=0; i<extXYZ.size(); i++)
+					extXYZ.set(i, Double.NaN);
+				for (int row=1; row<extCSV.getNumRows(); row++) {
+					double lon = extCSV.getDouble(row, 0);
+					double lat = extCSV.getDouble(row, 1);
+					double val = extCSV.getDouble(row, closestCol);
+					Location loc = new Location(lat, lon);
+					int locIndex = gridReg.indexForLocation(loc);
+					if (locIndex < 0) {
+						numSkipped++;
+						continue;
+					} else {
+						Preconditions.checkState(Double.isNaN(extXYZ.get(locIndex)));
+						extXYZ.set(locIndex, val);
+						numMatched++;
+					}
+				}
+				System.out.println("Filled in "+numMatched+"/"+extXYZ.size()+" points (skipped "+numSkipped+")");
+				
+				String destName = mapPrefix+"_"+rp.name()+".txt";
+				ZipEntry entry = new ZipEntry(destName);
+				zout.putNextEntry(entry);
+				GriddedGeoDataSet.writeXYZStream(extXYZ, zout);
+				zout.flush();
+			}
+		}
+		// write the region
+		ZipEntry entry = new ZipEntry("gridded_region.geojson");
+		zout.putNextEntry(entry);
+		OutputStreamWriter writer = new OutputStreamWriter(zout);
+		Feature.write(gridReg.toFeature(), writer);
+		writer.flush();
+		zout.close();
+	}
+
+}