diff --git a/rabpro/basin_stats.py b/rabpro/basin_stats.py
index 9d5c60c..9dda35c 100644
--- a/rabpro/basin_stats.py
+++ b/rabpro/basin_stats.py
@@ -382,7 +382,7 @@ def compute(
     occ_mask = ee.Image("JRC/GSW1_4/GlobalSurfaceWater").select("occurrence").lt(90)
 
     # For each raster
-    datas, tasks = [], []
+    datas, tasks, filenames = [], [], []
     for dt in control:
         if dt.band in ["None", None]:
             if dt.type == "image":
@@ -471,6 +471,8 @@ def remove_geometry(feat):
             filename = dataset_to_filename(dt.prepend, dt.data_id, dt.band)
             print(filename)
 
+        filenames.append(filename)
+
         task = ee.batch.Export.table.toDrive(
             collection=table,
             description=filename,
@@ -491,8 +493,10 @@ def remove_geometry(feat):
                 )
             )
             tasks.append(task)
+        # reset filename so it doesn't carry over to subsequent Dataset(s)
+        filename = None
 
-    return datas, tasks
+    return datas, tasks, filenames
 
 
 def _parse_reducers(stats=None, base=None):
diff --git a/tests/test_basin_stats.py b/tests/test_basin_stats.py
index 1792bf0..152ed57 100644
--- a/tests/test_basin_stats.py
+++ b/tests/test_basin_stats.py
@@ -26,7 +26,7 @@ def _clean_res(feature):
 
 
 def test_unvalidated_dataset():
-    urls, _ = rabpro.basin_stats.compute(
+    urls, _, _ = rabpro.basin_stats.compute(
         [
             Dataset(
                 "NASA/ORNL/DAYMET_V4",
@@ -49,7 +49,7 @@ def test_customreducer():
     def asdf(feat):
         return feat.getNumber("max")
 
-    urls, _ = rabpro.basin_stats.compute(
+    urls, _, _ = rabpro.basin_stats.compute(
         [Dataset("JRC/GSW1_4/YearlyHistory", "waterClass", stats=["max"])],
         gee_feature_path="users/jstacompute/basins",
         reducer_funcs=[asdf],
@@ -62,7 +62,7 @@ def asdf(feat):
 
 def test_categorical_imgcol():
 
-    urls, _ = rabpro.basin_stats.compute(
+    urls, _, _ = rabpro.basin_stats.compute(
         [Dataset("MODIS/061/MCD12Q1", "LC_Type1", stats=["freqhist"])],
         gee_feature_path="users/jstacompute/basins",
     )
@@ -73,7 +73,7 @@ def test_categorical_imgcol():
 
 def test_timeindexed_imgcol():
 
-    urls, _ = rabpro.basin_stats.compute(
+    urls, _, _ = rabpro.basin_stats.compute(
         [
             Dataset(
                 "JRC/GSW1_4/YearlyHistory",
@@ -91,7 +91,7 @@ def test_timeindexed_imgcol():
 
 def test_timeindexedspecific_imgcol():
 
-    urls, _ = rabpro.basin_stats.compute(
+    urls, _, _ = rabpro.basin_stats.compute(
         [
             Dataset(
                 "JRC/GSW1_4/YearlyHistory",
@@ -110,7 +110,7 @@ def test_timeindexedspecific_imgcol():
 
 def test_nontimeindexed_imgcol():
 
-    data, _ = rabpro.basin_stats.compute(
+    data, _, _ = rabpro.basin_stats.compute(
         [
             Dataset(
                 "JRC/GSW1_4/MonthlyRecurrence",
@@ -129,7 +129,7 @@ def test_nontimeindexed_imgcol():
 def test_img():
 
     stat_list = ["min", "max", "range", "std", "sum", "pct50", "pct3"]
-    data, _ = rabpro.basin_stats.compute(
+    data, _, _ = rabpro.basin_stats.compute(
         [
             Dataset(
                 "JRC/GSW1_4/GlobalSurfaceWater",
@@ -146,3 +146,22 @@ def test_img():
     assert float(res["mean"]) > 0
     # mean, count, id, and id_outlet are automatically returned
     assert res.shape[1] == len(stat_list) + 4
+
+
+def test_filenames():
+    stat_list = ["mean", "std"]
+    _, _, filenames = rabpro.basin_stats.compute(
+        [
+            Dataset("MERIT/DEM/v1_0_3", "dem", stats=stat_list, gee_type="image"),
+            Dataset(
+                "JRC/GSW1_4/GlobalSurfaceWater",
+                "occurrence",
+                stats=stat_list,
+                gee_type="image",
+            ),
+        ],
+        gee_feature_path="users/jstacompute/basins",
+    )
+
+    # make sure we're getting unique filenames
+    assert filenames.count(filenames[0]) == 1