diff --git a/examples/cosmodc2/Cluster_pipelines/1deg2-nersc.sub b/examples/cosmodc2/Cluster_pipelines/1deg2-nersc.sub
index 729e5b172..87c159f04 100644
--- a/examples/cosmodc2/Cluster_pipelines/1deg2-nersc.sub
+++ b/examples/cosmodc2/Cluster_pipelines/1deg2-nersc.sub
@@ -6,5 +6,6 @@
 #SBATCH --nodes=1
 #SBATCH --ntasks-per-node=32
 
-source $CFS/lsst/groups/WL/users/zuntz/setup-txpipe
-tx ceci examples/cosmodc2/Cluster_pipelines/pipeline-1deg2-CL-nersc.yml
+module load python
+conda activate ./conda
+ceci examples/cosmodc2/Cluster_pipelines/pipeline-1deg2-CL-nersc.yml
diff --git a/examples/cosmodc2/Cluster_pipelines/20deg2-in2p3.sub b/examples/cosmodc2/Cluster_pipelines/20deg2-in2p3.sub
index 3f4917d5d..f97ac5185 100644
--- a/examples/cosmodc2/Cluster_pipelines/20deg2-in2p3.sub
+++ b/examples/cosmodc2/Cluster_pipelines/20deg2-in2p3.sub
@@ -5,5 +5,6 @@
 #SBATCH --cpus-per-task=1
 #SBATCH --mem=128000
 
-source /pbs/throng/lsst/users/jzuntz/txpipe-environments/setup-txpipe
+source ./conda/bin/activate
+export HDF5_DO_MPI_FILE_SYNC=0
 ceci examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-in2p3.yml
diff --git a/examples/cosmodc2/Cluster_pipelines/20deg2-nersc.sub b/examples/cosmodc2/Cluster_pipelines/20deg2-nersc.sub
index 821539970..9f67d3899 100644
--- a/examples/cosmodc2/Cluster_pipelines/20deg2-nersc.sub
+++ b/examples/cosmodc2/Cluster_pipelines/20deg2-nersc.sub
@@ -6,5 +6,7 @@
 #SBATCH --nodes=1
 #SBATCH --ntasks-per-node=32
 
-source $CFS/lsst/groups/WL/users/zuntz/setup-txpipe
-tx ceci examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-nersc.yml
+
+module load python
+conda activate ./conda
+ceci examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-nersc.yml
diff --git a/examples/cosmodc2/Cluster_pipelines/config-20deg2-CL.yml b/examples/cosmodc2/Cluster_pipelines/config-20deg2-CL.yml
index 558e5054e..80f624a2b 100644
--- a/examples/cosmodc2/Cluster_pipelines/config-20deg2-CL.yml
+++ b/examples/cosmodc2/Cluster_pipelines/config-20deg2-CL.yml
@@ -10,7 +10,7 @@ TXSourceSelectorMetadetect:
     true_z: false
     shear_prefix: ''
 
-Inform_BPZ_lite:
+BPZliteInformer:
     zmin: 0.0
     zmax: 3.0
     nzbins: 301
@@ -32,7 +32,7 @@ Inform_BPZ_lite:
 
 
 
-BPZ_lite:
+BPZliteEstimator:
     zmin: 0.0
     zmax: 3.0
     dz: 0.01
@@ -64,7 +64,9 @@ CLClusterShearCatalogs:
     chunk_rows: 100_000  # rows to read at once from source cat
     max_radius: 5   # Mpc
     delta_z: 0.2   # redshift buffer
-    redshift_criterion: mean     # might also need PDF
+    redshift_cut_criterion: pdf
+    redshift_weight_criterion: pdf
+    redshift_cut_criterion_pdf_fraction: 0.9
     subtract_mean_shear: true
 
 
diff --git a/examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-in2p3.yml b/examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-in2p3.yml
index 357c05dd0..825a1a024 100644
--- a/examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-in2p3.yml
+++ b/examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-in2p3.yml
@@ -23,12 +23,12 @@ python_paths: []
 stages:
   - name: TXSourceSelectorMetadetect
     nprocess: 30
-  - name: Inform_BPZ_lite
+  - name: BPZliteInformer
     nprocess: 1
     aliases:
         input: spectroscopic_catalog
         model: photoz_model
-  - name: BPZ_lite
+  - name: BPZliteEstimator
     nprocess: 30
     aliases:
         model: photoz_model
@@ -50,12 +50,12 @@ config: ./examples/cosmodc2/Cluster_pipelines/config-20deg2-CL.yml
 
 inputs:
     # See README for paths to download these files
-    shear_catalog: ./data/cosmodc2/20deg2/shear_catalog.hdf5
-    #photometry_catalog:  ./data/cosmodc2/20deg2/photometry_catalog.hdf5
+    shear_catalog: /sps/lsst/users/mricci/TXPipe_data/data_link/cosmodc2/20deg2/shear_catalog.hdf5
+    #photometry_catalog:  /sps/lsst/users/mricci/TXPipe_data/data_link/cosmodc2/20deg2/photometry_catalog.hdf5
     fiducial_cosmology: ./data/fiducial_cosmology.yml
-    calibration_table: ./data/cosmodc2/20deg2/sample_cosmodc2_w10year_errors.dat
-    spectroscopic_catalog: ./data/cosmodc2/20deg2/spectroscopic_catalog.hdf5
-    cluster_catalog: ./data/cosmodc2/20deg2/cluster_catalog.hdf5
+    calibration_table: /sps/lsst/users/mricci/TXPipe_data/data_link/cosmodc2/20deg2/sample_cosmodc2_w10year_errors.dat
+    spectroscopic_catalog: /sps/lsst/users/mricci/TXPipe_data/data_link/cosmodc2/20deg2/spectroscopic_catalog.hdf5
+    cluster_catalog: /sps/lsst/users/mricci/TXPipe_data/data_link/cosmodc2/20deg2/cluster_catalog.hdf5
     #shear_tomography_catalog: ./data/example/outputs_metadetect/shear_tomography_catalog.hdf5
     #source_photoz_pdfs: ./data/example/inputs/photoz_pdfs.hdf5
 resume: true
diff --git a/examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-nersc.yml b/examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-nersc.yml
index 29ddf674a..885c96647 100644
--- a/examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-nersc.yml
+++ b/examples/cosmodc2/Cluster_pipelines/pipeline-20deg2-CL-nersc.yml
@@ -6,8 +6,7 @@
 
 #for NERSC
 site:
-    name: cori-batch
-    image: ghcr.io/lsstdesc/txpipe-dev
+    name: nersc-interactive
 
 
 #all the following steps should not depend on where you run 
@@ -23,12 +22,12 @@ python_paths: []
 stages:
   - name: TXSourceSelectorMetadetect
     nprocess: 30
-  - name: Inform_BPZ_lite
+  - name: BPZliteInformer
     nprocess: 1
     aliases:
         input: spectroscopic_catalog
         model: photoz_model
-  - name: BPZ_lite
+  - name: BPZliteEstimator
     nprocess: 30
     aliases:
         model: photoz_model
@@ -37,7 +36,7 @@ stages:
   - name: CLClusterBinningRedshiftRichness
     nprocess: 1
   - name: CLClusterShearCatalogs
-    nprocess: 1   #>1 does not work with mpi
+    nprocess: 32   #>1 does not work with mpi
   - name: CLClusterEnsembleProfiles
     nprocess: 1
 #    - name: CLClusterDataVector
diff --git a/examples/cosmodc2/pipeline-20deg2-laptop.yml b/examples/cosmodc2/pipeline-20deg2-laptop.yml
new file mode 100644
index 000000000..1a16d6550
--- /dev/null
+++ b/examples/cosmodc2/pipeline-20deg2-laptop.yml
@@ -0,0 +1,116 @@
+launcher:
+    name: mini
+    interval: 3.0
+
+site:
+    name: local
+
+
+modules: txpipe
+
+python_paths:
+  - submodules/WLMassMap/python/desc/
+  - submodules/FlexZPipe
+
+stages:
+  - name: TXRandomCat
+    nprocess: 12
+  - name: TXSourceSelectorMetadetect
+    nprocess: 12
+  - name: TXLSSWeightsUnit
+  - name: TXRandomForestLensSelector
+    nprocess: 12
+  - name: TXLensTrueNumberDensity
+    classname: TXTruePhotozStack
+    aliases:
+        tomography_catalog: lens_tomography_catalog
+        catalog: photometry_catalog
+        weights_catalog: none
+        photoz_stack: lens_photoz_stack
+  - name: TXSourceTrueNumberDensity
+    classname: TXTruePhotozStack
+    aliases:
+        tomography_catalog: shear_tomography_catalog
+        catalog: shear_catalog
+        weights_catalog: shear_catalog
+        photoz_stack: shear_photoz_stack
+  - name: TXPhotozPlotSource
+    classname: TXPhotozPlot
+    aliases:
+        photoz_stack: shear_photoz_stack
+        nz_plot: source_nz
+  - name: TXPhotozPlotLens
+    classname: TXPhotozPlot
+    aliases:
+        photoz_stack: lens_photoz_stack
+        nz_plot: lens_nz
+  - name: TXShearCalibration
+    nprocess: 12
+  - name: TXTruthLensCatalogSplitter
+    nprocess: 12
+  - name: TXTwoPointFourier
+    nprocess: 2
+    nodes: 1
+    threads_per_process: 6
+  - name: TXTwoPointTheoryFourier
+  - name: TXTwoPointPlotsFourier
+  - name: TXJackknifeCenters
+  - name: TXSourceMaps
+    nprocess: 8
+  - name: TXLensMaps
+    nprocess: 8
+  - name: TXAuxiliarySourceMaps
+    nprocess: 8
+  - name: TXAuxiliaryLensMaps
+    nprocess: 8
+  - name: TXDensityMaps
+  - name: TXSourceNoiseMaps
+    nprocess: 4
+    nodes: 1
+    threads_per_process: 1
+  - name: TXLensNoiseMaps
+    nprocess: 4
+    nodes: 1
+    threads_per_process: 1
+  - name: TXSimpleMask
+  - name: TXMapPlots
+  - name: TXTracerMetadata
+  - name: TXNullBlinding
+  - name: TXTwoPoint
+    threads_per_process: 12
+    nprocess: 1
+    nodes: 1
+  - name: TXTwoPointPlotsTheory
+  - name: TXDiagnosticQuantiles
+    nodes: 1
+    nprocess: 12
+  - name: TXLensDiagnosticPlots
+    nprocess: 12
+    nodes: 1
+  - name: TXSourceDiagnosticPlots
+    nprocess: 12
+    nodes: 1
+  - name: TXFourierGaussianCovariance
+    threads_per_process: 12
+  - name: TXTwoPointTheoryReal
+  - name: TXRealGaussianCovariance
+    threads_per_process: 12
+  - name: TXConvergenceMaps
+    threads_per_process: 12
+  - name: TXConvergenceMapPlots
+
+output_dir: data/cosmodc2/outputs-20deg2
+config: examples/cosmodc2/config-20deg2.yml
+
+inputs:
+    # See README for paths to download these files
+    shear_catalog: ./data/cosmodc2/20deg2/shear_catalog.hdf5
+    photometry_catalog: ./data/cosmodc2/20deg2/photometry_catalog.hdf5
+    fiducial_cosmology: data/fiducial_cosmology.yml
+    calibration_table: ./data/cosmodc2/20deg2/sample_cosmodc2_w10year_errors.dat
+    none: none
+
+resume: true
+log_dir: data/cosmodc2/logs
+pipeline_log: data/cosmodc2/log.txt
+
diff --git a/examples/cosmodc2/pipeline-20deg2.yml b/examples/cosmodc2/pipeline-20deg2-nersc.yml
similarity index 100%
rename from examples/cosmodc2/pipeline-20deg2.yml
rename to examples/cosmodc2/pipeline-20deg2-nersc.yml
diff --git a/txpipe/extensions/clmm/sources_select_compute.py b/txpipe/extensions/clmm/sources_select_compute.py
index d3e672f51..07dbf0a60 100644
--- a/txpipe/extensions/clmm/sources_select_compute.py
+++ b/txpipe/extensions/clmm/sources_select_compute.py
@@ -7,8 +7,12 @@
 from collections import defaultdict
 import yaml
 import ceci
+import warnings
 
 class CLClusterShearCatalogs(PipelineStage):
+    """
+    
+    """
     name = "CLClusterShearCatalogs"
     inputs = [
         ("cluster_catalog", HDFFile),
@@ -26,8 +30,11 @@ class CLClusterShearCatalogs(PipelineStage):
         "chunk_rows": 100_000,  # rows to read at once from source cat
         "max_radius": 10.0,  # Mpc
         "delta_z": 0.1,
-        "redshift_criterion": "mean",  # might also need PDF
+        "redshift_cut_criterion": "zmean",  # pdf / mean / true / median
+        "redshift_weight_criterion": "zmean",  # pdf or point
+        "redshift_cut_criterion_pdf_fraction": 0.9,  # pdf / mean / true / median
         "subtract_mean_shear": True,
+        "coordinate_system": "celestial",
     }
 
     def run(self):
@@ -74,6 +81,10 @@ def run(self):
                 
             nearby_gals, nearby_gal_dists = self.find_galaxies_near_each_cluster(data, clusters, tree, max_theta_max)
 
+
+            redshift_cut_criterion = self.config["redshift_cut_criterion"]
+            redshift_cut_criterion_pdf_fraction = self.config["redshift_cut_criterion_pdf_fraction"]
+            
             # Now we have all the galaxies near each cluster
             # We need to cut down to a specific radius for this
             # cluster
@@ -99,8 +110,21 @@ def run(self):
                 cluster_dec = clusters[cluster_index]["dec"]
 
                 # # Cut down to clusters that are in front of this galaxy
-                zgal = data["redshift"][gal_index]
-                z_good = zgal > cluster_z + delta_z
+                if redshift_cut_criterion == "pdf":
+                    # If we cut based on the PDF then we need the probability
+                    # that the galaxy z is behind the cluster to be greater
+                    # than a cut criterion.
+                    pdf_z = data["pdf_z"]
+                    pdf = data["pdf_pz"][gal_index]
+                    pdf_frac = pdf[:, pdf_z > cluster_z + delta_z].sum(axis=1) / pdf.sum(axis=1)
+                    z_good = pdf_frac > redshift_cut_criterion_pdf_fraction
+                elif redshift_cut_criterion == "zmode":
+                    # otherwise if we are not using the PDF we do a simple cut
+                    zgal = data["redshift"][gal_index]
+                    z_good = zgal > cluster_z + delta_z
+                else:
+                    raise NotImplementedError("Not implemented other z cuts than zmode")
+                
                 gal_index = gal_index[z_good]
                 distance = distance[z_good]
 
@@ -165,7 +189,7 @@ def run(self):
                 indices = np.zeros(0, dtype=int)
                 weights = np.zeros(0)
                 tangential_comps = np.zeros(0)
-                cross_comp = np.zeros(0)
+                cross_comps = np.zeros(0)
                 distances = np.zeros(0)
             else:
                 # Each process flattens the list of all the galaxies for this cluster
@@ -319,35 +343,35 @@ def compute_sources_quantities(self, clmm_cosmo, data, index, z_cluster, ra_clus
         
         # Depending on whether we are using the PDF or not, choose
         # some keywords to give to compute_galaxy_weights
-        if self.config["redshift_criterion"] == "pdf":
+        if self.config["redshift_weight_criterion"] == "pdf":
             # We need the z and PDF(z) arrays in this case
             pdf_z = data["pdf_z"]
             pdf_pz = data["pdf_pz"][index]
-            redshift_keywords = {
-                "pzpdf":pdf_pz,
-                "pzbins":pdf_z,
-                "use_pdz":True
-            }
-        else:
+            
+            # suppress user warnings containing string "nSome source redshifts are lower than the cluster redshift"
+            with warnings.catch_warnings():
+                warnings.filterwarnings("ignore")
+                sigma_c = clmm.theory.compute_critical_surface_density_eff(
+                    cosmo=clmm_cosmo,
+                    z_cluster=z_cluster,
+                    pzbins=pdf_z,
+                    pzpdf=pdf_pz,
+                )            
+        elif self.config["redshift_weight_criterion"] == "zmode":
             # point-estimated redshift
             z_source = data["redshift"][index]
-            redshift_keywords = {
-                "z_source":z_source,
-                "use_pdz":False
-            }
+            sigma_c = clmm_cosmo.eval_sigma_crit(z_cluster, z_source)
+        else:
+            raise NotImplementedError("Not implemented zmean weighting")
+            
 
         weight = clmm.dataops.compute_galaxy_weights(
-            z_cluster,
-            clmm_cosmo,
+            sigma_c = sigma_c,
             is_deltasigma=True,
-            use_shape_noise=True,
-            use_shape_error=False,
-            validate_input=True,
-            shape_component1=data["g1"][index],
-            shape_component2=data["g2"][index],
-            **redshift_keywords
+            use_shape_noise=False,
         )
 
+        coordinate_system = self.config["coordinate_system"]
         _, tangential_comp, cross_comp = clmm.compute_tangential_and_cross_components(
             ra_cluster,
             dec_cluster,
@@ -355,17 +379,12 @@ def compute_sources_quantities(self, clmm_cosmo, data, index, z_cluster, ra_clus
             data['dec'][index],
             data["g1"][index],
             data["g2"][index],
+            coordinate_system=coordinate_system,
             geometry="curve",
             is_deltasigma=True,
-            cosmo=clmm_cosmo,
-            z_lens=z_cluster,
-            z_source=z_source,
-            sigma_c=None,
-            use_pdz=False,
-            pzbins=None,
-            pzpdf=None,
+            sigma_c=sigma_c,
             validate_input=True,
-        )       
+        )
         
         return weight, tangential_comp, cross_comp
 
@@ -419,13 +438,30 @@ def iterate_source_catalog(self):
         # where and what to read rom the PZ catalog. This is in a QP
         # format where the mode and mean are stored in a file called
         # "ancil". The columns are called zmode and zmean.
-        # TODO: Support "pdf" option here and read from /data/yvals
-        redshift_criterion = self.config["redshift_criterion"]
+        redshift_cut_criterion = self.config["redshift_cut_criterion"]
+        redshift_weight_criterion = self.config["redshift_cut_criterion"]
+        want_pdf = (redshift_cut_criterion == "pdf") or (redshift_weight_criterion == "pdf")
+
+        point_pz_group = "ancil"
+        # TODO: We may extend this to use other options for the point redshift
+        # in the cuts / weighting
+        point_pz_cols = ["zmode"]
+
+        if redshift_cut_criterion == "zmean" or redshift_weight_criterion == "zmean":
+            point_pz_cols.append("zmean")
+
+        rename["zmode"] = "redshift"
+
+        # if the PDF is available, load it
+        # use keyword to decide what to cut on 
+        # use keyword to decide what to weight on
 
-        if redshift_criterion == "pdf":
+        if want_pdf:
             # This is not actually a single column but an array
-            pz_group = "data"
-            pz_cols = ["yvals"]
+            pdf_file = "source_photoz_pdfs"
+            pdf_group = "data"
+            pdf_cols = ["yvals"]
+            pdf_keywords = [pdf_file, pdf_group, pdf_cols]
 
             # we will also need the z axis values in this case
             with self.open_input("source_photoz_pdfs") as f:
@@ -433,10 +469,7 @@ def iterate_source_catalog(self):
                 # but that's the kind of thing they might change.
                 pdf_z = np.squeeze(f["/meta/xvals"][0])
         else:
-            pz_group = "ancil"
-            pz_col = "z" + self.config["redshift_criterion"]
-            pz_cols = [pz_col]
-            rename[pz_col] = "redshift"
+            pdf_keywords = []
 
 
         # where and what to read from the tomography catalog.
@@ -454,11 +487,12 @@ def iterate_source_catalog(self):
                 shear_group,
                 shear_cols,
                 "source_photoz_pdfs",
-                pz_group,
-                pz_cols,
+                point_pz_group,
+                point_pz_cols,
                 "shear_tomography_catalog",
                 tomo_group,
                 tomo_cols,
+                *pdf_keywords,
                 parallel=True
         ):
             # For each chunk of data we also want to store the original
@@ -486,7 +520,7 @@ def iterate_source_catalog(self):
             )
 
             # If we are in PDF mode then we need this extra info
-            if redshift_criterion == "pdf":
+            if want_pdf:
                 data["pdf_z"] = pdf_z
                 # also rename this for clarity
                 data["pdf_pz"] = data.pop("yvals")