diff --git a/CHANGELOG.md b/CHANGELOG.md
index 6dc90df..bbb8591 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,3 +1,12 @@
+# Version 4.0.0
+
+- Upgrade to AWS SDK release (1.x) versions.
+- Switch error handling to anyhow.
+- Switch logging to tracing.
+
+I'm treating the logging change, moving log level from a command line
+option to an environment variable, as a breaking change.
+
 # Version 3.0.2
 
 And, um, correct the broken download links in the README.
diff --git a/Cargo.lock b/Cargo.lock
index e72f097..2add5b6 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1136,7 +1136,7 @@ dependencies = [
 
 [[package]]
 name = "query-rds-data"
-version = "3.1.0"
+version = "4.0.0"
 dependencies = [
  "anyhow",
  "aws-config",
diff --git a/Cargo.toml b/Cargo.toml
index 2c470b0..02d81ca 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "query-rds-data"
-version = "3.1.0"
+version = "4.0.0"
 authors = ["Bruce Adams <bruce.adams@acm.org>"]
 edition = "2021"
 categories = ["database", "command-line-utilities"]
diff --git a/README.md b/README.md
index ad96da5..2918fab 100644
--- a/README.md
+++ b/README.md
@@ -71,6 +71,9 @@ $ cargo build  # The first build takes longer, with more output
 $ target/debug/query-rds-data --help
 Query AWS RDS Data from the command line
 
+You can set the environment variable `RUST_LOG` to adjust
+logging, for example `RUST_LOG=trace query-rds-data`.
+
 Usage: query-rds-data [OPTIONS] <QUERY>
 
 Arguments:
@@ -79,15 +82,15 @@ Arguments:
 
 Options:
   -p, --profile <PROFILE>
-          AWS source profile to use. This name references an entry in ~/.aws/config
+          AWS source profile to use. This name references an
+          entry in ~/.aws/config
 
           [env: AWS_PROFILE=]
 
   -r, --region <REGION>
           AWS region to target
 
-          [env: AWS_DEFAULT_REGION=]
-          [default: us-east-1]
+          [env: AWS_REGION=]
 
   -c, --db-cluster-identifier <CLUSTER_ID>
           RDS cluster identifier
@@ -106,16 +109,15 @@ Options:
 
           Possible values:
           - csv:  CSV output, including a header line
-          - json: An array of JSON Objects, {"field_name": field_value, …}
+          - json: A JSON Object. For example:
+            {"numberOfRecordsUpdated": 0, "records": [{"id": 1,
+            "name": "bruce", "amount": 0.05}]}
 
   -d, --database <DATABASE>
           Database name
 
           [env: AWS_RDS_DATABASE=]
 
-  -v, --verbose...
-          Increase logging verbosity (-v, -vv, -vvv, etc)
-
   -h, --help
           Print help (see a summary with '-h')
 
diff --git a/src/main.rs b/src/main.rs
index e8a18e7..10440df 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -22,12 +22,18 @@ use tracing_subscriber::{fmt, prelude::*, EnvFilter};
 enum Format {
     /// CSV output, including a header line.
     Csv,
-    /// An array of JSON Objects, {"field_name": field_value, …}.
+    /// A JSON Object. For example:
+    /// {"numberOfRecordsUpdated": 0,
+    ///  "records": [{"id": 1,
+    ///               "name": "bruce",
+    ///               "amount": 0.05}]}
     Json,
 }
 
-/// You can set the environment variable `RUST_LOG` to
-/// adjust logging, for example `RUST_LOG=trace aws-caller-id`
+/// Query AWS RDS Data from the command line
+///
+/// You can set the environment variable `RUST_LOG` to adjust
+/// logging, for example `RUST_LOG=trace query-rds-data`.
 #[derive(Clone, Debug, Parser)]
 #[command(about, author, version)]
 struct MyArgs {