2f:T435,For this assignment, you will explore impacts of calcium addition on sugar maple seedlings in two watersheds at Hubbard Brook Experimental Forest using data from Juice and Fahey (2019). Sugar maple (_Acer saccharum_) growth can be inhibited by soil acidification from acid rain, which leaches calcium (a nutrient important for plant growth) from soils. Learn more: [Decades of acid rain is causing loss of valuable Northeast sugar maples, Cornell researchers warn](https://news.cornell.edu/stories/2006/05/acid-rain-causing-decline-sugar-maples-say-researchers). Stephanie Juice, Tim Fahey and colleagues at Hubbard Brook Long Term Ecological Research (LTER) site (Vermont, USA) measured “general sugar maple germinant health by height, leaf area, biomass, and chlorophyll content” (Juice and Fahey 2019) for seedlings in untreated and calcium-treated watersheds to investigate the impact of soil calcium addition on sugar maple seedling growth. In this assignment, you will explore the sugar maple data to gain insights about the impact of watershed calcium addition.30:[] 2d:{"title":"Assignment (Data science workflows in JavaScript Workshop Series)","cells":[{"id":0,"value":"# Assignment (Data science workflows in JavaScript Workshop Series)","pinned":false,"hidden":false,"mode":"md"},{"id":5,"value":"## Logistics\n\n- **Fork this notebook** to make your own copy. Complete tasks below in your fork.\n- When you are ready to submit, ensure that your fork visibility is **Public unlisted** so that we can see your work. To update the setting, go to the Share button, the under Access choose “Can view (unlisted)” for Public visibility. Make sure to Save changes.\n- Submit assignment by sharing the URL in [this Google Form](https://forms.gle/cYLwZBsyvikVwgH59) by end of day **Friday, August 4th 2023**\n- Ask questions in the Observable Community Slack #learn-javascript channel.\n- Things in yellow are where you’re prompted to add content for the assignment.","pinned":false,"hidden":false,"mode":"md"},{"id":27,"value":"## About the data","pinned":false,"hidden":false,"mode":"md"},{"id":29,"value":"$2f","pinned":false,"hidden":false,"mode":"md"},{"id":33,"value":"**Data source:** Juice, S. and T. Fahey. 2019. Health and mycorrhizal colonization response of sugar maple (Acer saccharum) seedlings to calcium addition in Watershed 1 at the Hubbard Brook Experimental Forest ver 3. Environmental Data Initiative. https://doi.org/10.6073/pasta/0ade53ede9a916a36962799b2407097e","pinned":false,"hidden":false,"mode":"md"},{"id":43,"value":"## Task 1: Access the data","pinned":false,"hidden":false,"mode":"md"},{"id":45,"value":"Access the data from the [this data package on the Environmental Data Initiative](https://portal.edirepository.org/nis/mapbrowse?packageid=knb-lter-hbr.157.3). You will only need one file: **w1_acsa_seed_physical** (found in the Resources section on that page). You can either read in the remote file directly using the URL, or download the file and attach it in your notebook. Note that while this is a text file (.txt), values are comma-separated, so you can still use the CSV methods we used in the workshop.","pinned":false,"hidden":false,"mode":"md"},{"id":58,"value":"Below this cell, add a new JavaScript cell below, then add code to load data from the file above as `maples`.","pinned":false,"hidden":false,"mode":"md"},{"id":451,"value":"maples = d3.csv(\n \"https://portal.edirepository.org/nis/dataviewer?packageid=knb-lter-hbr.157.3&entityid=5c415a399b00430d35d13c31312f1e53\",\n d3.autoType\n)","pinned":true,"hidden":false,"mode":"ojs"},{"id":66,"value":"---\n## Task 2: Take a look at the data in a Data table cell","pinned":false,"hidden":false,"mode":"md"},{"id":74,"value":"Below, visualize the `maples` data in a Data table cell. Familiarize yourself with the variables. Here are variable summaries (for the variables you’ll use in this assignment) from the [data package metadata](https://portal.edirepository.org/nis/metadataviewer?packageid=knb-lter-hbr.157.3): \n\n- **Year**: year measured\n- **Watershed**: study watershed (Reference = untreated; W1 = Calcium treated)\n- **Elevation**: categorical description of watershed elevation\n- **Transect**: study transect\n- **Sample**: sugar maple seedling sample number\n- **StemLength**: seedling stem length (millimeters)\n- **Leaf1Area**: area of first leaf sampled from seedling (cm²)\n- **Leaf2Area**: area of second leaf sampled from seedling (cm²)\n- **LeafDryMass**: dry mass of both leaves (grams)\n- **StemDryMass**: dry mass of seedling stem (grams)\n- **CorrectedLeafArea**: total leaf area for both sampled leaves (Leaf1Area + Leaf2Area), corrected for the area removed for chlorophyll analysis (cm²)","pinned":false,"hidden":false,"mode":"md"},{"id":93,"value":"Below this cell, add a new Data table cell below, selecting maples as the data source:","pinned":false,"hidden":false,"mode":"md"},{"id":454,"value":"maples","pinned":true,"hidden":false,"mode":"ojs"},{"id":97,"value":"---\n## Task 3: Data wrangling","pinned":false,"hidden":false,"mode":"md"},{"id":99,"value":"Make a simplified version of the `maples` data that only includes observations from 2003. Using the method(s) of your choice (e.g. Array methods, Arquero, SQL, Data table cell, any other tools you prefer, or a combination!), create a simplified version of the `maples` data updated as follows:\n\n- Only keep (and rename) variables:\n - Year (rename to **year**)\n - Watershed (rename to **watershed**)\n - StemLength (rename to **stemLength**)\n - LeafDryMass (rename to **leafMass**)\n - StemDryMass (rename to **stemMass**)\n - CorrectedLeafArea (rename to **leafArea**)\n- Only keep observations from 2003","pinned":false,"hidden":false,"mode":"md"},{"id":105,"value":"Below this cell, add a JavaScript, SQL, or Data Table cell to wrangle the data as described above, naming your new subset `maples2003`.","pinned":false,"hidden":false,"mode":"md"},{"id":456,"value":"maples2003 = maples\n .map((d) => ({\n year: d.Year,\n watershed: d.Watershed,\n stemLength: d.StemLength,\n leafMass: d.LeafDryMass,\n stemMass: d[\"StemDryMass\"],\n leafArea: d.CorrectedLeafArea\n }))\n .filter((d) => d.year === 2003)","pinned":true,"hidden":false,"mode":"ojs"},{"id":127,"value":"---\n## Task 4: Exploratory data visualization","pinned":false,"hidden":false,"mode":"md"},{"id":129,"value":"In the area below, create three exploratory data visualizations for the 2003 maple seedlings. Where prompted, add a Markdown cell with a 1-sentences summary of a takeaway or insight gleaned from the visualization.\n\n**Note:** Throughout the rest of the assignment, you will use the 2003 subset `maples2003` you created above.","pinned":false,"hidden":false,"mode":"md"},{"id":131,"value":"Below this cell, use the **Chart cell** to create histogram of leafArea for seedlings recorded in 2003. Facet (vertically) by watershed to compare the distributions in the untreated (Reference) and calcium treated (W1) watersheds. ","pinned":false,"hidden":false,"mode":"md"},{"id":469,"value":"Histogram = Plot.plot({\n color: {\n scheme: \"Category10\",\n legend: true\n },\n marks: [\n Plot.auto(maples2003, {x: \"leafArea\", fy: \"watershed\", color: \"watershed\"})\n ]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":210,"value":"Below this cell, add a new Markdown cell, then describe and compare the distributions for leaf areas in untreated and calcium treated watersheds based on the histograms.","pinned":false,"hidden":false,"mode":"md"},{"id":475,"value":"Viewing the distributions of the maple leaf area using a histogram shows the seedlings in the watershed that received calcium tended to have a larger leaf area than those who did not receive supplementation. That is, the distribution of leaf area for seedlings receiving calcium was shifted to the right, and appeared to vary more than the reference seedlings given the larger spread of the distribution of leaf areas. ","pinned":false,"hidden":false,"mode":"md"},{"id":181,"value":"---","pinned":false,"hidden":false,"mode":"md"},{"id":183,"value":"Below this cell, insert a **Plot snippet from the Add cell menu to create a Scatterplot** of **stemLength** and **stemMass**, with fill color mapped to **watershed**. Then, visualize the linear trends by adding a `linearRegressionY` mark atop the scatterplot mark, with stroke color mapped to **watershed**.","pinned":false,"hidden":false,"mode":"md"},{"id":481,"value":"Plot.plot({\n marks: [\n Plot.dot(maples2003, {\n x: \"stemLength\",\n y: \"stemMass\",\n fill: \"watershed\",\n tip: true\n }),\n Plot.linearRegressionY(maples2003, {\n x: \"stemLength\",\n y: \"stemMass\",\n stroke: \"watershed\"\n })\n ]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":219,"value":"Below this cell, add a new Markdown cell, then describe and compare the relationship between stem length and mass in untreated and calcium treated watersheds.","pinned":false,"hidden":false,"mode":"md"},{"id":505,"value":"Seedlings in both reference and treated watersheds demonstrated similar weak positive linear relationships between stem length and stem mass. However, the regression line for seedlings receiving calcium was above than that of the reference cohort, indicating calcium may be beneficial in improving leaf stem mass, although this may be driven in part by several outliers in the calcium group with stem mass >0.06 g. ","pinned":false,"hidden":false,"mode":"md"},{"id":198,"value":"---\nBelow this cell, add a JavaScript cell to import PlotMatrix from [Fil’s Autoplot matrix notebook](https://observablehq.com/@observablehq/autoplot-matrix). Then in a second JavaScript cell, use PlotMatrix to create an interactive matrix of charts to visualize the different variable relationships and distributions.","pinned":false,"hidden":false,"mode":"md"},{"id":490,"value":"import { PlotMatrix } with { maples2003 as data } from \"@observablehq/autoplot-matrix\"","pinned":false,"hidden":false,"mode":"ojs"},{"id":493,"value":"PlotMatrix()","pinned":true,"hidden":false,"mode":"ojs"},{"id":204,"value":"---\n## Task 5: Analysis / modeling","pinned":false,"hidden":false,"mode":"md"},{"id":228,"value":"**Choose one** of the following to explore differences and patterns in sugar maple morphology in untreated and calcium-treated watersheds: \n\n- **Option 1:** Perform simple linear regression on stem dimensions (stemLength versus stemMass) for the untreated and calcium-amended watershed. For the purposes of this assignment, you can do these separately (e.g. create separate subsets for Reference and W1 watersheds, then use simple linear regression to estimate parameters for each). \n\n- **Option 2:** Perform k-means clustering with two clusters to see if, based on the four quantitative variables in `maples2003`, the seedlings are largely within separate clusters based on watershed. Add a visualization or table to describe the outcomes of cluster analysis.\n \n- **Option 3:** Perform principal component analysis using the four quantitative variables in `maples2003`, and create a biplot showing both observation scores (points) on PC1 and PC2, overlayed by the variable vectors (arrows).\n","pinned":false,"hidden":false,"mode":"md"},{"id":237,"value":"Below this cell, add any cells necessary to perform the analysis you chose from the options above.","pinned":false,"hidden":false,"mode":"md"},{"id":541,"value":"### Task 5: Option 1 _Simple Linear Regression_","pinned":false,"hidden":false,"mode":"md"},{"id":498,"value":"ML = require(\"https://www.lactame.com/lib/ml/6.0.0/ml.min.js\")","pinned":true,"hidden":false,"mode":"ojs"},{"id":500,"value":"maples2003LM_reference = new ML.SimpleLinearRegression(\n maples2003_reference.map((d) => d.stemLength),\n maples2003_reference.map((d) => d.stemMass)\n)","pinned":true,"hidden":false,"mode":"ojs"},{"id":536,"value":"maples2003LM_calcium = new ML.SimpleLinearRegression(\n maples2003_calcium.map((d) => d.stemLength),\n maples2003_calcium.map((d) => d.stemMass)\n)","pinned":true,"hidden":false,"mode":"ojs"},{"id":511,"value":"maples2003_reference = maples2003.filter((d) => d.watershed === \"Reference\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":529,"value":"maples2003_calcium = maples2003.filter((d) => d.watershed === \"W1\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":242,"value":"--- \n","pinned":false,"hidden":false,"mode":"md"},{"id":538,"value":"Plot.plot({\n marks: [\n Plot.dot(maples2003, {\n x: \"stemLength\",\n y: \"stemMass\",\n fill: \"watershed\",\n tip: true\n }),\n Plot.linearRegressionY(maples2003, {\n x: \"stemLength\",\n y: \"stemMass\",\n stroke: \"watershed\"\n })\n ]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":244,"value":"Below this cell, add a new Markdown cell and in 1-2 sentences summarize the findings from your analysis for Task 5. Then, in 1-2 sentences based on **all** of the exploration you have done in this assignment, share your preliminary thoughts on if and how calcium addition impacts sugar maple seedlings. ","pinned":false,"hidden":false,"mode":"md"},{"id":669,"value":"### Simple Linear Regression Takeaway","pinned":false,"hidden":false,"mode":"md"},{"id":508,"value":"The slopes for the reference group (${maples2003LM_reference.slope.toFixed(4)}) and calcium-treated group (${maples2003LM_calcium.slope.toFixed(4)}) were similar and demonstrated a very weak positive linear relationship between leaf stem length and mass. Even though the intercept for the calcium group (${maples2003LM_calcium.intercept.toFixed(4)}) was below that of the reference group (${maples2003LM_reference.intercept.toFixed(4)}), the regression line and distribution of leaf area and stem and leaf mass variables indicates the addition of calcium to soil used for maple seedlings may have a positive effect on plant growth. ","pinned":false,"hidden":false,"mode":"md"},{"id":544,"value":"### Task 5: Option 2 _K Means_","pinned":false,"hidden":false,"mode":"md"},{"id":564,"value":"Data Prep","pinned":false,"hidden":false,"mode":"md"},{"id":566,"value":"import { scale, asMatrix } from \"@chrispahm/hierarchical-clustering\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":571,"value":"maples2003_scaled = scale(maples2003)","pinned":true,"hidden":false,"mode":"ojs"},{"id":573,"value":"maples2003_array = maples2003_scaled.map((d) => [\n d.stemLength,\n d.leafMass,\n d.stemMass,\n d.leafArea\n])","pinned":true,"hidden":false,"mode":"ojs"},{"id":576,"value":"maples2003_kmeans = new ML.KMeans(maples2003_array, 2)","pinned":true,"hidden":false,"mode":"ojs"},{"id":578,"value":"maples2003DTC = maples2003.map((d, i) => ({\n ...d,\n cluster_no: maples2003_kmeans.clusters[i]\n}))","pinned":true,"hidden":false,"mode":"ojs"},{"id":590,"value":"Plot.plot({\n color: { legend: true },\n marks: [\n Plot.text(maples2003DTC, {\n x: \"stemLength\",\n y: \"stemMass\",\n text: \"cluster_no\",\n stroke: \"watershed\",\n fontWeight: 100,\n fontSize: 15,\n tip: true\n })\n ]\n})","pinned":true,"hidden":false,"mode":"ojs"},{"id":600,"value":"clusterCounts = d3.rollup(\n maples2003DTC,\n (v) => v.length,\n (d) => d.watershed,\n (d) => d.cluster_no\n)","pinned":true,"hidden":false,"mode":"ojs"},{"id":667,"value":"### K Means Takeaway","pinned":false,"hidden":false,"mode":"md"},{"id":605,"value":"The k-means clustering generally did an OK job distinguishing between the observations from reference and calcium-treated locations, with ${clusterCounts.get(\"Reference\").get(1)} reference observations miscategorized as calcium-treated and ${clusterCounts.get(\"W1\").get(0)} calcium-treated miscategorized as reference. ","pinned":false,"hidden":false,"mode":"md"},{"id":546,"value":"### Task 5: Option 3 _PCA_","pinned":false,"hidden":false,"mode":"md"},{"id":633,"value":"maples2003_pca_scaled = scale(\n maples2003.map(({ year, watershed, ...rest }) => rest)\n)","pinned":true,"hidden":false,"mode":"ojs"},{"id":636,"value":"maples2003_pca_array = maples2003_pca_scaled.map(Object.values)","pinned":true,"hidden":false,"mode":"ojs"},{"id":638,"value":"// Noticed that the array I created for K means is not the same as that for PCA.\nmaples2003_array === maples2003_pca_array","pinned":true,"hidden":false,"mode":"ojs"},{"id":618,"value":"maples2003PCA = new ML.PCA(maples2003_pca_array)","pinned":true,"hidden":false,"mode":"ojs"},{"id":620,"value":"maples2003PCA.getExplainedVariance() ","pinned":true,"hidden":false,"mode":"ojs"},{"id":623,"value":"maples2003PCA.getCumulativeVariance()","pinned":true,"hidden":false,"mode":"ojs"},{"id":625,"value":"import { viewof loadings } with { maples2003_pca_scaled as food_scaled } from \"@chrispahm/principal-component-analysis\"","pinned":false,"hidden":false,"mode":"ojs"},{"id":627,"value":"viewof loadings","pinned":true,"hidden":false,"mode":"ojs"},{"id":629,"value":"// import viewof scores from the notebook, with marCasadoScaled as food_scaled and marCasado as food:\nimport { viewof scores } with {\n maples2003_pca_scaled as food_scaled,\n maples2003 as food\n} from \"@chrispahm/principal-component-analysis\"","pinned":false,"hidden":false,"mode":"ojs"},{"id":631,"value":"viewof scores","pinned":true,"hidden":false,"mode":"ojs"},{"id":648,"value":"maples2003_combined = scores\n .map((d, i) => ({\n ...d,\n Year: maples2003[i].year,\n Watershed: maples2003[i].watershed\n }))\n .map(({ Name, ...d }) => d)","pinned":false,"hidden":false,"mode":"ojs"},{"id":655,"value":"chart = Plot.plot({\n marks: [\n Plot.dot(maples2003_combined, {\n x: \"PC1\",\n y: \"PC2\",\n fill: \"Watershed\",\n r: 4,\n tip: true\n }),\n Plot.arrow(loadings, {\n x1: 0,\n x2: (d) => d.PC1 * scaling_factor,\n y1: 0,\n y2: (d) => d.PC2 * scaling_factor\n }),\n Plot.text(loadings, {\n x: (d) => d.PC1 * scaling_factor,\n y: (d) => d.PC2 * scaling_factor,\n text: \"Variable\"\n })\n ],\n color: { legend: true }\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":657,"value":"scaling_factor = 5","pinned":true,"hidden":false,"mode":"ojs"},{"id":672,"value":"### PCA Takeaway","pinned":false,"hidden":false,"mode":"md"},{"id":660,"value":"The visualization of the first 2 principal components shows that the calcium-treated seedlings were projected to the right of the reference seedlings with leaf area and stem mass measurements increasing in the direction of the treated group along the first principal component that accounted for ${maples2003PCA.getExplainedVariance()[0].toFixed(2) * 100}% of the variance in the data. ","pinned":false,"hidden":false,"mode":"md"},{"id":246,"value":"---\n## Congratulations! Submit your assignment.\n\nOnce you’re done with the assignment, make sure to: \n\n- Update your notebook settings (click the Share button) so that Public access is “View (unlisted)”\n- Submit the URL to your notebook in [this Google form](https://forms.gle/cYLwZBsyvikVwgH59) by midnight on August 4, 2023\n- Await your course certificate, which we’ll send out in mid-August! \n","pinned":false,"hidden":false,"mode":"md"},{"id":262,"value":"_**Thank you** for joining us for the Data Science Workflows in JavaScript workshop series! We look forward to seeing you in our upcoming courses and workshops, and in our Community Slack!_ ","pinned":false,"hidden":false,"mode":"md"}],"files":"$W30","stdlib":"1","titling":"auto"}