Week 3 Lab (Answers) — The Dimension Matrix

import%20marimo%0A%0A__generated_with%20%3D%20%220.25.2%22%0Aapp%20%3D%20marimo.App(%0A%20%20%20%20width%3D%22medium%22%2C%0A%20%20%20%20app_title%3D%22Week%203%20Lab%20(Answers)%20%E2%80%94%20The%20Dimension%20Matrix%22%2C%0A)%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%20Week%203%20lab%20%E2%80%94%20the%20dimension%20matrix%20%E2%80%94%20Answers%0A%0A%20%20%20%20**MATH4120%20%E2%80%94%20Mathematical%20Modelling%20and%20Programming**%20Lancaster%20University%2C%202026%E2%80%9327%0A%0A%20%20%20%20---%0A%0A%20%20%20%20**This%20is%20the%20answers%20version.**%20Every%20blank%20below%20contains%20the%20completed%20solution%2C%20for%20instructors%20and%20for%20checking%20your%20own%20work.%0A%0A%20%20%20%20---%0A%0A%20%20%20%20Work%20through%20the%20exercises%20below.%20%20Press%20**Ctrl%2BEnter**%20(or%20**Shift%2BEnter**)%20to%20run%20a%20cell.%0A%0A%20%20%20%20The%20main%20task%20is%20to%20build%20a%20dimension%20matrix%20as%20a%20NumPy%20array%20and%20use%20%60null_space%60%20to%20find%20the%20%CE%A0-groups%20automatically.%20%20Anything%20that%20depends%20on%20your%20matrix%20will%20update%20as%20you%20fill%20it%20in.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_()%3A%0A%20%20%20%20import%20numpy%20as%20np%0A%20%20%20%20import%20matplotlib.pyplot%20as%20plt%0A%20%20%20%20from%20scipy.linalg%20import%20null_space%0A%0A%20%20%20%20return%20np%2C%20null_space%2C%20plt%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%23%20The%20tool%3A%20%60null_space%60%0A%0A%20%20%20%20%60null_space(D)%60%20returns%20the%20vectors%20%24%5Cmathbf%7Bk%7D%24%20such%20that%20%24D%5Cmathbf%7Bk%7D%20%3D%20%5Cmathbf%7B0%7D%24.%20Each%20column%20of%20the%20result%20is%20one%20set%20of%20exponents%20for%20a%20%CE%A0-group.%0A%0A%20%20%20%20After%20calling%20it%2C%20normalise%20the%20output%20so%20that%20the%20variable%20you%20are%20solving%20for%20has%20exponent%201%3A%0A%0A%20%20%20%20%60%60%60python%0A%20%20%20%20v%20%3D%20null_space(D)%5B%3A%2C%200%5D%0A%20%20%20%20v%20%3D%20v%20%2F%20v%5B0%5D%20%20%20%20%20%20%20%20%20%20%23%20set%20the%20first%20variable's%20exponent%20to%201%0A%20%20%20%20np.round(v%2C%203)%20%20%20%20%20%20%20%20%23%20display%20as%20rounded%20floats%0A%20%20%20%20%60%60%60%0A%0A%20%20%20%20The%20entries%20of%20%60v%60%20are%20then%20%24%5Bk_1%2C%20k_2%2C%20%5Cldots%5D%24%20such%20that%20%24%5CPi%20%3D%20x_1%5E%7Bk_1%7D%20x_2%5E%7Bk_2%7D%20%5Ccdots%20%3D%20%5Ctext%7Bconst%7D%24%2C%20and%20the%20scaling%20law%20for%20%24x_1%24%20follows%20by%20rearranging.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%23%20Exercise%201%20%E2%80%94%20terminal%20velocity%20of%20a%20raindrop%0A%0A%20%20%20%20A%20raindrop%20of%20weight%20%24W%20%3D%20mg%24%20falls%20through%20air%20until%20gravity%20is%20balanced%20by%20drag.%20At%20terminal%20velocity%20%24v%24%20the%20relevant%20parameters%20are%3A%0A%0A%20%20%20%20%7C%20Parameter%20%7C%20Symbol%20%7C%20M%20%7C%20L%20%7C%20T%20%7C%0A%20%20%20%20%7C-----------%7C--------%7C---%7C---%7C---%7C%0A%20%20%20%20%7C%20Terminal%20velocity%20%7C%20%24v%24%20%7C%200%20%7C%201%20%7C%20%E2%88%921%20%7C%0A%20%20%20%20%7C%20Weight%20%7C%20%24W%24%20%7C%201%20%7C%201%20%7C%20%E2%88%922%20%7C%0A%20%20%20%20%7C%20Air%20density%20%7C%20%24%5Crho%24%20%7C%201%20%7C%20%E2%88%923%20%7C%200%20%7C%0A%20%20%20%20%7C%20Cross-section%20%7C%20%24A%24%20%7C%200%20%7C%202%20%7C%200%20%7C%0A%0A%20%20%20%20%24n%20%3D%204%24%20parameters%2C%20%24k%20%3D%203%24%20base%20dimensions%20(M%2C%20L%2C%20T)%20%24%5CRightarrow%24%20%24m%20%3D%201%24%20%CE%A0-group.%0A%0A%20%20%20%20**Task.**%20Fill%20in%20the%20dimension%20matrix%20below.%20Each%20column%20is%20one%20parameter's%20exponents%20in%20the%20order%20%5BM%2C%20L%2C%20T%5D.%20Run%20the%20cell%20%E2%80%94%20the%20null%20space%20below%20will%20update%20automatically.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_(np)%3A%0A%20%20%20%20%23%20Dimension%20matrix%3A%20rows%20%3D%20%5BM%2C%20L%2C%20T%5D%2C%20%20columns%20%3D%20%5Bv%2C%20W%2C%20rho%2C%20A%5D%0A%20%20%20%20D_term%20%3D%20np.array(%5B%0A%20%20%20%20%20%20%20%20%5B0%2C%201%2C%201%2C%200%5D%2C%20%20%20%20%23%20M%0A%20%20%20%20%20%20%20%20%5B1%2C%201%2C%20-3%2C%202%5D%2C%20%20%20%23%20L%0A%20%20%20%20%20%20%20%20%5B-1%2C%20-2%2C%200%2C%200%5D%2C%20%20%23%20T%0A%20%20%20%20%5D%2C%20dtype%3Dfloat)%0A%20%20%20%20D_term%0A%20%20%20%20return%20(D_term%2C)%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%23%20Exercise%202%20%E2%80%94%20compute%20the%20%CE%A0-group%0A%0A%20%20%20%20**Task.**%20Run%20the%20cell%20below%20to%20compute%20the%20null%20space%20of%20your%20dimension%20matrix.%20Once%20%60D_term%60%20is%20filled%20in%20correctly%2C%20%60np.round(v_term%2C%203)%60%20should%20show%20%60%5B1.%2C%20-0.5%2C%200.5%2C%200.5%5D%60.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_(D_term%2C%20np%2C%20null_space)%3A%0A%20%20%20%20ns_term%20%3D%20null_space(D_term)%0A%20%20%20%20v_term%20%20%3D%20ns_term%5B%3A%2C%200%5D%0A%20%20%20%20v_term%20%20%3D%20v_term%20%2F%20v_term%5B0%5D%20%20%20%20%23%20normalise%3A%20v%20exponent%20%3D%201%0A%20%20%20%20np.round(v_term%2C%203)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20When%20%60D_term%60%20is%20correct%20the%20exponents%20%24%5Bk_v%2C%5C%2C%20k_W%2C%5C%2C%20k_%5Crho%2C%5C%2C%20k_A%5D%24%20are%20%24%5B1%2C%5C%2C-1%2F2%2C%5C%2C1%2F2%2C%5C%2C1%2F2%5D%24%2C%20giving%20the%20%CE%A0-group%3A%20%5C%5B%5CPi%20%3D%20v%20%5Ccdot%20W%5E%7B-1%2F2%7D%20%5Ccdot%20%5Crho%5E%7B1%2F2%7D%20%5Ccdot%20A%5E%7B1%2F2%7D%20%3D%20%5Ctext%7Bconst%7D.%5C%5D%20Rearranging%20for%20%24v%24%3A%20%5C%5Bv%20%3D%20C%5Csqrt%7B%5Cfrac%7BW%7D%7B%5Crho%20A%7D%7D.%5C%5D%20The%20constant%20%24C%24%20must%20be%20determined%20experimentally%20or%20from%20the%20full%20fluid%20equations.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%23%20Observation%20%E2%80%94%20physical%20interpretation%0A%0A%20%20%20%20The%20scaling%20law%20%24v%20%3D%20C%5Csqrt%7BW%2F%5Crho%20A%7D%24%20tells%20us%20how%20terminal%20velocity%20depends%20on%20each%20parameter.%0A%0A%20%20%20%20**Task.**%20Double-click%20this%20cell%20and%20complete%20the%20sentences%3A%0A%0A%20%20%20%20-%20A%20heavier%20drop%20(larger%20%24W%24)%3A%20terminal%20velocity%20**increases**%20%E2%80%94%20%24v%20%5Cpropto%20%5Csqrt%7BW%7D%24%2C%20so%20a%20heavier%20drop%20must%20fall%20faster%20before%20drag%20can%20balance%20its%20weight.%0A%20%20%20%20-%20Denser%20air%20(larger%20%24%5Crho%24)%3A%20terminal%20velocity%20**decreases**%20%E2%80%94%20%24v%20%5Cpropto%20%5Crho%5E%7B-1%2F2%7D%24%2C%20since%20denser%20air%20gives%20more%20drag%20at%20the%20same%20speed.%0A%20%20%20%20-%20Larger%20cross-section%20(larger%20%24A%24)%3A%20terminal%20velocity%20**decreases**%20%E2%80%94%20%24v%20%5Cpropto%20A%5E%7B-1%2F2%7D%24%2C%20exactly%20why%20a%20parachute%20(huge%20%24A%24)%20falls%20slowly.%0A%20%20%20%20-%20Why%20does%20the%20bob%20mass%20cancel%20from%20the%20pendulum%20but%20mass%20appears%20here%3F%20For%20the%20pendulum%2C%20mass%20enters%20both%20the%20driving%20force%20(gravity)%20and%20the%20inertia%20term%20and%20cancels%20algebraically.%20Here%2C%20%24W%24%20*is*%20the%20driving%20force%20itself%20(already%20a%20weight%2C%20not%20a%20separate%20mass)%2C%20and%20there%20is%20no%20second%2C%20independent%20mass%20term%20for%20it%20to%20cancel%20against%20%E2%80%94%20so%20it%20survives%20in%20the%20scaling%20law.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%23%20Exercise%203%20%E2%80%94%20be%20G.%20I.%20Taylor%3A%20the%20energy%20of%20an%20atomic%20bomb%20from%20one%20photo%0A%0A%20%20%20%20In%20the%20lecture%2C%20dimensional%20analysis%20gave%20the%20radius%20%24R%24%20of%20a%20blast%20wave%20at%20time%20%24t%24%20after%20an%20explosion%20of%20energy%20%24E%24%2C%20in%20air%20of%20density%20%24%5Crho%24%3A%20%5C%5BR%20%3D%20%5Cleft(%5Cfrac%7BE%5C%2Ct%5E%7B2%7D%7D%7B%5Crho%7D%5Cright)%5E%7B1%2F5%7D%5C%5D%20(taking%20the%20constant%20%24C%20%5Capprox%201%24).%20In%201950%20G.%20I.%20Taylor%20used%20this%20to%20work%20out%20the%20energy%20of%20the%20first%20atomic%20bomb%20(the%20Trinity%20test%2C%2016%20July%201945)%20from%20photographs%20that%20had%20been%20made%20public%2C%20while%20the%20energy%20itself%20was%20still%20top%20secret.%0A%0A%20%20%20%20Here%20is%20one%20of%20those%20photographs%2C%20taken%20**0.025%20s**%20after%20the%20explosion%2C%20with%20its%20**100%20m**%20scale%20bar.%20Drag%20the%20slider%20until%20the%20green%20circle%20hugs%20the%20top%20of%20the%20dome%20(ignore%20the%20dust%20along%20the%20ground).%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_()%3A%0A%20%20%20%20%23%20The%20Trinity%20photograph%20(0.025%20s)%2C%20stored%20as%20text%20so%20the%20notebook%20works%20anywhere%2C%20even%20offline.%0A%20%20%20%20%23%20Public%20domain%20(US%20Government)%3B%20source%20at%20the%20end%20of%20this%20exercise.%0A%20%20%20%20trinity_jpg_b64%20%3D%20%22%2F9j%2F4AAQSkZJRgABAQAAAQABAAD%2F2wBDAAsICAoIBwsKCQoNDAsNERwSEQ8PESIZGhQcKSQrKigkJyctMkA3LTA9MCcnOEw5PUNFSElIKzZPVU5GVEBHSEX%2FwAALCAIIAroBAREA%2F8QAGwABAAMBAQEBAAAAAAAAAAAAAAECAwQFBgf%2FxABCEAACAQMDAgUCAwUGAwgDAQAAAQIDBBESITEFQQYTIlFhFHEygZEVI1Kh0QczQpOxwVSS4SRDU2JygvDxJTREVf%2FaAAgBAQAAPwD8pk8vPBUAADJBJAJyACABkAZJAIyMgZBORkZGRkZGRkZJyMkZGRkZGScjJGRkZGScjIyMjJGRkZGRkZGSckZJyMkZJyMkZJyMjJGSdRGonUNQ1DI1E6nhoKTJjLdEt7vYrnBVkEkAAAAAEAkEEkAkAAEEgAZAAAAAAAAAAAAAAAAAAAAAAAAAyCyLPdsiWzZUgAAgkAAgAAAZAAAAAJBABIAAAAAABAAAAAAAAAAAAAAAABIJTNE1gpJ5e5UAEAkgkAgAAAAAAAAAAAAAAAAAEkAAAAAAAAAAAAAAAAAAkIuRIqAAAQSQAAAAAAAAAAAACQCAAAAAAAAAAAAAAAAAAAAAAACUWDKggkgEggAAAAAAAAAAAAADuAAAAAAAAAAAAAAAAAAAAAAAASi2A%2BSrAIABJBJAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABZE4D5IIAIJAIJBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABJKLpvBVkEAgkAAgkAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAklFtyGQQAQCQCCQAAQAAAAAAAAAAAACSCSAAAAAOQAAAAAAAAAAAAASCUWIZBAAIJAIJIBIBAAAAAAAAAAABJBIBAJBAAAAAAAAAAAAAAAAABIJRYhkEAAAAAAgkEAAAAAAAAAAEgAAEEgAgAAEkAAAAAAAAAAAAAEglFtyGQCAAAAQACQQAAASQSQSQSCACQAQSQSBgAAAAgEgEAAAAAAAAAAAAkAEosQyAGQAAAAAAQAAASQCQAAAAAMbgAnBAJIAGOABjIABAAAAAAAAAAHcAEgAlFiHyQARgAAAAAEEggAkgkAAAAAkYNaVtWrPFOlOX2R20eg39aSXk6V7yeDqp%2BFryTxOdKH55N14SqL%2B8uIr7RySvCmZJO5wnw9IfhPd%2FwDaeHjeJlPwxKLaVzF4eODJ%2BHKucedD9CJeHLhYxVpP9URLw1ex48uX2kZT6B1CD2oOX%2FpaZyVLC6pZ129SOPeLMHFx5WPuRggE5IYAIAAAAAAAAABLIAJABZEkMgAgAAAAAAAAAAAAAE4GDstemXV4%2FwBzRk1%2FE9kezbeFdlK6rf8AtgevR6NY26Sp26lJf4p7nZGMYv0RwvhBrKS9iPLlJJYb%2FwBTWFKb3bwbK3cpJTeUvctCzhnOHh9vYit0%2BGrK3M5dOWc4%2FQwqdP2WlPb55KO1aeG5LbZinKpTnxlm8ajk203n2Yq2tvcR%2Fe29Oaa3zE82v4Z6dcZcITov3i9jybrwdc08u2qwqx9nszw7qwubOWLijOHy1sc5AAIABJAAAGAAAB3A7AEgAkIuVZABAAAAAAAAAAAABOBgnGT0rDod1etPT5dP%2BKR9JZ%2BHbS1gpVF5tTPMt1%2Bh7FGEEsR9KxsiakWpPfZcmeFjbjuiF92XjFY1YWDbLaWyZZrLyuV8Gyilh4Wy57loyaWpspOXqaXLL21pdXmXQpynGPLS2L1OjdTW6oza9kcVanXoVdNaEotrOJLBjrcWsxy2XUVOWMYz7kqGFpf6FNDSbTazsmVVWcXvvsTKVKtBxqwUljvueNf%2BGLS5TnbPyZvhLhnzN70a7sW9dNyj%2FFHdHBggAEEggEgAgAAAAdgCRkAklFslWQAQAAAAAABgAAAAEk4PT6b0O76nJeXBxp96klsfWWPhqzsIqpP97UW%2BWei1FR9PpXwSm3FJR2TyVjHK29yzwks7%2FHcxWG8Yxklx31ZWCyg4rk2px3xv7l8e%2FGTSMlslj5Y0bt%2FoJUsvCe%2FY%2Bw6JKnb2NOnsmluesq9Nfhab%2BDz%2Br29pd281UhFyUXpfdP4PhpWjgn75K%2BXJP3fuJZb37bbGcscY%2FUznjGMcGakvYnVthPgs9MklJpp8o8fqHh%2B0vMypfuqvvFbP8j5a%2FwClXPTqjjWh6e01wziIAAAAAIAAAAJIBIAJCLEMgAAgAAABAAdgAAACcGtC3q3NWNKjBznLhJH2PSPCVOhprdQanJbqmuF9%2Fc%2Bhco%2BWlTjpguyWBFx3Rm4%2BptLA1Rit3l%2FBVVMv2SJ1Rk0nJJZ5ZWVSlTnJalLG2TPzKSaWXj%2BRo69KEsRlnO%2FBqq%2BtKW2FwS6sZLCe63La1h8M0SWlM29OjZvUaW93UpNpyaitj0KfUZKG0nsZVrqVaGFJtvZsznSh5aW%2BcHM6D0tdjCrT3x%2FsZYxlY55eDCdN7ZZg04vJRprfswnpxuSp%2BvZ8FbhQrqUKkdSlymtj5jqnh9006tmm13p919jwGmm01hr3IAAABBIIJIBIBBK5QIJBBIJRbJBABAAJBAAAHcAAAkAI9TpPRLjqlT0rRRT9VR8H3XTul2vS6KjQgtb%2FABTfLOlqTlqSfJPl68OTf2RWtc0bZaYpSn7Hm1budWcnjT3J%2FwAOfMztlpEwylnnPBZJyePjbJV03u%2B2cERpJNOWTWlbZllPGXsWjRlDs8Pk2p0UprOWiXBwlvwuDWm28rCZ04xNLG4eJfdl4w2f29zem4wxlLH25JqzzlrZMyU8rGfzJcE5b9yPpqbbef5mVS0ko7rK7HLUtZRy129ytzOVakoSglp3TS7HC4eW1l8ozm0pTabx2KSk9g5aofOOTy%2BpdIpXqc6eKdf37S%2B58tWoVLeo6dWLjJdmZAkgAAEEgEEgEDuS%2BSASCCSUWIYIfcAEAAALkAAAAkAEnv8ARPDs7uUa92nChyk9nP8A6H2OI0KMIU1GEUsJRXYvTlpj2b%2F0IlWqS2WPsVr3n08Mczfs%2BDzNMqtRy3eWbxo%2FiWO5vSt%2FMTW0cLuXjR37suqOWyrpbtLfJoqK9vg6KdJLnY0jRWyxt7lvLSWMZ%2F2MnSztjf2I8lrKe2exfymmvU3sWUXBrO5rFZbWdJLTai1x3KzqS2T4Ca937lm8rGcY7l4Qyk3%2FAKmvmPOG9iKsVLsmc1SlFrjBx1raDzjf8jG7UakYxVNRlHulycbo6pRhBZ%2B5jUh5UmpLdfJXHwzjv%2Bn07yjiot03pmuUfI3NtO1rOnUW67%2B5kQSQAAAACCQCCWsYfuiCQACUX%2FMqw%2BXkhgAgn7AAEAAkgAnsACUj6PofRMShdXsVp5hTl3%2BWfSOo3LdrCWMI3jFuEM4eMss8rKS5Iq1o29LLa1y4XseeoTrVHJpuUjrt6UoVNWOPg7Kds5ybayaq39WMY9zVUUVdLDZV0ksNdiyj32y%2FYusJb8GkcJcbFml8totHMJqWMv7Fqso3FRNpR2xsZzpYe3CWxDpOOM753L%2BVjCaw8FlS9LzwZTp4fGyM8ad0TB%2FO5pCp8kqSLKptu9w%2FUvj7GcqLktjnrUGlnHJxTt98nNXpyc8t5Xcwknhv%2BRWtRnBLXCUdXGUcF5Y072io1Np9n3R8rdWlS0q6ai27PszAAgAAAAAAFnvTj8NoqAQSCUXUVjkhkPuHnLyQAAudyASAQAT2IJIJAJSPpeh9EUVG7u48%2Fgg%2F9WfQadcnw%2F5GkKUuGvzOuEGlnLwvZiqo0aPmS39l7nDGDuJ%2BZPHPB6PlxqaUoJNcYOilbpYeN3sdboQpJNS9Wd0TGK1ZzlstKnlIpKMd8rcxlBe3H8yrS9Oxdadk0tuSZ6U%2FSi8Nv%2FnJpGOY78l1TS7bE%2BV6cPYiVF6VjOzJcHKXvsWjHCNE9NNxcVLJyTguDB08SbWxWO3GefYSnF4ZEJLL9%2F8AU3i3sn27G0Vtzn3RnUpaonLOiscbexy1KOY7nC4OD27bpsV7mpcUo05Y9L5PPrRWduVwzjr0KdxTlTrLKfD9mfOX1hOyqJS9VOX4ZLucgBAAAAAABaO9KS9mmVAAJCL%2FAJlWA%2BSAAAAAQCQAQSCcH0nh%2Fofm6by6j%2B75pwf%2BL5%2Bx9LOM5S9Sxntg6KFrJ7tbPc7HGMNLS3K1dENTaSx3PNqt3FXKy4rg7be11LHznB6FO3wm0s7G8aed84%2BCVFN55zwEvbb5NMfxGco5%2FQzlFuL2MXDDCXtgvCDymzXy9vn3NYw23WzNYp44LQWdpZLuGFpW%2BTNpRZTdvMffuXpvQnqWc%2B5WUVz2e5hUppvZnO4Y4eexkorOOSuEvubRmsYydNOWVnP6Ftm2s8mdSmm99jmqRWGnscFWkm3I5WsNJrjuctSPq478HZedPt30yN1bvEovEot8ngVaUK0Z0KyzTfD9j5q8tJ2ddwlx2fujmBAAAAAABaD%2FABL3iVIJABKL%2FmiMFe4YAAAAAAAABJ7PQOkfX3CqVk1bwe%2F%2FAJn7H2tOCjhpbR2S7I6VTlVlrm8r4OmT001GPHYrKW6clu%2FY4L2q5ScP4nua28GoKGrVBcHp0acVBbbnTHChznJKj6SsnjhFsJxT7kvLe25Et17Mya9ucGeG9%2BxGHk2jDbftwawXusGyinhPYsobJMso437k9n2wZOOuWXwJPDM5y7DmKREqezawc9WnjdGDT%2FIpjL3K5zLCz%2BRtSqSWx0Qlq27icW%2F6mM1jnc55Rk09JwratFrfS%2B5zXWKtRyUdPbBySjJrS5PSnlo5rimpPMc47Z7nn3NCN1RlSn%2BKO8Zex87UhKnNwksNPBQAgkEAAAAmG019yMYeAACSUWKsgl8kAAAAAAAAA6bK0neXUKNNbye79l7n6BZ2tO2t4UaS0xisLHf5O6EY5XKT7YOlYiltt7Cc0optLbhGUpuniTe74S7I4ZU3UknnLby2ejaU8YTXwdtPMXvlnQnnGV%2BhLwn3xxuUmtSjjOfgtS%2FMu4YeV3KyW%2B7%2FACKTXdcmag5PjktGnnKyXhDKwbxj6ce3c0jHfcvp1SbJcedijy%2BWVb90ZyeeMlJPd9hFLDe%2FwMvPzyVlFywc1WDWcGEnpy8FHhvYmE9zopzWEbqWcfOxScTnqwa34OVwWr2RSraN0tcU9OcZ%2BTza9Fp5OScXGOnbffB51d4qau%2Bdzzep2yqw8%2BC3X4vk8cAAEAAAAcblqv8AeSxw9yoAJJRIZUlkAALkPkAAAAAE4Pr%2FAA3090LZ3M4%2Buqts9kfRQjmUUt8bHVBYXH5vk0UkuX%2BhTZvM98djGpHZZX8zOEXrTwejRj6dmddODePY3kvnYacrfchJttJExjj7luOchrbsZ%2B7KvuXS9OotT%2BN0aLLmn7HRD8JdLHAlHbg56j7RKt5ik%2B%2FwU07vYpL9CEm3uTFEtZWEtzGpF432OOpDnK7GT9PBXOXjvg0pvCWX%2Bh005LO5pkzq7fKRzSi3Lh5KYawnw%2BUctxQTyebWp%2BvOnG%2FY8q9pYnNJPk5IPDcJcPseJeW7t67jj0veJzgAEAAAAFp7qL%2BCoAJJRYggEAAkPkgAkEEgglA7el2f1t9Tpv8ADnMvsfeUniUI01iK2SPQhDKWVlv2NJy0Q3KQlq1auPuarEt89hKOVullcGUYLP8A8Z6FNfu8KKzz9jqh%2BFd2aqHp3ReEMLJZxjjbJVU8S1cFZQ0pvJlP8Jm5bfciMk8mkXtjuXim3yaQ5W2%2FsdKaxhF4rYiW79vgwqLPPYppa5f6lXsv6lJR7ohLKTf5kpZe5bGpY7mc1iL3eTknB%2Fkc04uMjNx4%2FwBSYtp4N4SxL27mrnlZfHwxystfL%2BSmh5KOLby1yUqU8xz3OC5pJRzj8zxr6jOOcxxnfDPOt6VOVwo1XhPbKMetWadOai9Uqb2fuj5wAAgAAAAtzTXwyoAJJRYghhkDAADABIABBJKPq%2FDdl5dDz5%2Bl1Xt9j6S3glhvfY71HEIYa3%2Fkctao3JR4XctSh5kVssP%2BZ2Qp6o8YaLSpvvzj9SsaOJbxwdlKCwsbe51wjFR5N4QWMyXBpKO2y5EYLj4KOOOEZ1e%2BDCTWj8zmk%2FWsbkJtb9jSG63x9zaLSWFua016ktR0KPG%2FHcvHGEuA%2BNjJxbe5SSWl77mWNtyrj7P5IXP9TRQbfGC6jsZzWc7HPKnujnrU3Jb9vY55LTnO2%2B%2BSqxndIslnGP5F1vHPsTGXK4NsJrKeSuGt0itR6sJnLWpqcG87ni3yk5LX6tsI8C6TjLMdn2wZ0pynTlHOX7M8O6pOjcSjjCzlGIAIAAAALR%2FBJfZlQASSi5VlXySyCUsp%2FCyQAAASAAAbW1B3FxTpR5k8H30IRo0IxjhRgkoo9Oj5bowlHaS5RepUTisNL8zGpTjVmlHOlHoWtNaXnbHGDrUEsvGX7BQcm1%2FoKlPePBpCLjlv%2FwCjpTWOTaM8xWMGq4wOd8%2FBEppL5OarLVhrPBzzWU%2Fk5nnJCkuOTWLSWDWEsbGtN%2Br%2FAGOuHKTZfnLbw3yRjb%2FYyedykkZyXxtkr%2BbLKOrs0bJYxnk3t6Mqr2jt7mlS0cOMNHDXotSeFjHJw1oYfBy1ob7djFLLT9nySm0s8lpVdWFjDX8yMyUvZcl1Ucc8r7llUc90917kvd5wUnTUlnCZ5HUKeE2uEtj527p%2BnL2S2yefH0T22yY9Yo5hCql8M8gDsCAAAAC9L8TS7pooASCUXIfYqCCYvn7EAEsgAkAAEnveGrXVWqXEltBYj9z6SDc8LGyZ6NOpGEFl4SRXU51M8ZPRtbbVJtnpqgor2ZOjHpxn7E%2BW87LfnkiVPj5ISwlvyWxhb7o1j6Vk2pyWEnn7lnL52KvdtIwqemDbxycc6mXiJzuSa2TITw%2Fk0Ty%2BH%2BhtGS2R00c9zrjh%2FwBS2Ntw47bMzwUe3cpLeWO%2BeCuOxdLHv%2BZvRWqab2SPXopKKxsaSjtk86%2BoJYl3fO55NSnztwcVSHOdmckliW5GVjZZKtZeePglzbW73yU1JrGS0ZttG8Z5fyaNx07t57nm3yUo8cnzl5TSzt9jypx9X2LV4efaSh8bHzr5AAIBIAIBMHpmn7MSWJNezIBIRJcqQCCY8kAEsgEgAAFkj7HpVD6XptOOHqn6n%2BZ6FLKi44z3y1udtunUnnGcM9WhauSzhYiehQprL0pfY61BaePzCgm9sbrY0p0tnxuY1oNJfBzOWmSb7MtqW%2FyaJtLaWC0J6eWTKrh7LBWU21ly5MKj9Oz7nOvxbtezZk6mEk0sGcp4bWd2aU5%2BlbvP%2Bp0Q%2FB255Oqi9kdVN5xj8zV8p4ElndFGlJYMpJYM2s%2FqEk5ZNHjGDWhJQaxv757np0prC9vY2c1jb2Oe5cZQeXuePWaS2exwyim3ng5a0VqbjvBbGCWqWFt9yJ7PcpPbh4Mm2tiKVRt6ZfkdEZ4kt8Grm2lvk4rptRbcjxbuOf6nlXMYal5afG5jGWmWGjxLiEadxUi887YMAAOxBIAIALVPx5fdZKgkEouVKgDhgEgAAkgYJwDpsLd3V7SpLiUln7H2k8KSS%2FDFLG50W7XmbpvK4yezZwi2tNPGFj7nqUqWUtsPhpndSpLC24fJsovfb8iVTSXGGjVJLtyjC4gms557Hm1INPbYjjbJMaml4fJZT9POfsTKpt7YMZVG8e3czlPvnYwlP1bcGcnnsVymzWHbBtTb1aXwdtLtudlOS2SNnwu4a5xx7GUt5fHsZzy1hdzPcNNLYlvP3%2BTWOJcbG9Os6ccP8zaVz6Mx3wc1WvnbO5xVJt5Oecco555invycq2nuXcU17bmVSGd%2BxzuOFnv7lcPnubxaaWMZNI7xxgxuI5XdI8u5prQ2vsePWp6pM4qi0yPL6nHFzq%2FiimcQAAABAALS3jB%2FGCoJBKLbEMgEAcAEgAnAABJ9D4WtddatcNbQWlfdnty9U2scLY6LFKpUe2y9j6WxjKUFJacLvn2PVhFp52aZ2RgmnhbexdQ3%2FwBiY0%2FgvKO2Ut%2BxzVItZOCvTxJZ3MH3fPcpjvnchSwnuJPK90ZyzhMzlxjHPuYy%2BCqfsRKDb%2FoXjlPnk6YJ6U%2BWjsoNSW%2Bdjqg91lfobZwnu9xF4TIk1l5RlLHK3ZSXPsijb432Dm3hbFozxzuaKrrXbZ7EZa7lJP1bmEnk56jxtjJk8NP3MsRUuCySxnncznFr7GEo%2Fi9l3MnnG%2BVg1pxxFYX%2FAFNUZ1MODT5OCvFOLj2%2BTyLiniTytkeZXjh8M8zqSyqcvujzyAAAACAC3%2Fdv4ZUAklFipD5AAJAJI5JAAwSSj7nw%2FQVDo8Nlqq%2Bp%2FwCxevmMZbb8ZZ19Li5YxH9D6qyjpoxUklJ7s9SjDbjB0wWH7ZNVFOb%2FAELaPVjclrMcHNOKycdzHMcrf5Z5MpLU0wqmxXV7Eyks5xt8lW39smWMopjZ%2B5Dg9hKL1IjHr%2BxvSlskzrovc64yxxwjZS9JDkl%2FQrKfyZtvU9uSM9t9irllYRlN%2BzIjLG%2BCyqIsqvGXuQ6m%2FwAlJPOWjCo87Z29zOUd8GUv7zK2ReLTg18iUVhLYynFvjhCFsqspYeMb7LkRh%2FrwXVPCwYV0tD7vODgnFvOM%2FJ5lzjLWk8m6isZR5nUPVbReMaZYZ5jBAAABAJILR%2FDJfBUEglFir4IAABIAJGAATgvSg6lSMFzJpH6PQp%2BXSpUktoxUcI4q8W6qi2el06CUVlPLaPq6UfRFLlLc66EtX4mdkGmltnY1Wy4JTTeM4ZMot88mUoJZ%2Be5592tvv2PIuIqEs43MFPbC%2FmyZSxjBXVqe%2BzLdsvHJOdt8ZDptrIcE0VlDbYppblhm8IRSzudEUuUjqg00n3NFlrYPZZ7GbbxhkNpfHcq2sJ5yyrWFkzmu5XKxwHnJTV6%2BPsX1LuW1YTxnfYyngzknjK5RlOOMPD3Ihn4ZdSWMtEKLmmkuxZQ0JKLenl4I04lzsy%2BnHHY5LhYzg8%2Bqo42ezPKrvVLEnhe55VysJ5%2Bx513h2Mn3U0eWQMEAAAAgFqf94k%2BHsVJAJRYq9yAAASSAAASuSUjt6VS83qNCPZSyfeweqXP8jirZ%2Br34PYtNCnBQ%2FDlH0EK6impNZ%2B5pTust7Ho0KynFYa3OjXxnsTHd8mkW5J8kTWx5l%2FnVFLc8qtHUnnnJxTW7%2FkRlp%2F1IX4ucmtODnlI6oW%2BnCe77Eyiltj8jLS%2BUFDMeNyVTSm%2F9Cyi1uu5rFNrb%2FU3p7LnJpq%2BdiJcPfuRj33RWa1YwUeHyv8AqUz6TKUkymcZ3LKWWRpy8rkuljkZ33K6knuthJ7PGyMJYWMvgzUl2RpF6jSEXnhNdicP2RKwVf3OWuly90cFwkoYUUmzxLqMlPDXJ5V08ZT5TOGtSnVs62iLlpxJ47I8oEAgAAAgEp4afsyaixUklxkqSCUW%2FMqQAACSQACQSiUj2PDdvUr9UiqcXKSTaSP0KHRLqGl1JQi8cN7o4lYRlcSncascRjFcmsbetbyg3FqOc47nfTrqUm5Lc6IXUW93jc67W8goJNvjlHTLqCSWE3ktR6lj8Uee56dpc0qtN6Zb90y8m39jzL1ZmsLc82r6lj25OOrTcnlGTTCjvnk76EEqbeybNYT0wwluVeWs7vJSUHjnHyTCOUHDDys7krOUmtzRLLxjsaLjktuwlnIb2x3RTl5%2FkUa5M3l7GT2XK%2BxlKtTSfr%2FIj6mmnyT9XTS3yyJXsM4W65I%2Bti%2BVsiPrY4xghXscbJYM6tWUvwrZmfmSUc5wV%2BplF7b%2FAJGsL%2BSa22LfWuXwv9CXXnhrPL7EebPDeWZznKXs%2FujlnqmnHSkedX6bUrzSjLD92eVf9Hu6TzGm5x94nqeHOm1KMbmF7RcI1oaPUuzPiOp2M%2BndQr21RPNOWE%2FddmchAIAAABAL1OYv3iihPYEosVIHcAAnkkAAkFiUsn6n%2FZ90iFn0x39WH764%2FC32ij6W9k1T07qUv5I4aVGLlus4JrQ4e%2F6ERtNSTlhLP6kuzhFrCznkvCioNLg3UVpWyWNyrSWGsbo4pebSqa6dRrv8GtPqtxFYlPbuys%2BqVZPU3kwlfZeWvuTUvaTjlxe%2Fb2MHd0muWsfBpG4o4%2FEdVO4pNY1o0VSD%2FwAS5L646ViSZXlkx2LJrbdcBKOFsv1NIxWdng02imZOcU8akvfch3FOEXmaMJ9Rt4P8SZlLqtDVsm2YS6xHdKnuc9XqdWTajFJY7HJO5r1G22yIub99%2Fk1ivuNOfk0UXxgpoeX6f1IcXHIi8I0iszz77Fpbb4f2KypZXG7J0QoNKSztz8mlFUau2nS%2FdF3S8uTT%2FIaYtMjGXwisop8rgrGKcs%2FJa4g40dm%2FcUpSrU3F%2FjUdmfJeL7J16UbzT%2B8peip8rsz44ggEAAAEAs1%2B7i%2FloqSCUWKAAAEgkAnBISLpZO7pVhPqPULe1gm3Vmo7I%2Fc7a0p2drRoxwqdKKil9jz69bzKjenb7lKeZNJImXtjfsTSi3Fp9i6pv812LeWsJ43wQstvGVgz3e2%2F5FanqjhLg4pQak0%2F0Imlj7%2FBm4R7%2FoRVjmTWMYZjKjF7c7mc6fPODHRhbNoslJJ%2Bp7fJHmVYYxN%2B%2FJaN3WjxUZp9bcYxqI%2BqrPfX9skq5rLD1Mv9XXT%2FABsefXm%2F7xkONZ%2F43llPLqY3cnkRob7LOTVUN16c%2FJVW%2BGvjcs6MW%2BMvkhUsLCXBZUcs0jbPfK3YdDTuU0YeE%2FkNY%2BSssYwluVa%2FiWEaRik%2FbYnCeEXi1mOzM7%2B1lVgnTy98rfk5Yxq0px2lhPhLk9NtOMXLkacrbkoku%2FAlD5%2FMpobW2edn7E3CznHHsc2uUXGceVyZdRtIXVOae8akHFrHufmVejKhWnSmvVCWGZYIIBAAIJIBZbwkvbDKkglFigAABIJBJOCyRKiawp5Ptf7O%2Bn6%2BszuprMbeG33Z%2Bg9TuFGOIN77HmU3LVHKbf2O6EcSwlvzwZ1o%2FvvTvsXTlGLS7nbS6VeVI6%2FLSz%2FE8MpVpTtqjhWS1LhR32OadWK2i95PjuEoxWY8rY5pr239znqS3x88FUlh7bfJScPUkuCXB6sPd9ys4JN7GdX8Wyx90ZeU0yro5jn5M9CRPl7Z5IcN8ZCin%2BRpGkscFtC7rjuXWEljsXTUV6njJfbH8yaWl524ZaeNOE98lqNs7iEnBr8zky41XGSx2LqosYk0vY6KGqUdWhSXG5eV1ShVVOaxL4NNMZxck00UdCnjUzn0QbbXCDpKS2KeS0mnyT5TW%2FJVwbe%2BxaKjFtSeyOilUgmsbxfY0UqGptpKWDHynVrJQWpPhJHV9BcKGfJl%2BhxzptOSaaa5M9MpprDyWgnGXqWNtxOm3lpZOOssQaxuiyxKEW%2B%2Bx8f4p6U6dT6umts6Z%2FfsfMuDKuJRrBABABABaCTeH3TKkgksUAAABJKJLJF1BtmsaLfY6adpKTwotv2we70nwn1Hqcn9PbTai921hI%2FSPC3het0mwnGrBK4qPLxLZHu2%2FRoqoqldqel50%2FJ3KhTU8qEY9tkZys7dTc%2FLi2eN1y1VtpuoLEP8S9jx6fVaalGpBKoqW7SPqrDrSv4KVODhqjn18I4rrptzcV1WdXVKezcVwVvejeSoVKUtUo8qbzky0VK0Z042cZS%2Fii9kcV1066t6PmqKljlcYPKqUr5Q82VulF8b8nb0%2Fo1e9purXqKhr%2FDFptv7mF1Y3dpdeUo%2BcnvGUN8l5217Sj5v00tON1y%2F0OTz8t6vTh7ka1Oexo4rOcZfBmqVWSeINpeyMG8NtxeUyymotaljbuWWiSy0iqjHOHjB1QopR1f4UaTtf3bec7bHDWXlNbNNb5MqlZJL37FaV7JRSlH43RpFVsSmqiUO7JVXU1FSUnj9TWjVnSTiqjWdzO5r%2BrU3lnI5TqVo4T5Po5dSpQto0qNGTk1jODwbqCU5VJynCfZOOxvY3zhSeZRfu8msLp15LdJfc0axNKKbk%2By5OilQqSWZJwjvyuTopytHT9Tm543iy1FW05YVKTWN8s5butRo1EvKWh8Sb3K0unS6hGTo1YxmtknwebWVxZVfJqxxLvjgi0rVfrIxqKThP%2BR9r0%2BlC2tkqcctrOp8msL3W2tlFdznurSN7Jzg0ppbr3PBmnRuJQksNMmphrUms8GOtQaTxk56rU5uOV74NKcIum8Q2Tzg77bolp1ihcUriMsSxuuUfI9Y%2Fs7vrSt%2F2SP1FOWcaeV9z469sK1nWlSr05U5x5UlhnDKGCjRGCMDBGGCACU8STDWJNfIBKLYKkAAAEkko1hHLO%2B2tXVkktz9A8O%2F2e1L61jdXlTyKEuFj1M%2B0oeCOk0KUIqnLK3ck92erTtKdCiqFr%2B6prtFfzOuhSjShpi29%2BXyTUTlsng59flJ6nnfkzncYnHK%2BcHj9d6nKFpKNOmpOT0uMj5ay6L1KV3NU7dwpNerXsj6np%2FQq6t4O4u3CKeFGG233PoqNKFGOimnhd8mNynNaZc45ZS1t4U9Xlye757ZML2NW4ioUoppPEvYi4lCNBYprTHC3OfUqkk20njdZOao5Wl3Tbi9MvSpI6nU7wllrl%2Bxy1qHS6qk6vlxk3nKe%2BT4y%2Bt7il1KrC1zKi5Yi%2B6R9B0PpkZ0FXupuUk9odj2Kuf8MMRx2OGpCj5cvRGTW8so8W9kq13GpSjohFYlF8SfwX8hV6inGhpjjDXG51U7GlGS1U5JPbVykb3NnBUHHXGM58YPGrXNW0paKzyovGUc1CpW6jU00KUp74clwj6Oz8P0bTLusVpyW22yZ3OyjWoOjO3pRjjC9PCPF6tZw6fR1qH7uSwcNl0eVzGNXzVGlzst2e1R6XRnBR9Swu65%2BTjuLCFrU06fNlzud1OxgoJUqalqWXkpO1uLOpGpTUZQSw0nx8nmdUqXFSGJU4Sh8HJbdOoxks0tWpdt0jC%2F6ZWpKHl01FasRnFvLPX8OXCpUKv1Mc3CeHJ84PRp3FO5r%2BTByWXvLBFTpiqxxRm3vv8ABaPR1CdNTnJpct8v4N42ttTulGrTTfZtZReo6WvNGmk099Ox2ysaVXT5sFLbbUsnmXfTadpmpSo6oP8AElzD7HZRnRuLR%2FT1cNLj%2BpmqErm3W6gk%2BcFLLz1XnSinFJ7zlwd1TpcblNVopt%2F4lyedc%2BH5wjmjU1PPDOa96JdUrCUtEZy94veJydMtKUqma8k5p4znhHqXFvGpVcIpThFacpcHRaW30U41KcsvGGvc7p1ZVMOGNS5R4HinwvT6%2Fba4JQu6a9Mu0vhn471Dp9axuZ0Lim4VIPDTPOnFlGioJqLEnncqQCAXn%2BJNd0ioJRcoQAAATjbkkvFbnZbW7nJbbn6f%2FZ34ehVuZXd1b66VNely41fbufp8oQkksJpdjKrlKMYrd8I5H59OcttUVtt2Oa8vJ2uakPU8cI9C3rKtaU6iltJZyZ1XFxZyRqRUmtXqW%2F2OSnGhUg7irDzazk8LHB2yhOtRbk5RhNcdze1t4UKcacW3GSzu8nXFRTzF9jmuqsMpPd%2FB5911BWtvGmsqU6ijvzh9z1MPRpgkoruZXFKUoaNKknzg5VYQpUZTTerH%2Bh891DqsvraSpNLSvVGXuddLpErynKtOu1rXEXsTR6HbW8W2nOWc%2Bp9jg6fUj%2B2a9vOP95w32wfQRt4rVJNbbbLGCGttMeV%2FM87qcUo6Ix%2Fe1NlpZwqhiUZzTVOnwvk77Wca0VnGPfsztq%2BXThh8P9DnqStpKLShKSWyyeH1OhTd3Up%2BnS4505xuV8MXasb2fT6vpbTlCePfsfVV4OrBYlu2uC1KP7pZeWtm8ng%2BLZwnZU7eMlrnPbD4wd1rRhC0p0k4x1RTxn4ITcq3lxk4RitOU%2FxfZlPo1XqVlKrJTjtHO520bZRoqOrLS7cExjLD9OV3XucNfplOu06icE%2B0eDzkodNruk5JJvnHJpOqqtSVOrT0qKyngvY2lKP7xvVPfOFwdNraR1TWXDMm%2FVyduqNspKmsqSOmlOnWxh6n3XsTUoqp6cY9meY4zt683On6UsZfJ3069SVGk9LjBrDz%2FsdE4twwnl4x9zhhau2uYZj6Kiw0jd0KdhTlWppqPeOSkfVnTNYfsdFG501owUtSfp%2Bx1y9Uc5wZzrUowhTnL%2B%2B9K%2FQ8SdkqdypU4aYxezxnJ1StPNlmOYyS5RfyvJjq1avlnQntGccYaxgtF69%2FY%2BD8f9N%2Bqp0asaS82OVrivxI%2FM7i1lBvMcHFOOGZsgtU%2FFwUIJIGCX%2BCPxsQSC5QnAwGgoN8Jv8AIY7DAwWSRKibUoJyR9v4P8LXHWbmFXy2raMlqqNbH7TQtaFlbwpUacYwgsJJCVVanHG%2BM4Ip5nDUu62ZWf4d3ufP9RUqtxUp28XKTjlpf6mNr1uvY0oWl3bSlpeIziz1FUq19NaMXTpNbufLX2OeWJSc4RxFPDWN2d1CnSqQjOEXFJYwUdCfn%2BmT0e2SKidpGU6beuJe0u6VWl5bliq1%2Fi7nHdVJW9SCUm5zeleng55zVOvbuvTUoQqLVJrl%2FB7irzlUn6F5a%2FC%2FctUqR8ty%2FkeVUv5V6kFSitEd3vjcw6fa2tTXcThTlOc28yXHwbVb6lazlTi8Lsks7nk1eveRNKpTlLVlZxscsaFxXvFfUnoqNYjDGUl8nbHq9zZSp072j6J%2F40uH8mtbqlOpPFPLm1slwvucVlG4%2BqnVuX8rvhHTXuZZjT0KTk8P7HXaxoSXlyioSjtg561ZKThBylBLdJZPFqWtT6zFKTVJrKN50I04t1FKpqelylzFmULNTqQnQhFVqO%2Btt%2FoepRvaqp4qOPnZ5j%2FuRedVr0q0KdGClTw9TT7nhXs1ddThVm5PLxjsj2KlOVZQpqMk9OE3tsRaynG6p29OLcUs4e%2B511vPtKylKGFPunnHyd9pHzOJpxSzszs0YWzxsYV2lFxknLvhcnnS6XG7oRb9NXdrPucdGlXp3DjXjmonpc%2BcnqW1FU4SfOcN7CFOcq6dVJQzlYNpwpqDefSk3s%2BDgsak6F1KaUpQqS2T5PY%2Bop%2BbBa1ra%2FC%2BTj6jVp1Z0%2FKlicXmXtgimqlOcdM9orUo8qSOmhWco5aal7S7HBc3FWtU06dLS1LtnBW0vKvV6k6VSnphCGGvd%2B5ta2Lt678yb0LfS3wjtlWtlPU3pa7pbIzqXsYKMcbffk4uqV5KVvOMtMVVivsepSnB02991y%2B7JhmlT1Y3Zk45jpqcc7Iq6Lpxc4vC7JmlKbUlGWdTRt5EJRxUSkn2a2PjfFfgyheW9S46fSULiOW6cVtP%2Fqfkl3bTp1HGcWpJ4w0ccoP2I0b8ConqxgpjYYGH7Exg5PEVl%2ByJlTlGWJRcX7NYCj%2B7ltvlMrjAwSkWwTGMe8kaulTxnzI59gqVLGfNimaRtqDTbrx2R73hnqdl0h3brTVSNWKS2WU19zxalvbTnOarxWW3pfKIVrbpNyuI%2FZI0jY2zipfVU%2Ftk3%2FZNto1%2FX0N3xng%2BqhV8OVPCTtLmpbfXxjinUS9SfucHReg9Iq3lL63q1GNDmel4f2P1aw654f6ZQjbWd1bwowWyjLk7IeJOl1m4xvqGFvnWjCt4k6RTjhXtJuWVnUUh4s6TCNOmr2nKbXZ8YOO88YdOcH9NcUs5w8vgyufEHTqNq61GtCrKksz0PLZlQ630%2FrTt6sakKdOcuJvD2%2BD2rrrPTYYhK8o7YWNRzXHVunwpYjeUEmu0jGPiLp1nLypXVNL%2FANWcNnRb9fsK1byldw3Wd5Lc5K%2FXOk6qsZ3sN%2F8AznkXvXen27jOhd05TypLMlwcfUvHlPzI1bWKq8KdNPj5PU6h4hsL7pSVC7pxqYU3HOHF%2Bx0Wni7pX0UHUuYRrJbpy5ZxVvFFGu5KN1SjTbwvXukYVPE1KM4wp3FDEVtl8lLPxJaW9SdCVxR0t51OSyiV4ms5%2BIKdtroO2cNUrjzO%2FsXfiDpdW%2BjSndUp06c8pt7HXceJel0aj0XNLSt2k%2B5x9V8TdLrWnlU68J1JLP8AU8qx69TqqpQqypwUXiMk9zsl1uCqRj58dOMTk5Yyj27XrXSpOOK0EmsOTl3Omt17pitmoV6M5SeHutkTCdC4pQqW1el5UljZ%2FqZ1IWk5pU62KlJNqWf9DSpSo1Iz8y4p5UffbJ495fUbOhpnKEayWNpLDIsr61oUJzrVoVJSWXJtex4Nv1iXUfFdL6apiyUcyW3OD2XfWWpxr1o06udotrB6susWElTjOpBwgsvE1k8qn1m1t%2Brxup3VN0F6WsrKR6931%2ByuFOVtc0ZZjhKUsfc8%2B36xbWVeOi6hOTW6c9kdEfE9vWaVG5hnL1OTwj0qfWenKj531tLL%2FE9RhLqlhVk5UrmM37qeMGdGtRjUqa7uLm84bnsd0a1GhCFWVzBwm1FvV3Ou4nQlZSqOotONmmeDaX1vXcoVLlRcMqXqO246p0ulSS%2BrpbPZqSyUt%2Bq9J0S865p6ovClr7Gd51bpinGlG7pSmo52kjrteo2Pk09V3SjNLGXJbln1npdK5jrvaWXtp1rBhe9d6WqyhO6ox1P0ty4OddZ6VC4XlX9Km9L3UkeSvEtv50ozv4ObeG3LZo6aviXp8lKlK%2BpaHht5FHrvRndRpSv4LMcJuWU2XvfEXSXTpxhdUpONWKlmXG53WnibplW5cIXtFRistuWx1VOuWCuberHqNB0stNakd8%2Br9Mkk43tFY3%2FEjnvOqdPpQb%2BtpJwjq%2FFnY8268U9KVNOlf09SfbuTQ8b9KVNRqXEFJ8P%2BpZeMujxy5XsFld0edd9T8I9QparqpbSzvnTiWT5jxN0Lw1V13PTep0KElFSdFPKl9vk%2Bb6bPp%2FTL%2BnWnWjWwmnFxyjLqFOyvbqdzC6pU1N%2FgSxgihRsbalOTr0q1WSSipcL3OGdG1WZef3%2FCkRGhQcc%2BfFb8NHX0%2BdrZ39Ks60ZRg88clusXNv1C%2BdaNVKKioo4VChHP71b7cEujbOH98s59jOMLfOHUx84L6LfD%2Fe4ftgaKH%2FjfyOHJGWMsZfuSpbYazvkSm3Jtvdkan7k6n7jU%2Fcan7llUa7mka8l3Zf6qf8TH1M%2F4mHcSb%2FEHcTfdkK4nF5jNp%2FDLK6q5z5ks%2Fcl3VSX4pt%2FmR9TN%2FwCJ%2FqWV1Jcyf6h3c85jJr7MzdaT5kyjqN8shVGuGW86X8TI82XGWPNfuyPMl7sjzH7hTfPceY13JdR55Dqy9wqsk8qTQdacuZN%2FdllcTSxqf6l1dTX%2BJm0OqXVOGiFWUY%2ByexL6xetY%2BomvszJ9SunnNxV359TMp3NWo8zqSk%2FdvJDr1GsOcse2SIVpwlqjJxfungtO5qVHmU237tkedP8Aif6keY85y8l415Lu%2FwBR58%2F4mWVzNf4mW%2BqnjGp4Kq5qLibX2ZLu6r%2F7yX6l%2Fr6%2BjR509PtqeDoXXL9UfK%2Brq%2BXjGnVscsbytBtwqSi3zhmfnzb%2FABMsruab43IdzJvOcfYn6mo1jW%2F1KuvPvJ5I86TW7ZDqP3I8x%2B48x%2B48x%2B5KqP3LRqNcPBrC6ms7v9S%2F1lTC9T%2FUrK7qS5k%2F1Kq5mmmpMh3EvciVeTill5MnUfuVc2%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%2BQAAAOQAAAAAAAAAAAAAAAAAASASixQEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIAAAAAAAAEgksUIAAAAAAAAAAAAAAGR2AAAAAAAAAAAAAAAAA2%2BQwAAAAAAACQCUXMwQAMHodJ6H1HrledLpttKvKnHVN5SjBe7b2Q6v0PqHQriNHqVtKhOcdUG8OM17prZnnnv2PgnrvULKF3bWWadSOqlGdSMZ1V7xi3lo83p%2FR7%2FqnUo9Ps7adS6ba8vGHHHOc8Y75O%2FrPhHq3QraFzeUIStpy0KtRqRqQUvZtcM67H%2Bz7xD1GyoXdvZwdG4jrpudeEXKL4eGz5y4oVLa4q0Kq01KUnCSznDTw9y9lY3PUbuna2dGde4qvEKcFltns9X8E9e6FZ%2FV39i4W6eJThOM1B%2FwDmw9jqs%2F7O%2Bu3thb3lKnbRp3EFUpqpcxjJxfDwzgXhLq8%2BuPo9K2jVvIxUpRp1IyjGOM5ck8JGfXPDHVPDzpPqFBRp1s%2BXVpzU4Sxysrv8HkAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAkAlF8ozfIIAB9ROpOl%2FZtQVq3GFbqM43Ti%2BWoLQn8fiY82pdf2bzjcyco2nUoRtnL%2FAAqUG5xXxsnj5PnLdU3c0lVf7tzipfbO%2FwDI%2Bi8cXF3S8cX89c4St6qVtpeNEEl5en2WMPY9zq1xf0fH3ULTpNrTqX3VrSFvUg9tE6lOLnJPs9m3%2BZ5%2FUq%2FTfDnhq%2B6BZ3v7RvbypB3NWmsUKOh5xD%2BJ57nVceM%2BjXFv0q0vfDkbmFtaU7epVq1HCoklvoxx7p9z5rxV0eHQfEd50%2BjNzo05KVNy50ySks%2FOGbeD%2BtfsHr9O8dtUuKflzp1IUtpqLW7i%2BzXJ9Hb9L6Z1Ox6pHwp17qCuJ28qtayvKePOpp5a1LbJ09XXhWXQ%2FDf7dfUvqP2ZBQdmo6dOXznvnJ4fRZSo%2BBfE07JyjUdW3hUktpKi2%2Fbs3jJ5VKl1St4VupRr%2FwD4m2uYOdFy%2FwC8kmk0vtz9zxgAAAAAAAADS3oyuLilRi0pVJqCb4Tbwe5DwxGhV6g%2Bq38LO1srl2kqsKbqOrVWdoR27LLbxjYPwrKPWHazvqKs1bfWfWqLcHQxlSUec9tPvsY9X6LbW3T6HUul3sryxq1HRk6lLy6lKolnTKOWt1ummYeHej%2Ft7r1n0zzvI%2Bpnp8zTq07N8d%2BD1o%2BB7tWviCvWrRpx6LJwfpz50k%2BF7LGH%2BaMer%2BHun9HhKnW6lcyu1ShNQjZ4puUoqSWvVxvzgn9hdF%2FYMuqfti6xGoqHl%2FRLLqOGrC9fHz%2FIpe9AsOnWVOdz1C6dxUtoV1GnZ5ppzipRjrcvnGcHZZ%2BELK6l022l1C6hfdQowq00rPVRi5ZwnJSzjbd42OSl4atqXS7a66l1F2srxz8hxt5VKaUZOLc5p7Za7JvG5861h4zn5QAAAAAAAAAAAAAAAAJBILmbBAAPX6J4hr9GjcUXQo3dlcpKva3CzCeOHtumvdDrPiCr1ajQtadtQsrG2bdK1t09Kb5k29237s4Lq6jcxt1G3o0fJpKm3SjjzMZ9UvdvJ9FZ%2BN6lK3t%2Frek2HULu0goW93cQbnBL8KeNpY7ZOLpPiq76Z4grdYrU6d7cV4zjUVfOJa%2BXtx7fY26j4j6be2FW3oeGun2laoklXpSnqhv2TNbLxo6FvbRvuj2HUbizgoW1zcReuEV%2BFPDxJLtk8HqXUbjq3UK99eVPMuK8tc5Yxv8A0NuidZuug9TpX9k4qtTysTWYyTWGmvZo%2BhreOqFKzuqfRugWfTLq7g6da5pScnpfKimvTkyt%2FGtJdNsrO%2F6B02%2F%2BjpeTTq1lLVpzlLn5MKfjCpZ9ad%2F0zptlZ0Z0lRrWcYOVGtHvqT9%2F9ieoeM6l4rShQ6ZYWnT7avG4dnRp4hWmv4%2FfbY%2Bfuq8bm7rV40oUVVnKap01iMMvOF8IxAAAAAAA5Z6F70PqNheVrWtaVnUpScW4U20%2FlPG6MP2devizuP8AKl%2FQuukdRlxYXT%2B1CX9Dosuk9To3tCrLpl6406kZPTbzzhNP2PrKlN9ardXtOoWHVbW1ueoTvra5hYzqODeU4yj7OOOHs0ZuVx%2B2HS%2FYnVn0f6H9m%2F8A60vO8vnzMYxq1%2BrH5HH1fp9xS6NQ6P0npvVa9GNd3Na5rWU4OpPGmKjHfCSz33bOPw9Y9U6T1qheVuj9SlCmppqFtPO8JRWNvdn0d71rqt50ylay6D1PV%2BzZ21aStpLzK8lCPmPbf0wXyc%2FiC0vOtUnPzevKlGlB07Gp06o6cZxgo4T1YW6e%2BO54q6Z1V%2BGP2cujdSdZ3n1Dl9LPTpVPT7c5bPa6nbXfVenW9KVbr9GlStaVL6H9n1JU9cIpc6sYclnOO5Np1brtGdraz6f1iPS1YfRV7eFOXeDTnFcZy019i3hunddItrao59ft%2FKlquenxs3UpXO%2FEeFFNbPKPkrnonVK11VqR6RdUo1JuSgqMsQTfC%2Bxk%2Bg9WXPTbv%2FJkT%2BwOr%2F8A%2BZd%2F5Miv7C6rnH7Nus%2B3lMsvD%2FV28Lpd4%2FtRkX6r0C76NZ2Ne9pzozu1NqjUg4ygovG%2Bfc8sAAAAAAAAAAAAEgYBIRoZMAgAAAk9DoFpSvvEHTrW4i5Ua9zTpzinjKckmsn1nW%2FB9hZ%2BN%2BmULVur0XqNyqUXCb9DU9NSnq90z4u%2FoRteo3VCGdNKrOCzzhSaOcAgAAAAAAAAHff9ZveoXta6q16kZ1pOTjCclFfZZOf626%2F4mt%2FmP%2BpZdRvIrCu7hL2VWX9R%2B0bz%2Fi7j%2FNl%2FUftG8%2F4u4%2FzZf1I%2FaF2%2Bbqv%2FAJsv6kfXXT%2F%2FAKa3%2BZL%2Bo%2BtuXzcVv8x%2F1I%2Bsuf8AiK3%2BY%2F6j6y4%2F8er%2FAJj%2FAKlo393BJRuq6S9qsv6lHc13zWqP%2FwB7I8%2Br%2FwCLU%2F5mQ61R81J%2F8zI8yf8AHL9WRrl%2FFL9WT5k%2F45f8zI1yznVL9SVVqLicv%2BZnXd9Wub3p9lZ15KVOz1%2BW98%2Bt5eX9ziAAAAAAAAAAJ29iMjPwSM5GRlgAlFzNgEADBOMcoPnjAwQen4drwtvEvS61WShTp3dKUpSeEkpLLZ914X6vYV%2FF1%2F0nqlaDtJdTle2ddTWmFaM29pcYktv%2FALPgetSjPrfUJQkpRlc1WmnlNa2cBIRAAAAAAAADAAAAAAAAAAAAAAABONiAAAAAACRgYfsASkMFsce7JwVwRjYYDRGBgYGBgDAGCSMDBOBgYWCBgnCIxtwEhgY%2BRj5IwTj5GCMfJOCCWCMDBIDAwMDAwAAMDAAwMDAwBgYGBgYGBgYGBgYGBgYGBgDAwMDBOBgYGBgnBaOz33JcWuVgnHyz%2F9k%3D%22%0A%20%20%20%20return%20(trinity_jpg_b64%2C)%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(np%2C%20trinity_jpg_b64)%3A%0A%20%20%20%20import%20base64%0A%20%20%20%20import%20io%0A%20%20%20%20from%20PIL%20import%20Image%0A%0A%20%20%20%20trinity_img%20%3D%20np.asarray(Image.open(io.BytesIO(base64.b64decode(trinity_jpg_b64))))%0A%20%20%20%20return%20(trinity_img%2C)%0A%0A%0A%40app.cell%0Adef%20_(mo)%3A%0A%20%20%20%20R_photo_slider%20%3D%20mo.ui.slider(50%2C%20200%2C%20step%3D1%2C%20value%3D80%2C%20label%3D%22Fireball%20radius%20%24R%24%20(m)%22)%0A%20%20%20%20return%20(R_photo_slider%2C)%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(R_photo_slider%2C%20mo%2C%20np%2C%20plt%2C%20trinity_img)%3A%0A%20%20%20%20_pixels_per_metre%20%3D%201.55%20%20%20%20%20%20%20%20%20%20%20%20%20%20%23%20the%20100%20m%20scale%20bar%20is%20155%20pixels%20long%0A%20%20%20%20_cx%2C%20_cy%2C%20_ground_y%20%3D%20339%2C%20305%2C%20365%20%20%20%23%20centre%20of%20the%20blast%20and%20the%20ground%20line%2C%20in%20pixels%0A%20%20%20%20_r_px%20%3D%20R_photo_slider.value%20*%20_pixels_per_metre%0A%20%20%20%20_angle%20%3D%20np.linspace(0%2C%202%20*%20np.pi%2C%20400)%0A%20%20%20%20_x_circ%20%3D%20_cx%20%2B%20_r_px%20*%20np.cos(_angle)%0A%20%20%20%20_y_circ%20%3D%20_cy%20-%20_r_px%20*%20np.sin(_angle)%0A%20%20%20%20_above_ground%20%3D%20_y_circ%20%3C%3D%20_ground_y%0A%0A%20%20%20%20_fig_tr%2C%20_ax_tr%20%3D%20plt.subplots(figsize%3D(9%2C%206.7))%0A%20%20%20%20_ax_tr.imshow(trinity_img%2C%20cmap%3D%22gray%22)%0A%20%20%20%20_ax_tr.plot(_x_circ%5B_above_ground%5D%2C%20_y_circ%5B_above_ground%5D%2C%20%22.%22%2C%20color%3D%22limegreen%22%2C%20ms%3D3)%0A%20%20%20%20_ax_tr.set_title(f%22Trinity%20test%2C%200.025%20s%20after%20the%20explosion%20%20%20%7C%20%20%20R%20%3D%20%7BR_photo_slider.value%7D%20m%22%2C%20fontsize%3D12)%0A%20%20%20%20_ax_tr.axis(%22off%22)%0A%20%20%20%20_fig_tr.tight_layout()%0A%20%20%20%20mo.vstack(%5BR_photo_slider%2C%20_fig_tr%5D)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20**One%20photo%20gives%20the%20energy.**%20Rearrange%20%24R%20%3D%20(E%20t%5E2%2F%5Crho)%5E%7B1%2F5%7D%24%20for%20%24E%24%20on%20paper%2C%20then%20type%20it%20in%20below.%20(Powers%20in%20Python%20use%20%60**%60%2C%20so%20%24x%5E3%24%20is%20%60x**3%60.)%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_(R_photo_slider)%3A%0A%20%20%20%20t_photo%20%3D%200.025%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%23%20seconds%2C%20printed%20on%20the%20photo%0A%20%20%20%20rho_air%20%3D%201.2%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%23%20kg%2Fm%C2%B3%2C%20density%20of%20air%0A%20%20%20%20R_photo%20%3D%20R_photo_slider.value%20%20%20%23%20metres%2C%20from%20the%20slider%0A%0A%20%20%20%20E_photo%20%3D%20rho_air%20*%20R_photo**5%20%2F%20t_photo**2%20%20%20%23%20from%20R%5E5%20%3D%20E%20t%5E2%20%2F%20rho%0A%20%20%20%20kilotons_photo%20%3D%20E_photo%20%2F%204.184e12%20%20%20%23%201%20kiloton%20of%20TNT%20%3D%204.184%20%C3%97%2010%C2%B9%C2%B2%20J%0A%20%20%20%20return%20E_photo%2C%20kilotons_photo%2C%20t_photo%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(E_photo%2C%20R_photo_slider%2C%20kilotons_photo%2C%20mo)%3A%0A%20%20%20%20mo.md(rf%22%22%22%0A%20%20%20%20With%20%24R%20%3D%20%7BR_photo_slider.value%7D%24%20m%3A%20%24E%20%5Capprox%24%20%7BE_photo%3A.2e%7D%20J%2C%20about%20**%7Bkilotons_photo%3A.1f%7D%20kilotons%20of%20TNT**.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20**Two%20photos%20test%20the%20law.**%20If%20the%20formula%20is%20right%2C%20the%20radius%20grows%20like%20%24t%5E%7B2%2F5%7D%24.%20On%20a%20later%20photo%20(0.062%20s)%20Taylor%20measured%20%24R%20%3D%20185%24%20m.%20Comparing%20it%20with%20your%20measurement%20gives%20the%20power%20of%20%24t%24%2C%20which%20should%20come%20out%20close%20to%20%242%2F5%20%3D%200.4%24%3A%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_(R_photo_slider%2C%20np%2C%20t_photo)%3A%0A%20%20%20%20t_second%20%3D%200.062%20%20%20%23%20seconds%0A%20%20%20%20R_second%20%3D%20185.0%20%20%20%23%20metres%2C%20Taylor's%20measurement%0A%0A%20%20%20%20power_of_t%20%3D%20np.log(R_second%20%2F%20R_photo_slider.value)%20%2F%20np.log(t_second%20%2F%20t_photo)%0A%20%20%20%20power_of_t%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20**More%20photos%20give%20more%20trust.**%20Taylor%20didn't%20stop%20at%20two%3A%20he%20measured%20the%20fireball%20on%2025%20frames%2C%20from%200.1%20ms%20to%2062%20ms%2C%20checked%20they%20all%20followed%20%24R%20%5Cpropto%20t%5E%7B2%2F5%7D%24%2C%20and%20only%20then%20quoted%20an%20energy.%0A%0A%20%20%20%20**Discussion**%20(double-click%20to%20edit)%3A%0A%0A%20%20%20%20-%20Why%20can't%20a%20single%20photo%20tell%20you%20whether%20the%20formula%20is%20right%3F%20Because%20one%20photo%20always%20gives%20*some*%20number%20for%20%24E%24%2C%20right%20or%20wrong.%20With%20two%20photos%20the%20formula%20makes%20a%20prediction%20we%20can%20check%3A%20the%20radius%20must%20grow%20like%20%24t%5E%7B2%2F5%7D%24.%20If%20the%20slope%20above%20had%20come%20out%20as%2C%20say%2C%20%240.6%24%2C%20the%20formula%20would%20be%20wrong%20and%20the%20energy%20meaningless.%0A%20%20%20%20-%20Move%20the%20slider%20by%205%20m%20(about%204%25).%20By%20how%20much%20does%20%24E%24%20change%3F%20Why%20does%20that%20make%20it%20worth%20measuring%20many%20frames%20rather%20than%20one%3F%20About%2020%25%3A%20%24E%20%5Cpropto%20R%5E5%24%2C%20so%20a%204%25%20error%20in%20the%20radius%20becomes%20roughly%20a%2020%25%20error%20in%20the%20energy.%20Every%20photo%20has%20small%20reading%20errors%20like%20this%3B%20measuring%20many%20frames%20lets%20them%20average%20out%2C%20and%20checks%20the%20%24t%5E%7B2%2F5%7D%24%20law%20over%20a%20long%20stretch%20of%20time%20rather%20than%20between%20just%20two%20points.%0A%0A%20%20%20%20%2F%2F%2F%20details%20%7C%20How%20close%20did%20Taylor%20get%3F%20(open%20after%20you%20have%20answered)%0A%20%20%20%20Taylor's%201950%20answer%20was%20**16.8%20kilotons**.%20The%20real%20figure%20was%20secret%20at%20the%20time%3B%20it%20was%20later%20given%20as%20about%2021%20kilotons%2C%20and%20a%202021%20re-analysis%20puts%20it%20at%20about%2025.%20The%20right%20size%20of%20answer%2C%20from%20a%20published%20photo%20and%20an%20argument%20about%20units.%0A%20%20%20%20%2F%2F%2F%0A%0A%20%20%20%20**Sources.**%20Photo%3A%20Trinity%20test%20fireball%20at%200.025%20s%2C%20US%20Government%2C%20public%20domain%2C%20via%20%5BWikimedia%20Commons%5D(https%3A%2F%2Fcommons.wikimedia.org%2Fwiki%2FFile%3ATrinity_Test_Fireball-25ms.jpg).%20Data%3A%20G.%20I.%20Taylor%20(1950)%2C%20Proc.%20R.%20Soc.%20Lond.%20A%20**201**%2C%20175%E2%80%93186.%20Yield%20comparisons%3A%20%5BRevisiting%20Taylor%20and%20the%20Trinity%20Test%5D(https%3A%2F%2Farxiv.org%2Fabs%2F2403.19657)%20(2024).%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%23%20AI%20extension%20%E2%80%94%20physics-informed%20feature%20engineering%0A%0A%20%20%20%20A%20neural%20network%20trained%20to%20predict%20drag%20force%20%24F_D%24%20directly%20from%20raw%20inputs%20%24(L%2C%20V%2C%20%5Crho%2C%20%5Cmu)%24%20must%20implicitly%20discover%20the%20relevant%20dimensionless%20groups%20from%20data.%20Giving%20it%20the%20%CE%A0-groups%20as%20inputs%20instead%20%E2%80%94%20the%20drag%20coefficient%20%24C_D%24%20and%20the%20Reynolds%20number%20%24Re%24%20%E2%80%94%20is%20a%20simple%20form%20of%20**physics-informed%20machine%20learning**%3A%20you%20encode%20the%20structure%20of%20the%20problem%20before%20training.%0A%0A%20%20%20%20For%20a%20sphere%20of%20diameter%20%24L%24%20moving%20at%20speed%20%24V%24%20through%20a%20fluid%20of%20density%20%24%5Crho%24%20and%20viscosity%20%24%5Cmu%24%2C%20the%20two%20%CE%A0-groups%20from%20the%20dimension%20matrix%20are%3A%20%5C%5BC_D%20%3D%20%5Cfrac%7BF_D%7D%7B%5Crho%20V%5E2%20L%5E2%7D%2C%20%5Cqquad%20Re%20%3D%20%5Cfrac%7B%5Crho%20V%20L%7D%7B%5Cmu%7D.%5C%5D%0A%0A%20%20%20%20Below%20is%20a%20synthetic%20dataset%20of%20220%20random%20%22experiments%22%3A%20for%20each%20one%2C%20a%20sphere%20size%20%24L%24%2C%20a%20speed%20%24V%24%2C%20a%20fluid%20density%20%24%5Crho%24%2C%20and%20a%20fluid%20viscosity%20%24%5Cmu%24%20are%20all%20drawn%20independently%20and%20completely%20at%20random%20%E2%80%94%20spanning%20everything%20from%20a%20droplet%20in%20air%20to%20a%20boulder%20in%20syrup.%20The%20drag%20force%20%24F_D%24%20each%20one%20produces%20follows%20the%20real%20physical%20law%20every%20time%3B%20you%20just%20don't%20get%20to%20see%20%24L%24%2C%20%24%5Crho%24%2C%20or%20%24%5Cmu%24%20in%20the%20left-hand%20plot%20below%2C%20only%20%24F_D%24%20and%20%24V%24.%0A%0A%20%20%20%20Both%20panels%20below%20use%20logarithmic%20axes%20again%2C%20but%20for%20a%20different%20reason%20than%20the%20blast-wave%20demo%20in%20lecture%3A%20it's%20not%20primarily%20to%20read%20off%20a%20slope%20here%2C%20but%20because%20%24L%24%2C%20%24V%24%2C%20%24%5Crho%24%2C%20and%20%24%5Cmu%24%20all%20span%20several%20orders%20of%20magnitude%20%E2%80%94%20a%20log%20scale%20is%20what%20makes%20any%20pattern%20visible%20at%20all%2C%20rather%20than%20squashing%20everything%20near%20zero.%0A%0A%20%20%20%20**Task.**%0A%20%20%20%201.%20Replace%20the%20two%20%60np.zeros_like%60%20placeholders%20with%20the%20correct%20formulas%20for%20%24C_D%24%20and%20%24Re%24.%0A%20%20%20%202.%20Run%20the%20cell.%20%20The%20comparison%20plot%20below%20will%20update.%0A%20%20%20%203.%20Answer%20the%20discussion%20questions%20at%20the%20end.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(np)%3A%0A%20%20%20%20%23%20A%20wide%2C%20independent%20random%20survey%3A%20for%20many%20trials%2C%20a%20sphere%20size%2C%20speed%2C%0A%20%20%20%20%23%20fluid%20density%2C%20and%20fluid%20viscosity%20are%20all%20drawn%20independently%20of%20each%0A%20%20%20%20%23%20other.%20Real%20experiments%20wouldn't%20scramble%20everything%20like%20this%20%E2%80%94%20the%0A%20%20%20%20%23%20point%20is%20exactly%20to%20show%20what%20that%20scrambling%20does%20to%20a%20raw%20plot.%0A%20%20%20%20_rng%20%3D%20np.random.default_rng(12)%0A%20%20%20%20_n_drag%20%3D%20220%0A%0A%20%20%20%20_L_rand%20%20%20%3D%2010%20**%20_rng.uniform(-4%2C%20-1%2C%20_n_drag)%20%20%20%20%20%23%20sphere%20size%3A%200.1%20mm%20to%2010%20cm%0A%20%20%20%20_V_rand%20%20%20%3D%2010%20**%20_rng.uniform(-1.5%2C%201.0%2C%20_n_drag)%20%20%23%20speed%3A%20~0.03%20to%2010%20m%2Fs%0A%20%20%20%20_rho_rand%20%3D%2010%20**%20_rng.uniform(0%2C%203.3%2C%20_n_drag)%20%20%20%20%20%23%20density%3A%201%20to%20~2000%20kg%2Fm%C2%B3%0A%20%20%20%20_mu_rand%20%20%3D%2010%20**%20_rng.uniform(-4.3%2C%20-1%2C%20_n_drag)%20%20%20%23%20viscosity%3A%20~5e-5%20to%200.1%20Pa%C2%B7s%0A%0A%20%20%20%20%23%20The%20real%20drag%20law%20(the%20Schiller%E2%80%93Naumann%20correlation)%20generates%20F_D%20from%0A%20%20%20%20%23%20these%20%E2%80%94%20it%20bridges%20the%20Stokes%20and%20Newton%20regimes%20above%20smoothly%20across%0A%20%20%20%20%23%20the%20whole%20range%20of%20Re.%20Students%20never%20see%20this%20formula%2C%20only%20the%20F_D%20it%0A%20%20%20%20%23%20produces.%0A%20%20%20%20_Re_true%20%3D%20_rho_rand%20*%20_V_rand%20*%20_L_rand%20%2F%20_mu_rand%0A%20%20%20%20_C_D_true%20%3D%20np.where(%0A%20%20%20%20%20%20%20%20_Re_true%20%3C%3D%201000%2C%20(24.0%20%2F%20_Re_true)%20*%20(1%20%2B%200.15%20*%20_Re_true**0.687)%2C%200.44%0A%20%20%20%20)%0A%20%20%20%20_FD_rand%20%3D%20_C_D_true%20*%20_rho_rand%20*%20_V_rand**2%20*%20_L_rand**2%0A%0A%20%20%20%20drag_data%20%3D%20np.column_stack(%5B_FD_rand%2C%20_L_rand%2C%20_V_rand%2C%20_rho_rand%2C%20_mu_rand%5D)%0A%20%20%20%20return%20(drag_data%2C)%0A%0A%0A%40app.cell%0Adef%20_(drag_data)%3A%0A%20%20%20%20F_D_data%2C%20L_data%2C%20V_data%2C%20rho_data%2C%20mu_data%20%3D%20drag_data.T%0A%0A%20%20%20%20C_D_data%20%3D%20F_D_data%20%2F%20(rho_data%20*%20V_data**2%20*%20L_data**2)%0A%20%20%20%20Re_data%20%20%3D%20rho_data%20*%20V_data%20*%20L_data%20%2F%20mu_data%0A%20%20%20%20C_D_data%2C%20Re_data%0A%20%20%20%20return%20C_D_data%2C%20F_D_data%2C%20Re_data%2C%20V_data%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(C_D_data%2C%20F_D_data%2C%20Re_data%2C%20V_data%2C%20np%2C%20plt)%3A%0A%20%20%20%20_fig_ai%2C%20(_ax_raw%2C%20_ax_nd)%20%3D%20plt.subplots(1%2C%202%2C%20figsize%3D(12%2C%204))%0A%0A%20%20%20%20_ax_raw.scatter(V_data%2C%20F_D_data%2C%20color%3D%22%231a5276%22%2C%20s%3D30%2C%20alpha%3D0.55)%0A%20%20%20%20_ax_raw.set_xscale(%22log%22)%0A%20%20%20%20_ax_raw.set_yscale(%22log%22)%0A%20%20%20%20_ax_raw.set_xlabel(%22Velocity%20%24V%24%20(m%2Fs)%22%2C%20fontsize%3D12)%0A%20%20%20%20_ax_raw.set_ylabel(%22Drag%20force%20%24F_D%24%20(N)%22%2C%20fontsize%3D12)%0A%20%20%20%20_ax_raw.set_title(%22Raw%3A%20%24F_D%24%20vs%20%24V%24%20across%20many%20random%20spheres%20and%20fluids%22%2C%20fontsize%3D12)%0A%20%20%20%20_ax_raw.grid(True%2C%20alpha%3D0.3%2C%20which%3D%22both%22)%0A%0A%20%20%20%20_re_ref%20%3D%20np.logspace(np.log10(Re_data.min())%2C%20np.log10(Re_data.max())%2C%20300)%0A%20%20%20%20_CD_ref%20%3D%20np.where(_re_ref%20%3C%3D%201000%2C%20(24.0%20%2F%20_re_ref)%20*%20(1%20%2B%200.15%20*%20_re_ref**0.687)%2C%200.44)%0A%20%20%20%20_ax_nd.scatter(Re_data%2C%20C_D_data%2C%20color%3D%22%231a5276%22%2C%20s%3D30%2C%20alpha%3D0.55%2C%20zorder%3D5%2C%20label%3D%22data%22)%0A%20%20%20%20_ax_nd.loglog(_re_ref%2C%20_CD_ref%2C%20%22k--%22%2C%20linewidth%3D1.8%2C%20label%3D%22empirical%20drag%20law%22)%0A%20%20%20%20_ax_nd.set_xlabel(%22Reynolds%20number%20%24Re%24%22%2C%20fontsize%3D12)%0A%20%20%20%20_ax_nd.set_ylabel(%22Drag%20coefficient%20%24C_D%24%22%2C%20fontsize%3D12)%0A%20%20%20%20_ax_nd.set_title(r%22Nondimensional%3A%20universal%20%24C_D(Re)%24%20curve%22%2C%20fontsize%3D12)%0A%20%20%20%20_ax_nd.legend(fontsize%3D10)%0A%20%20%20%20_ax_nd.grid(True%2C%20alpha%3D0.3%2C%20which%3D%22both%22)%0A%0A%20%20%20%20_fig_ai.tight_layout()%0A%20%20%20%20_fig_ai%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20**Discussion**%20(double-click%20to%20edit)%3A%0A%0A%20%20%20%20Once%20your%20formulas%20are%20correct%2C%20the%20left%20panel%20should%20look%20like%20almost%20pure%20noise%20%E2%80%94%20at%20any%20given%20%24V%24%2C%20the%20points%20span%20several%20orders%20of%20magnitude%20in%20%24F_D%24%2C%20because%20%24L%24%2C%20%24%5Crho%24%2C%20and%20%24%5Cmu%24%20were%20all%20scrambled%20independently%20of%20%24V%24%20when%20the%20data%20was%20generated.%20There%20is%20no%20reliable%20way%20to%20predict%20%24F_D%24%20from%20%24V%24%20alone.%0A%0A%20%20%20%20The%20right%20panel%20shows%20every%20single%20one%20of%20those%20same%20220%20points%20landing%2C%20without%20exception%2C%20on%20one%20smooth%20curve%20%E2%80%94%20because%20%24C_D%24%20really%20is%20a%20function%20of%20%24Re%24%20alone%2C%20regardless%20of%20which%20sphere%20or%20fluid%20produced%20it.%20Nothing%20was%20hidden%20or%20cherry-picked%3A%20it's%20exactly%20the%20same%20data%2C%20just%20plotted%20against%20the%20combination%20the%20physics%20actually%20depends%20on.%0A%0A%20%20%20%20-%20Which%20representation%20would%20you%20rather%20give%20to%20a%20neural%20network%20as%20input%20features%3F%20The%20%24C_D%24%E2%80%93%24Re%24%20representation%20%E2%80%94%20it%20collapses%20every%20fluid%20and%20sphere%20size%20onto%20one%20universal%20curve%2C%20so%20the%20model%20only%20has%20to%20learn%20one%20relationship%20instead%20of%20a%20separate%20one%20for%20each%20combination%20of%20%24%5Crho%24%2C%20%24%5Cmu%24%2C%20and%20%24L%24.%0A%20%20%20%20-%20Why%20does%20the%20%24C_D%24%E2%80%93%24Re%24%20representation%20help%20the%20model%20generalise%3F%20Because%20it%20removes%20the%20redundant%20scale%20information%20the%20raw%20variables%20carried%20%E2%80%94%20two%20experiments%20with%20the%20same%20%24Re%24%20and%20%24C_D%24%20behave%20identically%20regardless%20of%20the%20actual%20fluid%20or%20sphere%20size%2C%20so%20the%20model%20needs%20far%20less%20data%20to%20learn%20the%20true%20relationship.%0A%20%20%20%20-%20What%20does%20this%20say%20about%20the%20value%20of%20dimensional%20analysis%20before%20applying%20machine%20learning%20to%20physical%20data%3F%20It%20shows%20dimensional%20analysis%20is%20a%20cheap%2C%20principled%20form%20of%20feature%20engineering%3A%20doing%20it%20first%20can%20turn%20a%20scattered%2C%20high-dimensional-looking%20problem%20into%20a%20clean%20low-dimensional%20one%20before%20any%20training%20happens.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_()%3A%0A%20%20%20%20import%20marimo%20as%20mo%0A%0A%20%20%20%20return%20(mo%2C)%0A%0A%0Aif%20__name__%20%3D%3D%20%22__main__%22%3A%0A%20%20%20%20app.run()%0A